Sei sulla pagina 1di 134

Tematiche organizzative

Il Project Management

Numero 30 - novembre 2011


TEMATICHE ORGANIZZATIVE

IL PROJECT MANAGEMENT

Gli articoli contenuti in questo fascicolo sono riservati


alluso personale, per finalit di studio e di ricerca allinterno della Banca d'Italia

1 PREMESSA

18 PROFILO STORICO DEL PROJECT MANAGEMENT


A. Villa; Gli articoli di PM Forum, giugno - luglio, 2001

33 CURRENT PRACTICE IN PROJECT MANAGEMENT: AN EMPIRICAL STUDY


J. Fortune, D. White; International Journal of Project Management, n. 20,
2002, pagg. 1-11

44 CRITICAL SUCCESS FACTORS IN ENTERPRISE WIDE INFORMATION MANAGEMENT


SYSTEM PROJECTS
M. Sumner; Proceedings of the Americas Conference on Information Systems,
1999, pag. 232-234

51 ONE SIZE DOES NOT FILL ALL PROJECTS: EXPLORING CONTINGENCY DOMAINS
A. J. Shenhar; Management Science, vol 47, n. 3, March, 2001, pagg. 394-414

73 ORGANIZATIONAL ALTERNATIVES FOR PROJECT MANAGERS


R. Youker; Project Management Quarterly, vol. 8, n. 1, March, 1977, pagg.
18-24

82 ON THE NATURE OF THE PROJECT AS A TEMPORARY ORGANIZATION


J. R. Turner, R. Muller; International Journal of Project Management, n. 21,
2003, pagg. 1-8

90 EVOLVING THE PROJECT MANAGEMENT OFFICE: A COMPETENCY CONTINUUM


G. M. Hill; Information system management, Fall 2004, pagg. 45-51

97 A FRESH LOOK AT THE CONTRIBUTION OF PROJECT MANAGEMENT TO


ORGANIZATIONAL PERFORMANCE
M. Aubry, B. Hobbs; Project Management Journal, vol. 41, n. 1, February,
2011, pagg. 3-16

111 LINKING PROJECT MANAGEMENT WITH BUSINESS STRATEGY


S. Srivannaboon; Working Paper, 2006

121 THE OFFICE OF STRATEGY MANAGEMENT


R.S. Kaplan, D.P. Norton; Harvard Business Review, October, 2005

NUMERO 30 - NOVEMBRE 2011


IL PROJECT MANAGEMENT 1

INTRODUZIONE

Nel Project Management Body of Knowledge (PMBOK), il principale manuale di


riferimento per il project management (da qui in poi PM), il progetto definito come:
sforzo temporaneo intrapreso per creare un risultato, un prodotto o un servizio unici.
Secondo la terminologia in uso nellingegneria economica, il progetto lo sforzo per la
definizione e lesecuzione di uno specifico obiettivo da realizzare entro dati vincoli di
tempo e di costo 2 .
Essenzialmente due aspetti accomunano tali definizioni: la temporaneit dello
sforzo e lunicit del risultato. Il progetto, dunque, ha un inizio e una fine ben definiti e il
suo obiettivo la creazione di un risultato nuovo. Tali caratteristiche rendono lattivit
progettuale di unazienda profondamente diversa rispetto a quella di linea, e richiedono
uno stile di gestione peculiare.
Il PM presuppone aree di conoscenza e strumenti che si applicano alla
pianificazione, al monitoraggio e al controllo della realizzazione di un progetto. Il suo
sviluppo sostanzialmente dovuto alla necessit, sempre pi pressante per le imprese, di
gestire i vincoli rappresentati dal triangolo tempi, costi, risultati. Con laumentare
delle interconnessioni tra le diverse competenze e dei ritmi e degli stimoli allinnovazione,
si sono ampliate nel tempo sia le applicazioni del PM, sia le aree di conoscenza necessarie
al project manager, che includono oggi necessariamente tutte quelle del moderno
management: da quelle pi tecniche (come costi e approvvigionamenti) a quelle pi soft
(come la qualit, la comunicazione o la gestione dei team).
Le molteplici implicazioni che il PM ha sulle competenze, sulle modalit di
interazione tra le strutture e talvolta in modo diretto sulle strutture stesse, ne fanno
indubbiamente una tematica organizzativa.
Allo sviluppo metodologico del PM si accompagnata una proliferazione di
definizioni, prassi, certificazioni, applicazioni. Lobiettivo di questa rassegna di mettere
in luce alcuni elementi di fondo: in che contesto utile adottare le tecniche di PM, in che
misura esso contribuisce a creare valore, quali sono i suoi limiti e le sue potenzialit. Gli
spunti per le applicazioni in Banca dItalia sono numerosi, essendo molteplici gli ambiti, a
livello formale e informale, nazionale e sovranazionale, in cui necessario attivare
collegamenti interfunzionali.

1
A cura di Paola Battipaglia, Fabrizio Ferriani e Fabrizio Mancini.
2
La definizione tratta dallAACE International Cost Engineering Terminology (RP no. 10S/90).

1
Il contenuto della rassegna sviluppato in tre sezioni: la prima dedicata alle
applicazioni del PM e ne evidenzia levoluzione nel tempo; la seconda presenta i principali
strumenti metodologici; la terza focalizzata sulle implicazioni organizzative che possono
derivare dalla scelta di valorizzare lapproccio per progetti nella gestione aziendale e
conclude discutendo le relazioni tra PM e strategia.

1. LAPPLICAZIONE CONCRETA DEL PROJECT MANAGEMENT

Lanalisi storica recente mostra come gi a partire dalla rivoluzione industriale


periodo caratterizzato da grandi cambiamenti e dallo sviluppo veloce delle conoscenze
scientifiche e delle innovazioni tecnologiche siano riscontrabili numerosi esempi di
utilizzo di fondamenti di PM, anche grazie allapplicazione di strumenti e tecniche
avanzate di gestione delle attivit. Naturalmente, queste attivit non furono n considerate
n descritte come progetti; solo con la prima met del Novecento i concetti di progetto e
di project management divennero comuni con la formalizzazione di tecniche e strumenti
ad hoc.
Nel 1917, Henry Gantt diffuse negli Stati Uniti il diagramma omonimo per
rappresentare graficamente su un asse temporale le singole attivit che compongono un
progetto. I primi diagrammi di Gantt furono impiegati per la realizzazione di grandi
infrastrutture (la diga di Hoover nel 1935, lInterstate Highway System nel 1956) e il loro
utilizzo facilit la programmazione e il controllo dello stato di avanzamento delle attivit.
La prerogativa dellattivit progettuale, che richiede unorganizzazione per
obiettivi piuttosto che per funzioni, port, a partire dagli anni 30, a unificare il
coordinamento delle attivit in un unico ruolo organizzativo. Nacque cos, in Exxon nel
1935, il ruolo del project engineer che, sebbene privo di effettivi poteri gestionali, assunse
il compito di coordinare le diverse funzioni aziendali coinvolte nei progetti. Nel 1942, con
il c.d. Progetto Manhattan (che, in poco pi di due anni, port alla realizzazione delle
prime due bombe atomiche), questo ruolo si evolse verso il moderno project manager: la
responsabilit e la piena gestione operativa del progetto furono affidate ad ununica
persona che rispondeva dei risultati raggiunti.
La complessit e la rilevanza strategica del progetto Manhattan stimolarono anche
lintroduzione di altri concetti chiave del PM: la chiara definizione degli obiettivi e
lallocazione delle responsabilit, la scomposizione delle attivit in sottoprogetti (Work
Breakdown Structure) e lassegnazione degli stessi a differenti organizzazioni.
Negli anni 50, la ricostruzione post bellica determin la necessit di sviluppare
numerosi progetti soprattutto nei settori dellimpiantistica e delle infrastrutture; inoltre
negli Stati Uniti, con la guerra fredda, lapplicazione del PM nel settori della difesa e in

2
quello aerospaziale rappresent un importante fattore strategico. Nello stesso periodo, il
progresso dellinformatica rese possibile un ulteriore ampliamento dei sistemi di gestione
dei progetti. In quegli anni, furono sviluppati parallelamente due strumenti tra loro simili:
il Critical Path Method (CPM) e la Project Evaluation and Review Technique (PERT). Il
CPM fu inizialmente introdotto nel 1956 da M.R. Walker e J.E. Kelley per ottimizzare la
gestione degli impianti della DuPont de Nemours. Le attivit della DuPont erano
fondamentalmente note e le relative durate potevano essere stimate con accuratezza; lo
strumento del CPM aveva lobiettivo di determinare il flusso di lavoro in grado di
ottimizzare lutilizzo delle risorse a parit di tempi. Lo strumento fu successivamente
commercializzato e trov ampio utilizzo nelle aziende del settore delle costruzioni 3 . Il
PERT fu invece sviluppato nel 1958 dalla Booz, Allen & Hamilton Inc. su commissione
della marina degli Stati Uniti in occasione del progetto Polaris 4 . Poich nella
realizzazione del progetto, la variabile dei costi era considerata trascurabile e lindirizzo
strategico era la minimizzazione della durata, il PERT fu focalizzato esplicitamente su
questo secondo aspetto e diede un contributo molto importante, permettendo di coordinare
efficacemente le diverse imprese coinvolte nel progetto che fu completato con due anni di
anticipo sulle previsioni.
Durante gli anni 60 lapplicazione del PM divenne massima nel settore
impiantistico e nelle costruzioni, ma dallambito dei progetti militari statunitensi 5 che
vennero i contributi metodologici pi significativi. In particolare, furono sviluppati sistemi
di controllo pi accurati (c.d. earned value) basati sullanalisi congiunta dei costi e delle
attivit lungo tutto il ciclo di vita del progetto e sul confronto tra il valore assorbito dal
progetto rispetto alle stime preventive. Nello stesso periodo progredirono anche gli studi
sulle implicazioni organizzative delle attivit progettuali, in seguito sistematizzate nel
modello di Youker, e sorsero le prime associazioni professionali (INTERNET in Europa
nel 1965 e PMI nel 1969 negli Stati Uniti).
Con la crisi economica esplosa negli anni 70, lattenzione alla pianificazione e
alla gestione dei progetti divenne un fattore strategico. Lo shock petrolifero determin
laumento degli investimenti spesso finanziati anche attraverso fondi pubblici nel
settore energetico. Accanto a casi di successo, anche per progetti particolarmente
complessi e rischiosi 6 , lincertezza del contesto determin casi in cui per errata
progettazione, difetti di governance o influenze ambientali esterne i progetti intrapresi
3
Nel 1961, la Catalytic Construction of Philadelphia fu la prima grande azienda ad acquisire il CPM per le
proprie attivit.
4
Il Polaris fu il primo missile balistico in grado di essere lanciato dai sottomarini atomici.
5
Dopo il successo del programma Apollo della NASA che port allallunaggio del 1969, le tecniche di PM
furono fortemente sponsorizzate allinterno del Dipartimento di Difesa americano.
6
E il caso dei progetti per lestrazione di petrolio dai giacimenti del Mar del Nord o del progetto Trans
Alaskan Pipeline System (TAPS) per il trasporto del petrolio dai giacimenti del Nord sino al porto nel Sud
dellAlaska.

3
diedero risultati inferiori alle aspettative, con la moltiplicazione dei costi e dei ritardi nei
tempi di realizzazione.
Si accrebbe, quindi, lesigenza di rendere pi efficienti i sistemi di gestione 7 e
inoltre, per la prima volta, fu considerato un fattore critico di successo nella realizzazione
dei progetti il coinvolgimento dei portatori dinteresse. A partire dalla seconda met degli
anni 80, questi sviluppi misero in discussione i sistemi di PM esistenti e ne indussero un
radicale rinnovo, anche grazie alla crescente disponibilit di strumenti informatici sempre
pi sofisticati. Nel 1985 fu pubblicato il primo PMBOK che compendiava tutte le
conoscenze ed esperienze fino ad allora acquisite.
Gli sviluppi successivi hanno sancito la definitiva maturit del PM attraverso
lutilizzo in nuovi settori industriali (informatico, automobilistico e farmaceutico), la
diffusione nei paesi emergenti e lapplicazione della metodologia oltre che alle fasi di
commitment, anche alla gestione delle opere realizzate.
In particolare, le recenti strategie di project financing, sempre pi attente a una
gestione accurata del rischio, hanno determinato la diffusione, soprattutto nel settore delle
infrastrutture, di progetti BOOT (Build Own Operate Transfer) sviluppati da project
companies capaci di gestire lintero ciclo di business: dalla ricerca dei finanziamenti fino
alla realizzazione del progetto e alla gestione redditizia del prodotto per conseguire il
ritorno dellinvestimento. E maturato, quindi, un nuovo modello di PM adatto a
supportare efficacemente la strategia aziendale, in quanto focalizzato sempre di pi sul
valore generato da un progetto piuttosto che sulla sua semplice realizzazione.
Tale valenza strategica ha aggiunto alle applicazioni del PM un terzo livello. Si
quindi passati dalla gestione operativa del singolo progetto (il PM in senso proprio) al
coordinamento di progetti correlati (il program management) arrivando, infine, al c.d.
portfolio management, utilizzato per definire, selezionare e indirizzare lintera attivit
progettuale di unazienda per conseguire gli obiettivi strategici prefissati.
Larticolo di Fortune e White (2002) evidenzia unampia e trasversale
applicazione del PM in diversi settori produttivi, con una percentuale significativa per
quanto riguarda i servizi ICT e lattivit di R&S, la logistica e la costruzione di
infrastrutture, il settore finanziario-assicurativo e la pubblica amministrazione. La
diffusione del PM in settori industriali a elevata complessit tecnologica e il suo utilizzo
nella gestione di progetti di frontiera pu essere ricondotta alla necessit di minimizzare

7
In particolare, alcune sentenze delle Corti americane sanzionarono le inefficienze nella gestione dei progetti
di realizzazione di impianti nucleari, vietando di scaricare sulla collettivit, attraverso laumento delle
tariffe, i maggiori costi sostenuti.

4
la probabilit di fallimento, presidiando i fattori critici di successo (si veda Block, 1983 o
Sumner, 1999) durante lintero ciclo di vita del progetto 8 .
Nel settore dellICT e della R&S lutilizzo del PM si dimostrato una scelta
vincente per lo sviluppo di nuovi prodotti ad alto contenuto tecnologico, come nel caso del
progetto ARP gestito dalla NASA per il lancio di un nuovo prototipo di elicottero senza
equipaggio (www.pmi.org). Altre aziende, come AT&T (www.pmi.org), hanno creato al
loro interno uno specifico project management center of excellence con lobiettivo di
uniformare le prassi metodologiche, ottenendo risultati concreti per quanto concerne le
capacit gestionali dei project manager. In altri casi, il PM ha rappresentato una sorta di
volano per il rilancio aziendale dopo un periodo di crisi, come PTC (Ward, 2009) che, in
un contesto di competizione mondiale, riuscita a incrementare sensibilmente il tasso di
successo dei propri progetti attraverso la definizione di uno schema di PM globale. I
risultati di eccellenza del PM nella gestione e nel rispetto di tempi, costi e richieste del
committente si osservano anche nellorganizzazione di grandi eventi complessi, come
mostrato da Cantamessa et al. (2007) nel case study dedicato alle Olimpiadi invernali di
Torino 2006.
Nellambito delle istituzioni pubbliche, a livello europeo si possono citare le
project cycle management guidelines della Commissione Europea (2004). Il project cycle
management stato adottato dalla Commissione Europea per la prima volta nel 1992,
come approccio metodologico per la definizione, la gestione e il controllo dei progetti. Le
project cycle management guidelines raccolgono unampia gamma di best practices e di
strumenti da utilizzare come riferimento per misurare la rilevanza, la fattibilit e lefficacia
dei progetti finanziati dalla Commissione. In Italia il Codice degli Appalti e il relativo
Regolamento di Attuazione, d.p.r. 207/2010, rappresentano un esempio di definizione
normativa della figura del project manager. Difatti, dallanalisi degli artt. 10 di entrambe le
fonti citate emergono chiaramente i margini di sovrapposizione tra funzioni e compiti del
responsabile unico del procedimento (RUP) e la figura del project manager.
NellEurosistema, con laumentare del numero e della complessit dei progetti comuni,
in crescita la pressione per ladozione sistematica di strumenti e prassi di PM anche in
ambiti non IT.

8
Il ciclo di vita di un progetto linsieme delle fasi in genere sequenziali che compongono il progetto. La
suddivisione del progetto in fasi serve a garantirne il controllo dellandamento da parte delle organizzazioni
coinvolte nella realizzazione. A unapposita metodologia affidata la definizione del numero e del livello di
dettaglio delle fasi, dei deliverable (ossia i prodotti misurabili e verificabili) di ciascuna fase e delle
procedure di controllo e convalida degli stessi. Secondo una classificazione tipica delle fasi, un progetto
prende avvio dallideazione o concezione, seguono la definizione dei contenuti e delle caratteristiche del
progetto, la pianificazione delle attivit, delle risorse richieste e dei costi connessi, la scansione temporale
delle attivit pianificate, lo sviluppo o esecuzione del progetto e la chiusura. Trasversale allintero ciclo di
vita la fase di monitoraggio e controllo.

5
Il livello di applicazione degli strumenti comunque vario (cfr. anche la sezione
successiva), al punto che alcuni autori propongono oggi alle imprese un modello di
autodiagnosi per valutare il livello di maturit nelladozione del PM. Il PM maturity
model 9 distingue cinque possibili stadi:
1) Iniziale: il caso di organizzazioni in cui coesistono molteplici approcci metodologici
non armonizzati tra di loro e manca un orientamento strutturale nei confronti del PM.
Ladozione del PM e dei relativi strumenti lasciata alla discrezionalit dei singoli
responsabili di progetto.
2) Controllato: si caratterizza per lutilizzo di una metodologia di PM formalizzata, e per
la presenza di meccanismi semplici per il controllo dei progetti. Tuttavia il PM
applicato solo ai grandi progetti strategici e il ruolo del project manager subordinato
ai ruoli esistenti nella organizzazione madre.
3) Standardizzato: si avvale di una metodologia di PM basata su standard formalizzati e
uniformi in tutta lorganizzazione. La Direzione coinvolta nel supporto alla
realizzazione dei progetti e i project manager svolgono un ruolo attivo e autonomo nel
coordinamento di team di progetto interfunzionali.
4) Integrato: prevede lutilizzo di metodologie avanzate di PM e la piena integrazione tra
i processi di PM e i processi ordinari dellorganizzazione, e tra i sistemi informativi del
PM e i sistemi informativi aziendali. Il sistema premiante di progetto allineato al
sistema aziendale di gestione delle risorse umane.
5) Ottimizzato: il caso in cui lintera organizzazione permeata dallapplicazione del
PM e sviluppa in via continuativa nuovi standard e best practices sulla base delle
lessons learned e del benchmarking con lesterno. La Direzione attivamente coinvolta
nel processo di miglioramento continuo del PM.

2. STRUMENTI DI PROJECT MANAGEMENT

Oggi il PM propone un approccio metodologico completo per gestire in modo


integrato le singole fasi in cui i progetti sono articolati, presidiando i fattori critici di
successo e minimizzando la probabilit di fallimento.
Larticolo di Fortune e White (2002) classifica gli strumenti e le tecniche di PM
sulla base della frequenza di utilizzo, sintetizzando i risultati di unanalisi empirica che ha
coinvolto oltre 200 project manager impegnati in diversi settori industriali, sia in imprese
pubbliche sia private. Dalla ricognizione effettuata, emerge la tendenza allutilizzo di un
numero contenuto di strumenti. Tra le principali limitazioni individuate dai project
manager allutilizzo integrale del PM vi sono i costi connessi alla documentazione da

9
Si vedano ad esempio i contributi sullargomento di Crawford (2002) e Kerzner (2005).

6
produrre, leccesso di rigidit con cui vengono definiti le fasi e gli strumenti del PM,
nonch il tempo da dedicare allattuazione della metodologia.
Daltra parte Shenhar (2001) evidenzia lattitudine delle imprese a selezionare gli
strumenti a seconda del grado di complessit dei progetti: per i progetti pi semplici ci si
limita sostanzialmente a pochi controlli formali, applicando gli schemi essenziali per il
controllo del budget e per la definizione e il monitoraggio delle macrofasi (milestone) di
progetto. Nel caso di progetti pi complessi invece richiesto, sin dalla fase iniziale, un set
pi completo e rigoroso di strumenti, per lindividuazione di blocchi di attivit, per la
pianificazione finanziaria, per la gestione dei rischi (risk management analysis), per il
controllo delle interdipendenze di progetto. Allaumentare della complessit dei progetti si
modifica anche il ruolo del project manager, che dismette sempre pi le competenze di
natura tecnica, per concentrarsi prevalentemente su quelle di tipo manageriale. Anche per
quanto riguarda le implicazioni organizzative, che saranno riprese nella terza sezione,
Shenhar (2001) sottolinea lesistenza di una proporzionalit diretta tra complessit dei
progetti e strutture. Progetti semplici presentano strutture organizzative elementari (il team
con il responsabile di progetto) che evolvono quindi verso forme organizzative a matrice,
fino a prevedere lesistenza di un PMO o di strutture a ombrello, con compiti di
coordinamento tra i diversi progetti e lintera struttura organizzativa.
Di seguito si presenta un elenco degli strumenti essenziali previsti nel PM,
classificati in base al ciclo di vita del progetto; sono contrassegnati con * quelli individuati,
nella rassegna di Fortune e White (2002), come di pi frequente applicazione.

- Project charter: sancisce in modo formale lavvio del progetto e rappresenta uno schema
riassuntivo per la definizione, a livello macro, di obiettivi, portatori di interesse e partecipanti
al progetto, con indicazione delle relative responsabilit. Ha una struttura generalmente snella
e rinvia ad altri documenti per analisi pi approfondite.
- Business case: un documento di sintesi che viene redatto per lapprovazione del progetto
da parte dellalta direzione, con un livello di complessit flessibile, che pu essere adattato ai
singoli casi. Il business case presenta informazioni sul contenuto essenziale del piano, sui
Inizializzazione

soggetti coinvolti, sui tempi, sugli impatti in termini di risorse, nonch sui benefici attesi. Nel
business case vengono anche individuati i potenziali rischi che possono incidere sulla
realizzazione del progetto e vengono definiti alcuni indicatori per valutare il grado di
raggiungimento degli obiettivi.
- Analisi costi-benefici (CBA)*: solitamente inserita allinterno del business case per
valutare lopportunit finanziaria di un progetto. La CBA prevede lelencazione analitica di
costi e benefici (tangibili e intangibili), sia dal punto di vista tecnico sia dal punto di vista
economico. Per ogni voce di progetto individuata prevista lattribuzione di una valutazione
quantitativa, che ne definisca limpatto in termini economico-finanziari. Sulla base dei
risultati della CBA, dovrebbero essere intrapresi solo i progetti che presentano un vantaggio
economico in termini di VAN (valore attuale netto), ossia quelli in cui la somma algebrica
scontata dei flussi di cassa relativi al progetto positiva.

7
- GANTT*: un diagramma a barre per la rappresentazione sequenziale delle attivit del
progetto. Consente di visualizzare in modo immediato la durata delle principali fasi che
compongono il progetto, facilitando il rispetto dei tempi.
- Work Breakdown Structure (WBS)*: un diagramma ad albero che definisce lambito
totale del progetto, dandone una rappresentazione logico-gerarchica. Partendo dal massimo
livello di aggregazione, la WBS suddivide il progetto sulla base di alcuni criteri alternativi,
tra cui la scomposizione del prodotto finale nelle sue sotto-componenti secondo unottica
orientata al prodotto o la scomposizione del progetto sulla base del suo ciclo di vita. Il
principale contributo della WBS fornire una descrizione analitica del progetto, per
semplificare lassegnazione delle responsabilit e il coordinamento tra i vari soggetti
coinvolti.
Pianificazione

- Critical Path Method (CPM)*: un diagramma reticolare che consente la rappresentazione


schematica delle attivit che compongono il progetto, nonch delle relative durate e costi. I
tempi vengono definiti in modo deterministico e vengono individuate le attivit critiche, ossia
quelle per le quali un eventuale ritardo produce effetti significativi sulle tempistiche di
progetto. La struttura reticolare consente inoltre di evidenziare le interdipendenze tra le
diverse attivit del progetto. Uno strumento simile al CPM il Program Evaluation and
Review Technique (PERT*) in cui la rappresentazione grafica delle attivit associata a una
stima probabilistica dei tempi di progetto.
- Matrice RACI (Responsible, Accountable, Consulted, Informed): una matrice utilizzata
per individuare i soggetti coinvolti nelle fasi di un progetto, attribuendo loro le relative
responsabilit. Vengono individuati quattro possibili ruoli allinterno del progetto: 1)
Responsible il soggetto effettivamente incaricato di realizzare il progetto; 2) Accountable
il soggetto ultimo competente per le decisioni e le azioni collegate al progetto, tipicamente
coincide con un rappresentante dellalta direzione; 3) Consulted rappresenta colui che viene
coinvolto per fornire prevalentemente pareri di tipo tecnico; 4) Informed colui che viene
informato sullo stato del progetto, pur non avendo un ruolo attivo nella sua realizzazione.
Verifica e controllo

- Project Tracking: sotto questa definizione sono inclusi diversi strumenti metodologici, che
hanno tutti come obiettivo principale il monitoraggio continuo dello stato di avanzamento del
progetto, consentendo di mettere in luce eventuali situazioni di criticit. Tra questi strumenti
vale la pena citare la curva a S, che utilizzata per visualizzare eventuali scostamenti tra
progetto e realizzazione, attraverso il confronto tra dati preventivi e dati consuntivi. La
curva a S generalmente utilizzata per evidenziare situazioni problematiche riferite sia ai
costi sia ai tempi.

- Project Closure Checklist: schema riassuntivo di controllo per la chiusura del progetto.
Vengono verificati tempi, costi, output di progetto e obiettivi, questi ultimi da valutare sulla
Chiusura

base di indicatori di risultato definiti in sede di pianificazione. La project closure checklist


spesso redatta congiuntamente alle lessons learned, con lintento di raccogliere le
informazioni di ritorno e le esperienze pi significative emerse durante tutto il ciclo di vita
del progetto.

8
La maggior parte degli strumenti metodologici citati sono solitamente raccolti
allinterno di software specialistici per il PM, che rappresentano uno dei supporti di
maggior successo nella gestione di tutte le fasi di progetto, come illustrato in Fortune e
White (2002). Dallarticolo emerge anche un ampio ricorso a strumenti di PM costruiti
autonomamente allinterno dellorganizzazione (in-house) e un uso limitato di strumenti
per la gestione dei rischi e per lanalisi delle condizioni dincertezza in grado di
pregiudicare il raggiungimento dellobiettivo (ad esempio, matrice probabilit-impatto,
analisi what-if).

3. LE IMPLICAZIONI ORGANIZZATIVE DEI PROGETTI


Poich una delle caratteristiche essenziali del PM lintegrazione di risorse,
abilit e competenze, del tutto naturale che i meccanismi organizzativi su cui pi
direttamente impatta il PM siano quelli che presidiano i collegamenti laterali e
lintegrazione. Quanto maggiore la necessit di attivare/potenziare questi meccanismi
rispetto al tradizionale schema verticale proprio del modello funzionale, tanto pi si
modificano gli assetti organizzativi, evolvendo verso il modello a matrice, fino
allorganizzazione per progetto.
Loriginale contributo di Youker (1977) riepiloga le caratteristiche delle diverse
forme organizzative e ne mette in evidenza le implicazioni per il PM. Nel modello
funzionale, basato sulla specializzazione e sullautorit di linea, il coordinamento risulta
di solito pi semplice per le attivit ripetitive, ma pi problematico per la gestione delle
attivit trasversali, e pi bassa generalmente anche lattitudine al cambiamento.
Allestremo opposto, nellorganizzazione per progetto, tutte le competenze e le risorse
necessarie sono staccate dalle attivit routinarie e affidate allautorit dei project manager
secondo unorganizzazione ad hoc (es. task force). Questa soluzione funzionale alla
realizzazione del progetto, in quanto la chiarezza dei ruoli e degli obiettivi aiuta la
coesione e la motivazione degli addetti. Daltra parte, presenta il problema della gestione
nel continuo delle risorse, che devono essere ricollocate alla fine del progetto. Il modello a
matrice prevede la coesistenza della dimensione verticale per funzioni e di quella
orizzontale per progetti, impegnando le risorse sia nelle attivit di linea sia in quelle
orientate a obiettivi trasversali; nelle diverse tipologie di matrice (debole, bilanciata, forte),
il project manager e il manager funzionale hanno rispettivamente minore e maggiore grado
di potere e autonomia nella gestione delle risorse. In teoria il modello a matrice sfrutta i
punti di forza di entrambi i modelli estremi; in pratica la sua applicazione trova il maggiore
ostacolo nella coesistenza di due linee di riporto, che pu creare situazioni di conflitto
soprattutto quando non sono chiaramente definiti i ruoli, le responsabilit e lautorit dei
soggetti coinvolti.

9
In definitiva, lassetto organizzativo ottimale per la gestione del progetto non
scontato, ma dipende da una serie di fattori quali la complessit, i vincoli temporali,
limportanza del progetto, il tipo di tecnologia impiegata (standard o innovativa). La
tensione tra la dimensione verticale specialistica e gerarchizzata e quella orizzontale,
interdisciplinare e orientata agli obiettivi definiti per un orizzonte temporale limitato di
fatto ineliminabile, ma pu essere vista come salutare per lorganizzazione, in quanto
fornisce uno stimolo continuo alla ricerca dellequilibrio tra i due modelli produttivi.
Lidea dellequilibrio ottimale tra queste dimensioni ben sintetizzata dalla figura 1,
estratta da Youker (1977), in cui strutture organizzative e risorse umane dedicate a ciascun
progetto vengono messe in relazione tra di loro sotto forma di organizational continuum.
Dallorganizational continuum di Youker (1977) emerge la soluzione di compromesso
rappresentata dalla struttura matriciale, in grado di mediare tra gli estremi organizzativi
della forma funzionale e dellorganizzazione per progetto.

Fig. 1

Come precedentemente sottolineato, la maggior flessibilit dellorganizzazione


matriciale deve tuttavia scontare le potenziali situazioni di conflitto tra manager funzionali
e di progetto. In questo modo diventa fondamentale, per il successo del PM allinterno di
unorganizzazione di tipo matriciale, che lalta direzione definisca chiaramente le
responsabilit e lautorit del project manager, cos come i ruoli delle strutture funzionali.
In particolare gli obiettivi, le scadenze e il budget per il progetto devono essere esplicitati e
sottoscritti anche dai manager funzionali. Senza questi punti fermi, secondo Youker (1977)
il project manager sar un diplomatico frustrato che lotta per convincere le varie funzioni
a rispettare i tempi e il budget.

10
A prescindere dalla tipologia organizzativa adottata, Youker (1977) sottolinea la
necessit che il project manager possieda sia competenze tecniche sia spesso soprattutto
capacit manageriali. Queste ultime diventano addirittura cruciali nel caso specifico
dellorganizzazione per progetto, al punto tale che prima di adottarla opportuno valutare
se sono disponibili professionisti con le competenze adeguate. Da questo punto di vista,
larticolo di Youker (1977) riprende lanalisi sulle competenze del project manager,
sviluppato anche negli articoli di Sumner (1999) e Fortune e White (2002), evidenziandone
la trasversalit rispetto alle forme organizzative prescelte.
Un approccio diverso allo studio del PM e delle sue implicazioni organizzative
viene proposto in Turner e Muller (2003), che analizzano il rapporto tra progetti e
organizzazione alla luce dei modelli principale-agente 10 . Turner e Muller (2003)
definiscono il progetto come unagenzia per il cambiamento, ossia come
unorganizzazione flessibile e dinamica, concepita per rompere linerzia delle strutture pi
statiche, come ad esempio lorganizzazione funzionale. Secondo unulteriore prospettiva di
analisi, il progetto visto come unagenzia per lutilizzo delle risorse in cui si distingue, in
primo luogo, labilit del project manager nel massimizzare le risorse da destinare al
progetto e, in secondo luogo, la capacit di ottimizzarle per il raggiungimento con successo
degli obiettivi. Infine, il progetto declinato come unagenzia per la gestione
dellincertezza, in cui vengono ripresi i concetti di rischio, urgenza e necessit di
integrazione assenti nella tradizionale organizzazione funzionale.
Dallo studio di Turner e Muller (2003) emerge una molteplicit di parametri da
utilizzare nel definire un progetto nel suo complesso. Questo fa s che nellorganizzazione
di un progetto siano contemporaneamente enfatizzati: lesplicito approccio al rischio, la
necessit di un sistema di monitoraggio e controllo dei risultati, la comunicazione,
lorientamento al cambiamento, la gestione delle risorse. E proprio la combinazione di
innovazione, flessibilit e controllo che rende interessante lapproccio per progetti anche
per le organizzazioni di tipo tradizionale, al fine di superare il grado di inerzia e di rigidit
che le contraddistingue.

3. 1 UN APPROCCIO STRUTTURALE ALLA GESTIONE DEI PROGETTI: IL PROJECT MANAGEMENT


OFFICE
Per rispondere allesigenza di gestire un elevato numero di progetti con successo e
applicare efficacemente le metodologie di PM integrandone il contenuto allinterno
dellorganizzazione nellottica del miglioramento continuo, un numero crescente di
aziende, nonch alcune Banche centrali come la Federal Reserve di New York e la Central
10
I modelli principale-agente, introdotti nel campo dalle scienze economiche, studiano le difficolt che
emergono in condizioni dinformazione incompleta e asimmetrica, allorch un agente opera per conto di un
principale e vi un potenziale disallineamento tra gli interessi dell'agente e quelli del principale.

11
Bank of Ireland, hanno optato per la creazione di un project management office (da qui in
poi PMO). Cantamessa et al. (2007) definiscono il PMO come una funzione aziendale,
solitamente di staff, con la responsabilit di riportare allalta direzione lo stato dei
progetti in corso e di far assimilare gradualmente il project management in tutta
lorganizzazione. Al PMO affidato quindi il compito di sviluppare e promuovere la
cultura del PM allinterno delle organizzazioni, assicurando il supporto e il coordinamento
in sede di applicazione metodologica. Hill (2004) introduce il concetto di competency
continuum del PMO che ricorda almeno in parte la classificazione introdotta dal PM
maturity model secondo cui possibile individuare cinque stadi di PMO: il project office,
il basic PMO, lo standard PMO, ladvanced PMO, il center for excellence. Le tipologie di
PMO definite da Hill (2004) sono associate a livelli crescenti di complessit rispetto
allattivit svolta da parte del PMO. Per ogni stadio viene valutato il grado di formalismo,
la quantit di risorse dedicate al PMO, il livello di autonomia organizzativa e
lallineamento con la strategia complessiva dellorganizzazione madre. Partendo dalla
gestione di pochi progetti semplici, realizzati in unottica microscopica, con controlli di
base e senza metodologie formalizzate, si giunge alla gestione macroscopica di progetti
integrati tra di loro (programmi), realizzata secondo le migliori best practices
metodologiche, con personale appositamente dedicato, attraverso una collaborazione tra le
strutture organizzative finalizzata al raggiungimento degli obiettivi strategici. Un esempio
di PMO riferito al contesto italiano si trova in Cantamessa et al. (2007). Per la gestione dei
progetti di sviluppo e manutenzione del territorio, un ente pubblico locale si era
inizialmente dotato di un PMO con un duplice ruolo di supporto sia nei confronti dei
singoli RUP sia nei confronti della Direzione con lobiettivo di garantire una metodo-
logia comune per la realizzazione dei progetti e la gestione integrata delle opere pubbliche
da realizzare. Successivamente, nel momento in cui le tecniche di PM sono state
pienamente assimilate nella gestione dei progetti, anche attraverso la realizzazione di un
apposito sistema informativo per le opere pubbliche basato sui principi metodologici di
PM, si ritenuto di poter rimuovere il PMO dalle strutture permanenti.
I benefici derivanti dalla costituzione di un PMO sono esaminati anche da Aubry e
Hobbs (2011), i quali presentano uno studio sul contributo alla performance organizzativa
da parte del PM in generale, e del PMO nello specifico. La loro analisi si fonda su una
precedente ricerca empirica 11 in cui avevano riscontrato che oltre il 50% dei rispondenti
metteva in discussione lutilit e lefficacia del PMO, percepito pi come un ulteriore
appesantimento burocratico che come uno strumento di facilitazione.
Gli autori individuano due possibili risposte per giustificare questo risultato: da un
lato i manager tendono generalmente a privilegiare una visione microscopica dei benefici
del PMO, legandoli esclusivamente alle loro aree di interesse, senza tener conto

11
Aubry e Hobbs (2007).

12
dellorganizzazione nel suo complesso. Dallaltro lato per apprezzare pienamente il
contributo del PMO necessario non considerare esclusivamente lottica di tipo
finanziario-contabile ma anche, e soprattutto, tenere conto di altre misure della
performance, adottando una prospettiva multidimensionale. Rispetto a questultimo punto
Aubry e Hobbs (2011) individuano una lista di 79 indicatori per la valutazione del PMO,
costruiti facendo riferimento a quattro direttrici:

- Rational goals and efficiency che riprende il concetto di profittabilit finanziaria del
PMO, ritorno dellinvestimento ed efficienza (economica);
- Open systems and effectiveness, che misura la capacit del PMO di essere un
catalizzatore per la crescita e linnovazione, nonch la sua potenzialit nel leggere e
dare risposte concrete agli stimoli ambientali percepiti;
- Human relations che valuta lapporto del PMO con riferimento allo sviluppo delle
risorse umane, alla maggior coesione e al benessere lavorativo;
- Internal processes, che giudica i benefici per lorganizzazione in termini di capacit di
controllo, comunicazione e condivisione delle conoscenze, standardizzazione dei
processi.
Le quattro dimensioni sopraelencate concorrono a sintetizzare una prospettiva di
rango superiore, in cui si valuta il contributo del PMO alla overall output quality, come
rappresentato in figura 2, tratta da Aubry e Hobbs (2011).

Fig. 2

13
Nel complesso, sia in Hill (2004) che in Audry e Hobbs (2011), il PMO assume il
ruolo di una struttura capace di dare un valore aggiunto trasversale allorganizzazione, in
grado di operare indipendentemente dalle specificit del singolo progetto.

3. 2 IL PROJECT MANAGEMENT E IL CONTROLLO STRATEGICO


Lallineamento tra la gestione dei progetti e la strategia aziendale un presupposto
indispensabile affinch le organizzazioni conservino il proprio vantaggio competitivo.
Questo punto ampiamente discusso nellarticolo di Srivannaboon (2006) che individua
sette prospettive in grado di qualificare il rapporto tra i vantaggi competitivi (competive
attributes) dellimpresa e i principali elementi che contraddistinguono la gestione dei
progetti.
Nello specifico, Srivannaboon (2006) cerca di spiegare il legame che intercorre tra
la strategia complessiva dellimpresa e la strategia, lorganizzazione, il processo-
evoluzione, gli strumenti, la valutazione e la cultura del progetto (cfr. figura 3).
Fig. 3

Dalla figura emerge un primo livello di allineamento, di natura strategica, in cui i


progetti selezionati per supportare la strategia aziendale vengono raccolti allinterno del
project portfolio. In questa fase sono coinvolti i processi di pianificazione strategica e di
project portfolio management (PPM), dove questultimo il supporto per tradurre

14
operativamente le decisioni strategiche formulate dallalta direzione. In questa fase di
pianificazione, gli strumenti che consentono di classificare i progetti sulla base della loro
priorit strategica risultano particolarmente utili per garantire il buon successo delle
iniziative proposte e la corretta allocazione delle risorse disponibili. Ciascun progetto deve
essere valutato in relazione al suo grado di allineamento alla mission, alla vision e ai valori
aziendali; tale valutazione deve realizzarsi sulla base di informazioni aggiornate e
quantificabili, escludendo quindi i progetti che godono esclusivamente della
sponsorizzazione da parte di figure manageriali di vertice.
La figura 3 evidenzia anche la presenza di unattivit di monitoraggio continuo
durante le fasi del ciclo di vita del progetto, con lobiettivo di verificarne lallineamento
con la strategia aziendale. Questultimo punto sottolinea la relazione biunivoca tra PM e
strategia aziendale, nonch limportanza di interazioni e feedback da parte dei project
manager, in quanto i progetti da realizzare influenzano, sulla base degli stimoli ambientali,
i vantaggi competitivi dellimpresa.
La visione del rapporto tra PM e strategia trattata anche nel libro di Cantamessa
et. al (2007), in cui viene aggiunta unulteriore chiave interpretativa, basata sulla visione
resource-based dellimpresa. Secondo questo approccio le imprese si distinguono non solo
sulla base delle risorse che le compongono, ma anche e soprattutto in funzione delle
routine organizzative che attivano e collegano tali risorse e che costituiscono la vita
ordinaria dellimpresa. Alle routine organizzative riconosciuta la capacit di assicurare
alle imprese un vantaggio competitivo, eventualmente introducendo cambiamenti in
sintonia con il contesto esterno di riferimento. Nella visione resource-based il PPM
assicura il collegamento tra la strategia aziendale e lattivit operativa includendo nel
project portfolio sia progetti orientati al cambiamento sia progetti di tipo operativo, in cui
venga per valorizzata la componente di apprendimento, quella cio che consente
allimpresa di conservare il proprio vantaggio competitivo. Cantamessa et al. (2007) nel
descrivere la visione resource-based dellimpresa, sottolineano la difficolt che esiste nel
definire il project portfolio, a causa dei vincoli rappresentati dalle risorse disponibili. La
soluzione proposta dagli autori prevede una collocazione dellattivit di PPM sotto
linfluenza diretta dellalta direzione. In generale la letteratura concorde nellattribuire
una collocazione in chiave strategica della struttura che si occupa della gestione dei
progetti. Kaplan e Norton (2005) si spingono oltre, proponendo unulteriore tipologia di
unit organizzativa, lOffice for Strategy Management (OSM), con lobiettivo specifico di
colmare i gap che esistono tra la formulazione della strategia aziendale e la sua messa in
pratica. Nonostante lOSM non sia una struttura organizzativa tradizionalmente legata ai
principi metodologici del PM, larticolo utile soprattutto come riflessione sulla necessit
di mettere in atto modalit efficaci di integrazione quando la necessit di gestire progetti
complessi, allineandoli alla strategia, diventa unesigenza stabile per almeno una parte
dellorganizzazione.

15
Kaplan e Norton (2005) presentano lOSM come ununit organizzativa snella e
flessibile, a diretto contatto con il vertice aziendale, il cui compito di superare il
frazionamento organizzativo, attraverso la comunicazione della strategia aziendale, la
revisione e il monitoraggio periodico degli obiettivi strategici, ladozione di standard
uniformi per la valutazione della performance, il pieno coinvolgimento delle funzioni
chiave dellazienda nel conseguire la strategia complessiva dellorganizzazione.

16
Altri riferimenti

Archibald, R.D., (2003), Project management, ed. Franco Angeli, Milano.


Aubry, M., Hobbs, B., (2007) A multiphase research program investigating project management offices
(PMOs): the results of phase 1, Project Management Journal, 38(1), 7486.
Block, R., (1983) The politics of projects, Yourdon Press, Prentice-Hall.
Cantamessa, M., Cobos, E., Rafele, C., (2007), Il project management, ed. Isedi, Torino.
Crawford, J. K. (2002), Project management maturity model: providing a proven path to project
management excellence, Dekker.
Dombkins, D. (2009) Redefining our profession Part 2: the history and future of project management,
PM World Today, Vol. XI (II).
European Commission, Europe Aid Cooperation Office, (2004) Project Cycle Management
Guidelines.
Gaddis, P.O. (1959) The project manager, Harvard Business Review, (May/June) 8997.
Kerzner, H. (2005), Using the Project Management Maturity Model: Strategic Planning for Project
Management, Wiley .
Project Management Institute (PMI), A Guide to the Project Management Body of Knowledge
(PMBOK), 2000.
Ward, J.L. (2009), PTC improves customer satisfaction through global project management, PM
World Today, Vol. XI (II).
Weaver, P. (2006), A brief history of scheduling back to the future, Conference paper.
Weaver, P. (2008), The origins of modern project management, PM World Today, Vol. X (III).
www.pmforum.org
www.pmi.org

17
Tratto da Gli Articoli di PM Forum www.pm-forum.it

Profilo storico
del project management
Dalle prime forme embrionali di gestione per
progetto alla situazione attuale e di trend del P.M.

(Parte prima: fino agli anni sessanta)

Autore: Alberto VILLA


Giugno 2001

Linsieme di strumenti organizzativi, gestionali e produttivi volti alla realizzazione di


grandi progetti viene oggi comunemente indicata come project management .
Sebbene anche grandi opere del passato abbiano richiesto forme embrionali di
project management, solo con il progetto Manhattan del 1942 che questa disciplina si
presenta in forme razionali e coscienti. Nei decenni successivi, caratterizzati da
realizzazione di grandi opere ingegneristiche, il project management si consolida nelle
societ impiantistiche e stabilisce anche regole formali di base. Infine, negli anni novanta
il project management conquista nuovi campi applicativi, come nelle telecomunicazioni,
nellenergetica, nei servizi e nella gestione dl obbiettivi strategici di business aziendali.
Questo articolo ripercorre la storia del project management indicandone i possibili
futuri sviluppi anche in funzione della certificazione professionale del project manager.

Sin dallet della pietra lumanit ha completato progetti intesi come impresa
temporanea dedica alla realizzazione di obiettivi con carattere di univocit attraverso
limpiego mirato di risorse e tecnologie.
Stonehenge lesempio pi antico e pi enfatico di un progetto di successo
realizzato al limite delle tecnologie e risorse disponibili. Con segno opposto, la biblica
torre di Babele il modello del progetto piagato da crisi organizzativa e di comunicazione
e perci condannato allinsuccesso.
Da allora ad oggi sono stati realizzati una serie innumerevole di progetti, ma solo
negli ultimi cinquantanni che si compreso che la loro realizzazione richiedeva processi e
strumenti organizzativi, gestionali e produttivi specifici, definiti nel loro insieme project
management.
La definizione e teorizzazione dei processi di management che si sviluppa tra la fine
dellottocento e gli anni quaranta la premessa necessaria per lo sviluppo delle teorie di
project management. Il bisogno che ne veicola lo sviluppo quello della gestione del
cambiamento che si verifica in questo periodo [I]. Ci si riferisce allo sviluppo di
tecnologie, infrastrutture, nuovi prodotti che avviene ad una velocit senza precedenti e
soprattutto con una compressione crescente dei tempi intertorrenti tra scoperta
scientifica e sua applicazione (figura 1).
Il contributo pi significativo in questo periodo di Henry Gantt (1861 - l919). l
suoi studi di programmazione delle attivit cantieristiche lo portano allideazione dello
strumento operativo che sar ribattezzato con il suo nome. Il diagramma di Gantt
permette di elencare in uno stesso grafico le attivit che compongono un progetto e di

Pag. 1
18
Tratto da Gli Articoli di PM Forum www.pm-forum.it

rappresentarne le
Figura 1 durate, le risorse
Lasso Temporale Tra Scoperta Scentifica ed Innovazione necessarie e in parte
100 i legami logici e di
monitorarne
90
<- 1. Produzione lavanzamento [2].
Elettricit
80
I diagrammi di
<- 2. Fotografia

70 Gantt si sono rivelati


uno strumento cos
60 potente che, a
distanza di
Anni

50
<- 3. Telegrafo ottantanni, sono
40 ancora saldamente in
uso.

<- 6. Produzione energia


30

<- 5. Penicillina
Sul piano

nucleare
<- 4. Dinamite

organizzativo, si

<-7.Transistor
20

10
inizi a comprendere
che una gestione
0 <----------- 1750 a 1800 ---------- <----------- 1800 a 1850 ----------- <----------- 1850 a 1900 ----------- <----------- 1900 a 1950 -----------
>
efficace dei progetti
Periodo
richiedeva una
responsabilit unica e
focalizzata, che superasse le divisioni delle competenze funzionali aziendali. della
seconda met degli anni Trenta la creazione della figura di project engineer alla Exxon.
Sebbene questo rappresenti un importantissimo passo, si deve sottolineare come la
figura del project engineer avesse solo funzioni di coordinamento, senza alcun potere
effettivo.
Quasi contemporaneamente, una figura professionale simile introdotta in Bechtel,
sebbene con maggiore enfasi sulla responsabilit integrati-va del progetto.

Il Progetto Manhattan

E solo con il progetto Manhattan che i concetti organizzativi e gestionali tipici del
project management iniziano a essere applicati razionalmente [3] 1
Il progetto, iniziato nel 1942, ha portato in soli 700 giorni dalla prima fissione a
catena ottenuta da Fermi alla realizzazione delle prime due bombe atomiche sganciate su
Hiroshima e Nagasaki.
Al di l del giudizio morale sul loro utilizzo, il programma ha superato difficolt di
ordine scientifico, tecnologico e organizzativo senza precedenti e solo lirrepetibile
motivazione di arrivare prima del Terzo Reich ha potuto renderlo di successo. Basti
pensare che il progetto ha richiesto lo sviluppo delle basi teoriche della teoria della fisica
dei reattori nucleari e della chimica del plutonio, la costruzione di reattori plutonigenici e
la realizzazione di impianti di arricchimento di uranio 235.
Il project manager, gen. L. M. Groves, riportava direttamente al Presidente Roosvelt
che agiva da project sponsor.
Fondamentale stato il contributo delle maggion societ di ingegneria del tempo.
Stone & Webster, Bechtel, Kellog, DuPont, che hanno fornito strutture organizzative,
risorse umane e capacit esecutiva per questa complessa impresa. Importanti scelte
organizzative ne fanno il primo caso di applicazione razionale dei principi dei project
management:

1
E importante notare che questo salto si verifica come risposta a un programma caratterizzato dalla
complessit, innovazione e importanza vitale per la sopravvivenza della stessa democrazia

Pag. 2
19
Tratto da Gli Articoli di PM Forum www.pm-forum.it

o obiettivi del programma chiaramente identificati e definiti;


o ciascuna parte significativa del progetto assegnata a unorganizzazione
responsabile con uno specifico programma temporale derivato dal programma
generale; la figura 2 riporta la parte alta della WBS (Work Breakdown Structure)
del progetto;
o chiara allocazione delle responsabilit a tutti i livelli; lautorit era delegata con
attribuzione dei poteri e delle responsabilit conseguenti;
o utilizzo delle strutture produttive esistenti, cos che ogni entit partecipante
doveva mettere a disposizione le proprie competenze ed esperienze e non
inventare da zero;
o il gen. Groves e il suo staff a esercitare il ruolo di decisore e integratore delle
macroattivit del progetto, supportato da Stone & Webster a cui era stato
assegnato un contratto di consulenza simile agli attuali PMC (Project
Management Contract).

Project Manager
_________________
Gen. L.M.Groves

Stone & Webster


(Project management consultant)

Impianto Arricchimento Impianti Separazione Assemblaggio Bomba


Uranio 235 Plutonio
OAKRIDGE Handford Los Alamos

SEPARAZIONE SEPARAZIONE IMPIANTO REATTORE THIN MAN FAT MAN


ELETTROMAGNETICA U235 GASSOSA RIPROCESSAMENTO GRAFITE
WESTINGHOUSE DU PONT
EASTMAN KODAK KELLOG DU PONT BECHTEL Hiroshima Nagasaki

Fig.2 Work Breakdown Structure del Progetto Manhattan

Gli anni Cinquanta


Gli anni Cinquanta vedono la frenetica ricostruzione post bellica, con la conseguente
crescita economica caratterizzata da importanti realizzazioni impiantistiche e
infrastrutturali. Sul piano politico lintenso confronto della guerra fredda fra Occidente e
Oriente porta a una corsa agli armamenti mai registrata precedentemente nella storia
delluomo.
Lo sviluppo impiantistico vede le societ dingegneria progredire nellorganizzazione
per progetti, identificando un punto integrativo di responsabilit complessiva (il project
manager), che utilizza competenze e risorse aziendali condivise con altri progetti. La
figura del project manager costituisce levoluzione di quella del project engineer creata
negli anni Trenta alla Exxon. Lesigenza di comprimere i tempi di realizzazione porta allo
sviluppo e consolidamento della metodologia di realizzazione fast track dei progetti
impiantistici.
Il metodo fast track richiede che le attivit dingegneria, acquisti e costruzione siano
sviluppate parzialmente in parallelo. La conseguente accresciuta necessit di
pianificazione, controllo e integrazione esercitata dal team di progetto che collabora con
il project manager.
Lintroduzione delle metodologie di project management nelle societ di ingegneria
va di pari passo con quelle nelle societ di committenza. Exxon, Mobil, Shell, per citare le
maggiori, si organizzano per utilizzate il project management per la realizzazione degli
investimenti operati con contratti per lo pi rimborsabili.
Se notevole il debito dello sviluppo di project management alle societ d
ingegneria c da riconoscere che negli anni Cinquanta il contributo pi importante

Pag. 3
20
Tratto da Gli Articoli di PM Forum www.pm-forum.it

arrivato dal settore della difesa e spaziale degli Stati Uniti [3]. Infatti, sollecitazioni
politiche e la pressione dellopinione pubblica, ossessionata dalla psicosi della minaccia
sovietica e del suo possibile vantaggio tecnologico, spingono per la messa a punto di
sistemi gestionali e di strumenti di pianificazione che permettano il rispetto dei tempi di
realizzazione dei progetti militari del dipartimento della difesa, con i programmi Atlas e
Polaris si rivedono totalmente i sistemi organizzativi. LUSAF (United States Air Force)
obbligata a organizzarsi per progetti.
Il concetto di ciclo di vita utilizzato per organizzare per fasi successive
lesecuzione di un progetto. Nella fase di definizione lobiettivo produrre le specifiche
funzionali del sistema e il progetto concettuale. Nella fase di front end si richiede di
pianificare dettagliatamente lesecuzione, valutare i tempi e stimare i costi di
realizzazione sulla base del progetto di base, sviluppato contestualmente. Nella fase
successiva si eseguono lingegneria, gli acquisti e la fabbricazione del prototipo prima di
procedere alla produzione di serie. Vengono completamente definiti il progetto di
committenza e il progetto del main contractor, responsabile dellimplementazione del
sistema (system supplier).
Siamo cos entrati nei moderni concetti di project management
Lurgenza percepita nei programmi militari e lesigenza di un migliore controllo
temporale ed economico dei progetti portano la Marina USA e la societ DuPont a
sviluppare indipendentemente due metodi di pianificazione e controllo temporale,
denominati rispettivamente PERT (Project Evaluation Review Technique) e CPM (Critical
Path Method). I due metodi sono stupefacentemente molto simili, sebbene siano stati
sviluppati in modo totalmente indipendentemente. In maniera generica vengono definiti
sistemi di controllo reticolare.
Lo sviluppo contemporaneo dei due sistemi non casuale, ma conseguenza
dellintroduzione commerciale dei grandi computer e della grande capacit di calcolo
relativa al tempo che li caratterizzava. Lidea alla base un modello matematico che
descriva le attivit e le sequenze logiche che le legano nella realizzazione di un progetto.
Il sistema di equazioni del modello matematico, i cui coefficienti contengono le durate
temporali stimate di ciascuna attivit, permetter di calcolare il minimo tempo di
esecuzione del progetto identificando la sequenza di attivit che lo caratterizzano
(cammino critico). Si pu ottenere anche una probabilit di completamento del progetto
in determinati tempi attribuendo funzioni di distribuzione della durata di ciascuna attivit.
Nella formulazione iniziale il PERT poteva operare il solo controllo temporale,
mentre il CPM permetteva anche considerazioni sulluso ottimale delle risorse.
Come tentativo di migliorare questi modelli allUniversit di Standford si studia nei
primi anni Sessanta il precedence diagramming in cui le attivit sono rappresentate da
cerchi collegati connessi tra loro da linee.
Una delle caratteristiche pi importanti del precedence diagramming che il legame
tra due attivit viene specificato dando lo sfasamento tra i relativi inizi o completamento.
Il precedence diagramming conosciuto come reticolo con attivit nei nodi mentre il
PERT/CPM come reticolo con attivit nelle frecce di collegamento.
Questa differenziazione si poi stemperata nei modelli reticolari incorporati nei
programmi di controllo temporale utilizzati nei personal computer di cui parleremo nel
seguito.
A testimoniare il grande sviluppo degli anni Cinquanta, la rivista Harvard Business
Review pubblica nel 1959 larticolo The Project Manager di P O. Gaddis. Larticolo
esamina il ruolo innovativo svolto dal project manager nellorganizzazione, le modalit di
empowerment, i limiti della delega a volte meno ampia della responsabilit, i metodi
gestionali e anche comportamentali, con particolare riguardo alla soluzione dei conflitti.
Costituisce in questo senso il riconoscimento ufficiale di un ruolo, quello del project
manager fondamentale e centrale nella gestione progetti.

Pag. 4
21
Tratto da Gli Articoli di PM Forum www.pm-forum.it

Anni 50 Anni 60 Anni 70 Anni 80 Anni 90

Diagramma di Gantt (1916)


Project Engineer (Exxon 1935)
Work Breakdown Structure
System Support Contractor (USAF)
Ciclo di Vita del Progetto (USAF)
Project Manager Office (USAF)
Project Evaluetion and Review Method (PERT)
Critical Path Method (CPM)
Organizzazione Progetto a Matrice
Organizzazione Progetto a Task Force
Cost Schedule Control System (CSCS)
(sistema pianificazione DoD)
Value Engineering
Configuration management
Precedence diagrams (attivit nei nodi)
Concurrent Engineering
EPC lump sum turn key (contratti)
Teorie Gestione Conflitti
Fondazione Project Management Institute (PMI)
Fondazione Associazione Europea Project Management
(INTERNET poi IPMA)
Progetto aereo supersonico Concorde (inizio e fine)
Archibald Managing high tech projects
Kerner Project management handbook
Non recourse financing (finanziamento a Progetto)
Bul operate transfer project (BOOT)
Metodologia di design to cost
Life cicle costing
Assicurazione qualit progetti
Project Management body of knowledge (PMBOK)
Certificazione professionale project management PMI
Progetto tunnel sotto la Manica
Ruolo della committenza PM gestione investimenti
Fattori di successo del project management
Project management nuovi prodotti
Cleland Project Management handbook
Internet project start-up manual
Contratti di alleanza fra Cliente e Contrattore
Sistemi informativi di progetto su mini computer
Sistemi informativi di progetto su PC
Sistemi informativi aziendali (PC+LAN)
Sistemi informativi di progetto per team dispersi
(Intranet)

Fig. 3 - Tappe significative dello sviluppo del Project Management

Gli anni Sessanta


Gli anni Cinquanta hanno messo le fondamenta e gli anni Sessanta ci
consegneranno i sistemi di project management cos come li conosciamo oggi. Non a
caso la maturit del project management si coniuga a successi notevoli, primo di tutti il
programma Apollo coronato dal fantastico allunaggio di Armstrong e Aldrin il 20 luglio
1969. E anche il successo del segretario della difesa americano McNamara, che
sponsorizza lintroduzione di sistemi di gestione e pianificazione importanti, che

Pag. 5
22
Tratto da Gli Articoli di PM Forum www.pm-forum.it

elenchiamo nella loro denominazione originale: planning/programming and budgeting


system, lyfe cycling costing, quality assurance, value engineering, configuration control e
integrated control system [3].
Molti di questi sistemi erano stati definiti e sperimentati precedentemente in altri
settori e organizzazioni, come la Banca Mondiale per citarne uno. Tuttavia, il
Dipartimento della Difesa degli Stati Uniti (DOD) che li fa suoi, li sviluppa e li applica
compiutamente nellesecuzione di progetti. Il contributo pi importante , senza dubbio,
lo sviluppo di un sistema di pianificazione e controllo progetti, integrato nei tempi e nei
costi. Ci riferiamo al CCSCS (Contractor Cost Schedule Control System), che rappresenta
uno spartiacque nella pianificazione e controllo progetti.
Gli aspetti salienti sono:
il riconoscimento che il sistema di pianificazione e controllo ha necessit di operare in
un sistema formale di project management;
lidentificazione della WSS come elemento chiave per la pianificazione e il controllo
temporale ed economico;
pianificazione eseguita partendo dallalto per arrivare al particolare (top down) e
controllo iniziato dal basso della WBS per salire allalto (bottom up);
la definizione di una metrica che permetta di comparare lavanzamento effettivo ACWP
(Actual Cost Work Perforrned) sia con quello programmato BCWS (Budget Cost Work
Scheduled) e sia con quello effettivo valorizzato a costi di budget BCWF (Budget Cost
Work Performed)

In particolare, questultimo parametro definito earned valuee spesso con questo


nome si identifica, impropriamente, lintero sistema d pianificazione e controllo.
La filosofia sottesa che per uneffettiva comprensione dello stato del progetto
necessario conoscere la correlazione tra lavanzamento temporale e quello economico.
Ci reso possibile comperando il progress effettivo a costi di budget con quello effettivo
a costi impegnati e con quello pianificato a costi standard. I due valori forniscono,
rispettivamente, gli scostamenti temporali ed economici della singola attivit o del singolo
progetto. Su questa base si possono definire indici che descrivono lo stato del progetto e
si pu formulare una corretta stima al completamento. Ad oggi nessun sistema o modello
ha sostituito o reso superati i criteri di pianificazione e controllo dellearned value.
Per quanto riguarda la teoria dellorganizzazione, in questo periodo si consolida la
convinzione che limplementazione di un progetto con successo legata allambiente
organizzativo e che alcune scelte esaltano lefficacia del sistema di project management
mentre altre la deprimono.
Molti sono i lavori teorici e di analisi comparativa delle esperienze acquisite che
fanno progredire notevolmente la cultura organizzativa di project management. Si discute
molto sulla necessit di integrazione e coordinamento in funzione della complessit del
progetto stesso [5]. P. Lawrence e J. Lorsh, in uno studio sulla gestione delle tecnologie e
progetti, giungono alla conclusione che maggiore lincertezza ambientale e tecnica tanto
pi sentita la necessit di project management.
Molti studi hanno analizzato le strutture organizzative a task force e a matrice,
identificandone punti di forza e debolezza, soprattutto in funzione del ruolo e dellautorit
del project manager.
Il modello classico di Youker ne fa splendida sintesi. J. Galbraith applica queste
teorie alla Boeing, partendo dai modelli integrativi e proponendo le relative risposte
organizzative.
Lesame dellesperienza di conduzione di vari progetti porta a concludere che la
diversit culturale e i diversi obiettivi tra chi opera nella struttura permanente e chi in
quella temporanea di progetto costituiscono causa di conflitto, soprattutto nelle
organizzazioni a matrice. Vengono analizzate le cause di conflitto e le strategie di

Pag. 6
23
Tratto da Gli Articoli di PM Forum www.pm-forum.it

soluzione adottate dai project manager. Si sviluppano teorie sulla positivit della
conflittualit, purch gestita allottenimento degli obbiettivi di progetto.
Viene focalizzata la teoria dei sistemi aperti, considerando le organizzazioni come
esseri viventi predisposte alladattamento delle evoluzioni ambientali attraverso forti
differenziazioni e interdipendenza [1].
In questi anni ladozione del project management diventa massiva nellindustria
impiantistica, in quella aeronautica spaziale e in quella delle costruzioni [6].
Approssimativamente si pu calcolare che queste industrie rappresentassero il 20% de l
Pil dei paesi progrediti. Lindicatore importante perch ci permette di concludere che gi
alla fine degli anni Sessanta circa un quinto della ricchezza generata era gestita con
tecniche di project management.
Questo indicatore ci aiuta a comprendere come i tempi fossero maturi per la nascita
di societ professionali di project management, che ne curassero la diffusione e fossero
riferimento per i professionisti del settore.
Negli Stati Uniti viene fondato il Project Management Institute (PMI) e in Europa
Internet che verr ribattezzata a met degli anni Novanta IPMA (lnternational Project
Managernent Association), per ragioni di omonimia con il sistema informatico telematico.
Importantissimo il ruolo che queste due associazioni avranno nello sponsorizzare,
diffondere e rafforzare la cultura di project management

Pag. 7
24
Tratto da Gli Articoli di PM Forum www.pm-forum.it

L Italia segue un percorso particolare, dettagliato nel riquadro.

Project Management In Italia

In Italia il Project Management arrivato negli anni60 con le grandi societ dingegneria che lo hanno
sempre considerato come uno del fattori critici ti successo del proprio business (realizzazione chiavi in mano
di impianti industriali). Infatti. operando gi da allora in un mercato globale, hanno dovuto acquisire
rapidamente eccellenza nei processi di gestione progetti per raggiungere il successo competitivo e reddituale
che le caratterizza.
Mancando la grande committenza militare ed aerospaziale, il paese deve interamente a questo settore
industriale ed in particolare alla sua associazione di categoria ANIMP, lazione di diffusione della cultura di
gestione progetti.
LANIMP, costituita nel 1974, ha esercitato la sua azione promozionale sia con la rivista, con testata
Impianti sino al 1988 ed Impiantistica Italiana poi, sia con i convegni, in cui ha favorito scambio
desperienze e osmosi della cultura di project management ad altri settori produttivi dellindustria e del
terziario.
Le tappe formali di questo contributo iniziano nel 1982 in cui LANIMP riconosciuta come il
rappresentante Italiano dallassociazione europea di project management. INTERNET.
Nel 1984 lANIMP organizza il primo convegno italiano di project management a Sorrento con ampia
partecipazione di altri settori produttivi oltre allimpiantistica a denotare una crescita gi avanzata. Nel 1986,
viene fondata la Sezione di Project Management da 24 soci fondatori.
Nel 1992 la sezione di PM organizza con successo il congresso mondiale di project management a Firenze
con partecipazione di oltre 100 delegati da tutto il mondo.
Molti sono i contributi realizzati dalla sezione di project management
Vale la pena ricordare qui i seminati di Project Management e la certificazione professionale dei project
manager.
I Corsi e i Seminari sono iniziati nel 1985 e proseguono con successo. Sebbene mirati inizialmente alla
formazione dei dipendenti delle societ di ingegneria a seguito della costituzione delle altre sezioni Animp,
hanno visto ben presto anche la partecipazione di societ ed organizzazioni provenienti da altri settori
produttivi o a servizi.
La Certificazione Professionale e stato un obiettivo che la sezione di PM. persegue dal lontano 1989, con la
creazione di gruppi di lavoro ad hoc, che hanno visto la partecipazione di molti esperti appartenenti ad
aziende italiane che avevano gi introdotto i criteri di gestione per progetti.
Questo lungo processo si concluso alla fine del 1999 e nel 2000 sono iniziate le certificazioni dei
responsabili di progetto.

Lassenza di un settore militare e spaziale capace di fare committenza lascia alle


sole societ di ingegneria luso del project management. Enorme il debito che il paese
deve a questo settore e alla sua Associazione di categoria: lANIMP.

Bibliografia
[1] Minati G: Knowledge management. PSOA editore,1987
[2] Archibald R.D.: Managing High technology Programs and Projects. John Wiley & sons, 1986
[3] Morris PW: The management of Projects. Thomas Telford, 1994
[4] Lepri F: Lazienda liberata. Andrea Tentati Editore, 1995
[5] Stuckembruck LC: Project Managemenr. PMI, 1983
[6] Kimmons RL: Project Management, a Refarence for Professional. Dekker, 1989

Fine prima parte

Pag. 8
25
Tratto da Gli Articoli di PM Forum www.pm-forum.it

Profilo storico
del project management
Dalle prime forme embrionali di gestione per
progetto alla situazione attuale e di trend del P.M.

(Parte seconda: dagli anni settanta ai nostri giorni)

Autore: Alberto VILLA


Luglio 2001

Gli anni Settanta

Gli anni Settanta vedono il project management consolidarsi definitivamente in tutti


i settori dellimpiantistica e delle costruzioni. Pi in generale, la grande incertezza
ambientale conseguente ai notevoli ratei inflativi, le oscillazioni economiche quali
laumento del prezzo del petrolio del 1973 e la conseguente crisi economica marcano
definitivamente il project management come la tecnica organizzativa che meglio
permette di far fronte ai mutamenti degli scenari di riferimento.
negli anni Settanta che la gestione progetti impara che nellambiente esterno al
progetto esistono importanti decisori (gli stakeholder) che vanno gestiti.
Da lezioni dolorose Cleveland e altri teorizzano come pianificare e gestire il
coinvolgimento degli stakeholder [9].
Lo esperimentano i progetti europeo e statunitense di aereo passeggeri
supersonico, rispettivamente Concorde ed SST. Lamericano SST gestito senza opportuno
coinvolgimento dellopinione pubblica viene dismesso perch il Senato non concede
lestensione dei fondi per la sua esecuzione. Il vettore europeo, concepito prima della
crisi energetica quando si pensava che il fattore critico del trasporto fosse la velocit,
riesce solo nel 1976 a ottenere lautorizzazione di atterraggio negli Stati Uniti.
Una cattiva gestione del progetto di committenza e la rincorsa allinnovazione
tecnologica ne allungano i tempi di realizzazione e moltiplicano esponenzialmente i costi
consuntivi rispetto al preventivo iniziale.
Limportanza e il ruolo dellopinione pubblica riaffermato dal caso del
megaprogetto Trans Alaskan Pipeline System (TAPS). Il progetto di circa 8 miliardi di
dollari per il trasporto del petrolio dai ricchissimi giacimenti del Nord sino al porto al Sud
dellAlaska non avrebbe mai ricevuto lautorizzazione per la realizzazione a causa delle
opposizioni ambientaliste se non fosse arrivato lo shock energetico del 1973 con
lembargo proclamato dai Paesi OPEC agli Stati Uniti. Dal punto di vista del project
management, il TAPS costituisce un esempio eccellente di project management di
committenza e rimane un esempio per tutti i megaprogetti.
I progetti di investimento nel Mar del Nord rappresentano un altro anello della
spirale virtuosa di crescita del project management [9]. Il prezzo elevato del greggio
negli anni Settanta ha reso profittevole lestrazione nei giacimenti sottomarini nel Mar del
Nord, caratterizzato da severissime condizioni ambientali. Il rischio e la complessit
esecutiva hanno spinto i committenti a cercare esperienza e capacit di project
management con contratti lump sum di servizi di project management (PMC) da
utilizzare nella gestione di alcuni contrattisti.
Sebbene il ciclo virtuoso che lapplicazione del project management instaura negli
investimenti impiantistici sia una costante in questi anni, impressionante il cattivo

Pag. 1
26
Tratto da Gli Articoli di PM Forum www.pm-forum.it

utilizzo fattone nellimpiantistica nucleare, soprattutto negli Stati Uniti [3, 9]. Principali
indiziati sono stati i sistemi di project management di alcuni enti elettrici nel gestire gli
investimenti in impianti nucleari.
Numerose ed esemplari sentenze delle Corti americane hanno stigmatizzato
linefficienza dei loro sistemi di management, ritenendoli alla base dellescalation dei costi
e dei ritardi nel completamento dei progetti e impedendo che i costi di questa inefficienza
venisse scaricata sui cittadini con aumento di tariffe (figura 4) 1

Questa sentenza del tribunale amministrativo di New York del 31 marzo 1985 molto importante.
Infatti evidenzia come la corretta applicazione del project management sia entrata a far parte della dottrina
giuridica al punta tale che possibile giudicare se stata esercitata la dovuta diligenza nella gestione di un
progetto di investimento.
La sentenza si riferisce alla gestione del progetto dellimpianto nucleare Shorenam esercitata dal
committente. Sulla base di questa sentenza stato proibito allente elettrico Lilco di ammortizzare nei prezzi
dallenergia erogata i 1200 MM$ dl costi consuntivi eccedenti la stima di investimento originale.
Lammortamento stato assorbito dal capitale di rischio in termini di valore della singola azione e
dividendi attribuiti agli azionisti. Ha garantito quindi ai cittadini dello stato di New York di non essere
penalizzati dalla cattiva gestione del progetto di investimento del loro ente elettrico.

Recita la sentenza:

Lente elettrico (Long Island Lighting Company) non ha ottemperato al dovere di sviluppare un
piano di progetto adeguato per:
supervisionare (oversee) la gestione del progetto eseguita da Stone & Webster
identificare ruoli e responsabilit
sviluppare un accurato ed appropriato sistema di reporting che permettesse di monitorare,
misurare e controllare costi e tempi
staffare adeguatamente il team di supervisione e prepararlo alla missione di sorveglianza
richiesta al committente
Pertanto ne deduciamo che durante tutta lesecuzione del progetto, lente elettrico non ha esercitato la
cura e diligenza che si richiede al committente nel controllo del tempi e costi del progetto.
Perci le limitate informazioni sullo stato del progetto che venivano presentate al consiglio di
amministrazione erano inadeguate per assumere decisioni corrette ed esercitare un indirizzo appropriato sul
progetto Shorenam.

Fig.4 - Stralcio dallo sentenza del tribunale amministratIvo dallo stato di New York del 11 marzo 1985, caso
27563, relativo al sistema di project management dellente elettrico newyorkese Long lsland
Lighting Company

Ci sono varie ragioni che hanno portato a questo colossale fallimento:


lutilizzo dellenergia nucleare stata promossa dai singoli Governi in
considerazione della sua presunta strategicit; questo ha favorito la
creazione di mercati nazionali protetti, la creazione di oligopoli e reso difficile
il controllo sociale del relativo sviluppo;
tra il 1964 e il 1975 c stata unesplosiva crescita dellindustria nucleare negli
Stati Uniti (211 unit ordinate per complessivi 180.000 MWe), con limitata
esperienza nel gestire progetti cos rischiosi e incapace di valutare rischi reali
generati da fattori di scala di queste dimensioni;
lente di sicurezza (NRC) che, sulla base della crescente protesta degli
antinuclearisti, diventato generatore continuo di nuovi criteri con effetto

1
I progetti di impianti nucleari hanno mostrato escalation senza precedenti dei costi, con il massimo toccato
dalla Utility dello stato di Washington. Il costo di cinque impianti passato da 4 bn di dollari del 1973 a 24 bn di
dollari nel 1981. In generale i tempi di realizzazione degli impianti sono passati da 6 a 10 anni.

Pag. 2
27
Tratto da Gli Articoli di PM Forum www.pm-forum.it

retroattivo e conseguente disruption nello sviluppo dellingegneria e


costruzione.

Gli anni Ottanta

Gli anni Ottanta vedono il raggiungimento della piena maturit dei metodi di project
management e la loro diffusione ad altri settori industriali e ai servizi. Lindustria
informatica lo adotta estensivamente nella realizzazione di software e, pi in generale,
nellimplementazione di sistemi informativi. Vari settori dellindustria manifatturiera, per i
quali il tempo di introduzione al mercato di nuovi prodotti costituisce un fattore critico di
successo, lo adottano come parte della loro strategia competitiva. Questo
particolarmente vero per i settori farmaceutici e automobilistici [12].
Il nuovo prodotto gestito come un progetto concede inizialmente significativi
vantaggi competitivi a chi adotta questa organizzazione. Ricordiamo, a titolo di esempio,
la societ giapponese Toyota, che per prima ha ridotto il tempo al mercato di un nuovo
modello di autovettura dai 60 ai 36 mesi, con riduzione delle ore di ingegneria da 3 a 1.8
milioni.
Sebbene la drastica riduzione di tempi e di risorse utilizzate non sia attribuibile
solamente allapplicazione del project management, tuttavia costituisce una indicazione
forte dei vantaggi che ne derivano. Non a caso si accentua il trend di diffusione di questa
metodologia sino alle aziende che operano nei servizi, al settore non-profit e alle
organizzazioni di tipo governativo.
Levoluzione del hardware e del software [7] che si realizza con lintroduzione dei
personal computer porta alla nascita di programmi commerciali di project management
autosufficienti (stand alone) e molto orientati agli utenti. Sono soprattutto sistemi di
programmazione temporale (scheduling system), ma anche programmi di controllo
economico e di gestione documenti. La caratteristica di essere installati sul singolo PC
impedisce spesso la raccolta dei dati aziendali e comunque luso integrato di un data base
omogeneo di progetto. Il problema verr superato negli anni Novanta con la diffusione
delle reti aziendali e dei sistemi client/server
Nel campo petrolifero e della generazione di energia si sempre pi diffusa la
tendenza a far ricorso a finanziamenti non recourse, concessi, sulla base del solo cash
flow generato dal progetto, alle societ create per la gestione del business specifico
(project company). Il progetto con finanziamento non recourse ha posto forte il problema
di un project management pi strategico, orientato alla redditivit dellinvestimento e ai
controllo dei rischi relativi e non solo al completamento nei tempi e costi
dellinvestimento.
La crisi debitoria dei paesi terzo mondo, drammaticamente iniziata dal Messico con
la dichiarazione di insolvenza nellagosto 1982, ha significativamente cambiato le
strategie di project financing e ha spinto a rafforzare le attivit di risk management.
Tralasciando le implicazioni sulla gestione degli enti finanziari, sempre di pi si richiesto
che i progetti evidenziassero una effettiva salute economica e finanziaria, verificata anche
attraverso una estesa analisi di rischio. Le fasi di front end dei progetti sono state quindi
meglio affrontate, soprattutto negli aspetti di preventivazione, pianificazione e definizione
delle strategie di implementazione.
In questa direzione vanno anche i progetti BOOT (Build Own Operate Transfer),
mirati soprattutto alle infrastrutture. Prendono avvio dalle serie restrizioni finanziare che i
Governi hanno sperimentato durante gli anni Ottanta, ma anche dal periodo favorevole
alle privatizzazioni messe in atto in settori non tipici delle organizzazioni statali. Lidea
che sorregge il BOOT che si formi una societ che gestisca lintero ciclo del business,
trovando i finanziamenti, seguendo la costruzione ed esercendo il sistema in maniera

Pag. 3
28
Tratto da Gli Articoli di PM Forum www.pm-forum.it

profittevole per realizzare il ritorno dellinvestimento2. Al termine lopera viene trasferita


allo Stato senza oneri.
Questo tipo di progetti richiede una infrastruttura redditiva dal punto di vista
finanziario, una organizzazione governativa recettiva, un gruppo di investitori privati che
pongano il capitale di rischio, un consorzio di societ di ingegneria e costruzione con
adeguata esperienza per la realizzazione e un gruppo di istituzioni finanziarie che
finanzino il debito su basi non recourse.
I primi esempi di BOOT sono stati il canale sotto la manica e il secondo
attraversamento del porto di Hong Kong.
Le forme di imprenditorialit che caratterizzano queste project company sono
centrate sul project management non solo nella fase di committenza, ma anche in quella
di esercizio del business.
Le difficolt incontrate dalla societ del canale della manica nellespletare le attivit
di committenza sono state contemporanee allaffermazione del concetto delle
responsabilit del committente nelleseguire gli investimenti capitali. A questo proposito
sono risultate fondamentali le sentenze delle Corti americane, che hanno negato ad
alcuni enti elettrici USA lammortamento sulle tariffe di molta parte degli investimenti per
la realizzazione di impianti nucleari con la motivazione che molti dei costi incrementali
erano stati generati dalla carente gestione del progetto del committente.
La centralit del committente per una gestione efficace degli investimenti poi
stata ribadita dagli studi del CII (Construction lndustry Institute), una organizzazione di
ricerca creata negli Stati Uniti per cercare di rimediare a due decenni di costante
decremento della produttivit nella industria delle costruzioni. Due i concetti sviluppati in
quegli anni dal CII e risultati di particolare importanza: constructability e partnership.
La constructability nasce dal concetto che risparmi, tempi di esecuzione pi brevi e
miglior uso delle tecnologie di costruzione si potesse ottenere attraverso una
progettazione e pianificazione attenta alle esigenze della costruzione.
La partnership tra committente e contrattore nasce con la finalit di superare la
conflittualit che pu nascere nellesecuzione di contratti e che, al termine del progetto, si
traduce spesso in claim e contenziosi, che arrivano sino alle camere di arbitrato. La
soluzione prospettata quella di sviluppare legami di lungo periodo, al di l degli specifici
progetti, che permettano una collaborazione concreta tra i due team di progetto e un
orientamento di reciproco sostegno per il raggiungimento degli obbiettivi condivisi.
La strategia di partnership evolver poi nei contratti di alleanza strategica negli anni
Novanta. La significativa esperienza acquisita e la relativa banca dati sui progetti
implementati negli anni Settanta e Ottanta ha portato a definire i fattori critici di successo
nei progetti.
Sia il PMI sia lassociazione di project management europea IPMA compiono sforzi
significativi per mettere a punto standard di project management con lobbiettivo di porre
le basi per definire indicatori di prestazioni, sintetizzare lo stato dellarte della professione
sia in termini di conoscenze teoriche che di pratiche organizzative e gestionali
comunemente adottate.
Di assoluta rilevanza il PMBOK (Project Management Body Of Knowledge) [13]
sviluppato dal PMI nella seconda met degli anni Ottanta, successivamente revisionato e
divenuto standard ANSI nel 1998.
Il PMBOK si pone lobbiettivo di sommare le teorie e competenze del project
management. Lo fa descrivendo un modello di project management organizzato in otto
aree di conoscenza focalizzate secondo cinque processi di management: avviamento,
pianificazione, controllo, esecuzione e chiusura.
Le nove aree di conoscenza sono: project integration, time management, scope
management, cost management, quality management, risk management, human
resources managernent, communication managemnent e procurement management

2
Riscotendo un pedaggio o comunque un prezzo del servizio

Pag. 4
29
Tratto da Gli Articoli di PM Forum www.pm-forum.it

Il PMBOK stato utilizzato come sistema di riferimento per valutare le conoscenze e


competenze dei project management. La maturit raggiunta dal project management si
coniuga con la necessit del cliente di avere sempre pi garanzie che il project manager
proposto sia adeguato alla missione. Parimenti chi opera nel settore desidera che le
proprie competenze vengano riconosciute.
Per soddisfare queste due esigenze viene lanciata dal PMI nel 1985 la certificazione
professionale, che ha avuto una partecipazione ristretta nei primi anni, ma poi una rapida
diffusione per arrivare ad oggi a circa 20.000 professionali certificati.
Anche la IPMA, che il punto di riferimento nel Project management di tutti i paesi
europei, ha lanciato una propria certificazione professionale e dal 1996 certifica project
manager, e dai 2000 presente in Italia come Certificazione ANIMP/IPMA.
A differenza della certificazione PMI, che impostata secondo i criteri tipici della
realt statunitense, la certificazione ANIMP/IPMA caratterizzata dallapproccio tipico
della cultura europea.

Gli anni Novanta

Negli anni Novanta, con la raggiunta maturit, il project management ha vissuto


una impressionante e salutare crescita sia nei settori di applicazione sia nella diffusione,
anche nei paesi in via di sviluppo.
Lo sviluppo dei sistemi software e pi in generale dellInformation Technology (IT)
ha potenziato e migliorato la facilit duso dei sistemi di project control e la sua
integrazione con i sistemi di azienda, sino a disporre di vere e proprie suite o MIS di
progetto [7].
Lofferta commerciale si differenziata, presentando sia semplici software stand
alone installati su pc sia sistemi integrati, capaci di applicazione client/server con moduli
di programmazione temporale, controllo costi, gestione delle risorse, comunicazione,
interfacce grafiche e risk analysis. Senza dubbio questi sistemi informativi permettono
una maggiore efficacia nella gestione progetti, come:
processo decisionale pi informato;
miglior integrazione come conseguenza della disponibilit di DB di progetto;
uso ottimizzato delle risorse;
simulazioni di progetto pi realistiche (what if):
migliore e pi accurato reporting:
migliore e pi estesa informazione ai partecipanti al progetto.

Va poi considerato che laccresciuta capacit di raccogliere ed elaborare dati e di


sintetizzarli a pi livelli ha liberato energie allinterno dei project team verso una
maggiore attenzione alla comunicazione e alle relazioni, instaurando un circolo virtuoso
nel controllo del progetto stesso [4]. Inoltre, deve essere sottolineato il ruolo positivo
svolto dalle societ di software nel diffondere la cultura e lapplicazione del project
management in molti settori non tradizionali: servizi, consulenza, ecc.
Il project management si esteso nellapplicazione a progetti critici per la strategia
competitiva di azienda e i sistemi di controllo si sono estesi a includere aspetti pi
strategici. quali quelli richiesti dal management by project. In questa ottica, particolare
attenzione stata dedicata alla definizione del portafoglio progetti e al suo costante
controllo per assicurarne lallineamento con le strategie dazienda [11,12].
In accordo a Cleland [9], negli anni Novanta il project management si mosso dai
settori tradizionali, quali ingegneria e difesa, per diventare il cardine centrale nella
gestione dei processi competitivi nella direzione di impresa, quali:
re-engineenng dei processi produttivi e di business per aumentare le capacit
competitive dellimpresa:
utilizzo di project team per lintroduzione di nuovi prodotti, servizi e tecnologie in
tempi pi brevi e a costi inferiori;

Pag. 5
30
Tratto da Gli Articoli di PM Forum www.pm-forum.it

analisi delle prenotazioni del business aziendale e adeguamento ai valori di


benchmarking dei leader di mercato:
sviluppo di nuovi business utilizzando team di progetto per analizzare il mercato,
realizzare alleanze e sviluppare la nuova unit di affari;
iniziative di Qualit Totale utilizzando team di progetto per integrare e
implementare le strategie di miglioramento di prodotti, processi e servizi.
Un esempio significativo di applicazione del project management al nuovo prodotto
quello del modello 777 della Boeing. Laereo stato progettato da pi di 235 team di
sottosistema geograficamente disperse. importante sottolineare che la parte del
collegamento e integrazione stata realizzata dallaccesso contemporaneo dei vari team
allelaborazione del modello 3-D dellaviogetto.
C da notare che molti dei 235 team appartenevano a fornitori che hanno
partecipato al progetto sviluppando in contemporanea i propri subsistemi.
Lo sviluppo accelerato delle nuove tecnologie e dei business collegati stato reso
possibile da un utilizzo massivo del project management. Ci riferiamo soprattutto alle
telecomunicazioni, Internet, computer e nuovi materiali.
I progetti complessi hanno ricevuto notevolissimo contributo dalle certificazioni di
qualit ISO 9001, 9002 e 9003. Infatti, sono caratterizzati dal contributo di molte entit,
soprattutto subcontrattori di ingegneria, fornitori e contrattasti di costruzione. I processi
di certificazione hanno creato tra tutte queste organizzazioni una condivisione
metodologica dei principi su cui basare i processi di lavoro. un allineamento che prima
della diffusione massiva delle certificazioni doveva essere svolto su base di progetto o su
iniziativa specifica di una azienda.
Il project management ha avuto un utilizzo pressoch totale nella gestione delle
alleanze strategiche e dei relativi contratti. Nel settore energetico, le grandi aziende
hanno mirato alla riduzione dei costi fissi, riducendo lo staff per la gestione degli
investimenti. Questo ha spinto ad alleanze con i contrattori di ingegneria per gestire la
realizzazione degli investimenti in impianti non pi possibile con le risorse interne.
Al contempo, rappresenta una opportunit per le societ di ingegneria per
espandere il modo con cui forniscono valore e servizi ai clienti.

Trend futuro

Lutilizzo di Internet, con le sue caratteristiche di uniformit di standard sottesi e di


ubiquit degli utenti, presenta opportunit importanti sia nellutilizzo lntranet (allinterno
dellorganizzazione) che Extranet (tra organizzazioni diverse). In questo sembra costituire
la risposta alla globalizzazione, ai team geograficamente dispersi, alla sempre pi estesa
partnership e allottimizzazione della supply chain.
prevedibile che ci saranno differenziazioni nelle metodiche di project management
per alcuni settori specifici di business e in particolare telecomunicazioni e , compreso
lnternet. Tuttavia, ragionevole prevedere che le omogeneit saranno superiori alle
differenze, come accaduto al marketing.
Le spinte in atto di globalizzazione e competizione globale porteranno sempre pi a
fusioni e joint venture con team di progetto multinazionale e comunque di culture e
spesso lingue madri diverse, team geograficamente dispersi, maggiore attenzione al
rischio e alla sua mitigazione, soprattutto investendo in maggiore pianificazione di
progetto (fase di front end) e alla sua allocazione (allaince e supply chain optimization).
Sul piano delle risorse umane, si sta intensificando lesigenza di una migliore
preparazione teorica dei project manager e del team di progetto, garantita, da una parte,
dalle certificazioni delle associazioni professionali e, dallaltra, da una educazione
universitaria o post universitaria. Gi molte universit stanno offrendo master in project
management e, addirittura, la North Caroline University lo offre on line. Parimenti,
lesigenza di formazione ha moltiplicato i corsi base e fatto avanzare offerte da societ di
formazione professionale.

Pag. 6
31
Tratto da Gli Articoli di PM Forum www.pm-forum.it

Il team di project management per sua natura temporaneo. Molte societ ed enti
che non considerano la gestione progetti tra i processi caratteristici del loro business
opteranno per loutsourcing. La necessit di reperire temporaneamente sul mercato
efficaci team di progetto e project manager aprir un segmento di servizi importante.

Conclusioni

L articolo ha rivisitato la breve e positiva storia del project management.


La sua dimostrata efficacia ne ha fatto uno strumento gestionale che rester a
lungo nei processi di gestione di imprese, enti di ricerca e organizzazioni governative.
Per chi, come noi, opera allinterno delle societ di ingegneria c anche lorgoglio
per il significativo contributo dato dal settore allo sviluppo e diffusione del project
management in Italia.

Bibliografia
[1] Minati G: Knowledge management. PSOA editore,1987
[2] Archibald R.D.: Managing High technology Programs and Projects. John Wiley & sons, 1986
[3] Morris PW: The management of Projects. Thomas Telford, 1994
[4] Lepri F: Lazienda liberata. Andrea Tentati Editore, 1995
[5] Stuckembruck LC: Project Management. PMI, 1983
[6] Kimmons RL: Project Management, a Reference for Professional. Dekker, 1989
[7] AA.VV.: Project Management Software Survey. PMI, 1999
[8] Kerridge AE: Engineering & Construction Project Management. Guilf publishing company, 1989
[9] Cleleand DI: Project Owners Beware. Project Management Journal n.5, 1986
[10] farewell E: The Next Generation of Projects management. PMI 1998 seminar, Long Beach, California,
USA, 1998
[11] Sommer RJ: Portfolio Management for Project. A New Paradigma. PMI seminar & Symposium, 1998
[13] A guide to the Project Management Body of Knowledge (PMBOK guide), 2000 edition, PMI supplement to
PM network.

Pag. 7
32
International Journal of Project Management 20 (2002) 111
www.elsevier.com/locate/ijproman

Current practice in project management an empirical study


Diana White, Joyce Fortune *
Centre for Complexity and Change, The Open University, Walton Hall, Milton Keynes MK7 6AA, UK

Abstract
This paper reports the ndings of a survey designed to capture the `real world' experiences of people active in project manage-
ment. The survey took the form of a questionnaire that was sent to 995 Project Managers and which achieved a response rate of
23.7%. Each respondent was asked to describe a recent project and identify factors that were regarded as critical to that project's
outcome. The extent to which the project gave rise to side-eects was explored and particular emphasis was placed on the use that
had been made of any of the many project management methods, tools and techniques that are available. Respondents were also
asked to judge the eectiveness of the methods, tools and techniques they had used and to report any limitations or drawbacks they
had encountered. The results showed that most respondents used only a small number of methods, tools and techniques with pro-
ject management software and Gantt charts being the most widely used aids. Almost half of the respondents reported drawbacks to
the methods, tools and techniques they had employed. The criteria for judging project success most cited in the project management
literature (on time; to budget; and to specication) were the criteria used by the respondents to judge their projects' success. How-
ever, two further criteria were reported as being of particular relevance. These were both concerned with the consequences of the
project on the organisation involve. In contrast to the nding of many surveys of project success rates, a remarkably high propor-
tion (41%) of the projects reported upon here were judged to be completely successful, though it should be noted that the judge-
ments were made by Project Managers who had worked on the projects being judged. # 2001 Elsevier Science Ltd and IPMA. All
rights reserved.
Keywords: Survey; Project management; Project success criteria; Methods, tools and techniques

1. Introduction delivered. It reports on the results of a recently con-


ducted survey designed to capture the `real world'
Project management is now well developed and well experiences of project managers with a view to deter-
accepted as a domain for the exercise of professional mining the extent to which those involved in the man-
expertise and as an area for academic research and dis- agement of projects actually make use of the methods
course. Numerous methods and techniques have been and techniques that are available and how eective the
developed, covering all aspects of managing projects methods and techniques used are felt to be.
from their genesis to their completion, and these have
been disseminated widely in books and journals and
through the work of professional bodies. However, 2. Research methodology
project management remains a highly problematical
endeavour. A great many projects exceed their budgets, In order to capture the experience of project man-
run late or fail to meet other objectives. Indeed, surveys agers, a questionnaire was designed that would:
[13] suggest that this applies to well over a half of IT
projects. . identify any common criteria used for dening
This paper aims to shed light on the reasons for the project success;
mismatch between the promise oered by project man- . explore the extent to which projects give rise to
agement methods and techniques and the outcomes unexpected side-eects;
. establish a common list of `critical success factors';
. identify the methods, methodologies, tools and
* Corresponding author. Tel.: +44-1908-652105; fax: +44-1908- techniques in current use in the eld of project
653718. management;
0263-7863/01/$20.00 # 2001 Elsevier Science Ltd and IPMA. All rights reserved.
PII: S0263-7863(00)00029-6
33
2 D. White, J. Fortune / International Journal of Project Management 20 (2002) 111

. identify any limitations or drawbacks to the 3.1.2. Industry sectors


methods, methodologies, tools and techniques The `Finance, Insurance and Banking' industry sector
being used. accounted for just over 16% (38) of responses. The
breakdown is shown in Fig. 2.
First, a pilot survey was conducted with a questionnaire Where projects were carried out for a client the
being sent to 30 project managers representing seven orga- respondent was asked to indicate the client's industry
nisations. 20 responses were received and used to generate sector. The breakdown is shown at Fig. 3. The `Trans-
revisions to the questionnaire which was then sent to 995 portation and Communication' industry sector accounted
project managers representing 620 organisations in both for just over 27% (24) of responses.
the public and private sectors. All of the recipients were
asked to take part in the survey only if they had been 3.1.3. Number of employees
actively involved in the management of a project and to The number of employees in the respondents' and
base their responses on their most recently concluded pro- (where applicable) the clients' organisations, classied
ject even if that project had been curtailed or abandoned. into ve groups, are shown in Fig. 4. Nearly 66% (155)
In order to maximise the chance of obtaining answers to of respondents worked for organisations that employed
sensitive questions the main survey was anonymous. 1000 or more employees.

2.1. Questionnaire design 3.1.4. Numbers involved


Respondents were asked to indicate the number of
The revised questionnaire comprised 18 questions people directly involved in the project from their own
with a mixture of yes/no, scale, multiple choice and organisation and (where applicable) to indicate the
open questions. It explored the following areas: numbers directly involved from the client's organisation
and any supplier's organisation(s). Their responses,
. information about the respondent and the project classied into four groups, are shown in Fig. 5. The
upon which the replies were based; mode number of people directly involved in a project
. the criteria used for judging the project's outcome; (including client and supplier organisations, where
. any unexpected side-eects arising from the project; applicable) was 10.
. factors critical to the project's outcome;
. the methods, tools and techniques used; 3.1.5. Role of respondent
. the limitations or drawbacks experienced with the The surveys were addressed to project managers but,
methods, tools and techniques used. as Fig. 6(a) shows, only 47% (113) of respondents
identied themselves by the title `Project Manager'.
However, a higher proportion, 63% (146), indicated
3. Analysis of data and discussion of results that they had managed the project being considered (see
Fig. 6b). The other titles by which people identied
3.1. The respondents and their projects themselves and the parts they played in the projects can
also be seen in Fig. 6.
Of the 995 questionnaires that were sent out in the
main survey, 236 were returned (23.72% response rate). 3.1.6. Main decision maker
All of the questions were answered in nearly every case Respondents were requested to indicate who was
but two respondents failed to describe the part they ultimately responsible for decisions concerning the pro-
played in managing their projects, one respondent failed ject. 43% (102) of respondents stated that they were the
to answer the question on limitations or drawbacks main decision makers. 33% (77) were involved in the
experienced with the methods, methodologies, tools or decision making process, but 24% (57) of the respon-
techniques and four failed to respond to the question on dents did not have any input into the main decision
unexpected side-eects. making processes.

3.1.1. Project type 3.1.7. Project duration


On the basis of the responses received the projects Project duration, broken down into seven categories,
described were classied into 16 types (Fig. 1). Just over is shown in Fig. 7. 37% (87) of the projects lasted
25% (60) of respondents were involved with informa- between 6 and 12 months and the mode duration of a
tion technology projects, whilst reorganisation in either project was 6 months.
the public or the private sector accounted for nearly
19% (44) of the projects. 37.2% (88) of the projects 3.1.8. Project completion
were carried out for a client. 62.7% (148) of projects 91% (215) of the projects were completed. 9% (21) of
were carried out within the respondent's organisation. the projects did not run through to completion.

34
D. White, J. Fortune / International Journal of Project Management 20 (2002) 111 3

Fig. 1. Project type frequency of mention. N=236.

Fig. 2. Industry sector breakdown respondent's organisation. N=236.

3.2. Criteria used for judging project outcome complete success (Fig. 8). This success rate is far higher
than that reported in the literature.
3.2.1. Project outcome
Project outcome was measured using a bi-polar 3.2.2. Criteria used for judging success
semantic dierential continuous line scale. For analysis Respondents were asked to indicate and rank (from 1
the line was divided into seven equal sections and coded to 5) the criteria they used for judging success. The
from 1 (abandoned) to 7 (complete success). In the opi- ranks were re-coded (1=5, 2=4, 3=3, 4=2, 5=1 and
nion of the respondents, 41% (97) of the projects were a criteria not chosen=0). The sum of the re-coded

35
4 D. White, J. Fortune / International Journal of Project Management 20 (2002) 111

Fig. 3. Industry sector breakdown client's organisation. N=88.

Fig. 4. Number of people employed.

Fig. 5. Numbers directly involved in project frequency of mention.

responses is displayed in Table 1 where it can be seen These criteria were important to the respondents to the
that `Meets client's requirements' was the criteria most survey but in addition to these standard measures a
often ranked rst by respondents, followed by `Com- signicant number of respondents identied another
pleted within schedule' and `Completed within budget'. citerion as important. This was the t between the project
Much of the literature suggests that criteria against and the organisation and the consequences of the pro-
which the success of projects is judged are time taken, ject for the performance of the business. For instance,
cost and the extent to which requirements are met. they talked about the need to meet organisational

36
D. White, J. Fortune / International Journal of Project Management 20 (2002) 111 5

Fig. 6. (a) Role of respondents. N=236; (b) respondent's part in project. N=234.

Fig. 7. Project duration frequency of mention. N=236.

Fig. 8. Project outcome. N=236.

objectives and to minimise business disruption and arisen (see Table 2). Their descriptions of the unex-
about the project's yield in terms of business and other pected side-eects, grouped into desirable side-eects
benets. and undesirable side-eects, are displayed in Table 3.
Examination of this data suggests that nearly 70% (113)
3.3. Unexpected side-eects of the side-eects could be attributed either directly or
indirectly to lack of awareness of the environment. This
Unexpected side-eects was another area explored by may imply that many of the tools and techniques the
the survey. 46% (108) of respondents reported that their respondents used were poor at modelling `real world'
project gave rise to unexpected side-eects or outputs, problems or that insucient account was taken of pro-
with many of them expressing surprise that these had ject boundaries and environments.

37
6 D. White, J. Fortune / International Journal of Project Management 20 (2002) 111

3.4. Factors critical to the project's outcome added by the respondents]. The count of factors indi-
cated by the respondents is shown in Table 4. `Clear
In the section of the questionnaire exploring this area, goals and objectives' was the most mentioned factor,
respondents were presented with a list of 19 critical fac- selected by 87% (206) of respondents.
tors culled from a study of the literature [417] identi- Respondents were asked to indicate and rank the
fying or examining factors believed to be critical to the three factors they believed to be most critical to their
outcome of a project. They were asked to indicate which project's outcome. Their responses are displayed in
of the factors they regarded as `critical'. Respondents Table 5. From this table, it is evident that the number of
were also encouraged to add any other factors they felt times a critical factor was mentioned by respondents
to be important [a further ve (including `other') were does not correspond to the importance given to the
factor in terms of ranking. For instance, `Realistic
Table 1 schedule' was the second most frequently mentioned
Criteria used for judging project success factor (mentioned by 78% (185) of respondents) how-
Criteria Sum of re-coded Sums ever only 7% (17) of respondents ranked it rst and
ranking ranked 11% (27) ranked it second.
As Table 5 shows, the three critical success factors
Meets client's requirements 970 1
Completed within schedule 850 2
mentioned most frequently by respondents to the survey
Completed within budget 766 3 were:
Meets organisational objectives 188 4
Yields business and other benets 86 5 . Clear goals/objectives
Causes minimal business disruption 71 6 . Support from senior management
Meets quality/safety standards 48 7
Other criteria 20 8
. Adequate funds/resources

These results are similar, in large part, to those found


in the literature [5,13,18]. However, there were some
Table 2
Unexpected side-eects
dierences of opinion between the respondents and the
prevailing wisdom in this area. For example, alongside
Side-eects Count `project mission' and `top management support', Pinto
Yes (desirable) 40 and Slevin [13] emphasise `the provision of adequate
Yes (undesirable) 30 communication channels and control mechanism' but
Yes (both desirable and undesirable) 38
No 108
Don't know 16 Table 4
Did not respond 4 Factors critical the project's outcome frequency of mention
Total 236 Factors Count

Clear goals/objectives 206


Realistic schedule 185
Table 3 Support from senior management 176
Side-eects descriptions frequency of mention Adequate funds/resources 164
End user commitment 159
Desirable side-eects Count Clear communication channels 144
Eective leadership/conict resolution 138
Increased business/sales/opportunities 34
Eective monitoring and feedback 135
New understanding/knowledge gained 25
Flexible approach to change 133
Improved business/sta relations 18
Taking account of past experience 121
Greater consistency of working 6
Recognising complexity 121
Total 83 Taking account of external inuences 120
Eective team building/motivation 117
Undesirable side-eects Count Eective management of risk 117
Training provision 98
Undesirable organisational impact/conict 15
Contextual awareness 94
Problems with sta/client/contractors/suppliers 13
Provision of planning and control systems 88
Technical limitations came to light 12
Appreciating the eect of human error 53
Lack of awareness of environment 12
Considering multiple views of project 47
Underestimation of cost/time 10
Having access to innovative/talented people 8
Changes to goal/objectives 6
Other factor(s) 7
Poor IT awareness/knowledge 6
Having relevant past experience 3
Conicting priorities 5
Support from stakeholder(s)/champion(s) 3
Total 79 Having a clear project boundary 2

38
D. White, J. Fortune / International Journal of Project Management 20 (2002) 111 7

Table 5
The three factors believed to be most critical to the project's outcome

Critical factors 1st most critical 2nd most critical 3rd most critical Sum of
factor count factor count factor count counts

Clear goals/objectives 76 40 18 134


Support from senior management 28 25 24 77
Adequate funds/resources 14 35 23 72
Realistic schedule 17 27 22 66
End user commitment 23 18 23 64
Eective leadership/conict resolution 9 8 21 38
Flexible approach to change 7 15 12 34
Clear communication channels 4 13 16 33
Taking account of past experience 15 5 7 27
Eective management of risk 6 10 9 25
Contextual awareness 5 8 11 24
Eective monitoring and feedback 3 8 12 23
Recognising complexity 8 3 8 19
Provision of planning and control systems 3 9 7 19
Taking account of external inuences 8 3 6 17
Eective team building/motivation 3 4 8 15
Training provision 2 3 3 8
Considering multiple views of project 2 0 2 4
Having access to innovative/talented people 2 0 2 4
Appreciating the eect of human error 0 1 1 2
Support from stakeholder(s)/champion(s) 1 1 0 2
Having a clear project boundary 0 0 1 1
Total 236 236 236 708

only 37% (88) respondents regarded the `Provision of the standard text books on project management [2438].
planning and control systems' to be a critical factor. Again, respondents were encouraged to add to the list
Cash and Fox [6] maintain that `successful projects of options if necessary.
always have a champion', but `Support from stake- The 44 options provided were grouped as follows:
holder(s)/champion(s)' was only considered critical by
1% (3) of respondents. Furthermore, although it is 1. Methods/methodologies
widely argued [3,11,16,1923] that many projects fail 2. Project management tools
due to inadequate management of risk, only 49.5% 3. Decision making techniques
(117) of respondents considered `Eective management 4. Risk assessment tools
of risk' to be critical. Therefore, although there was 5. Computer models/databases/indexes
agreement between the literature and the survey ndings 6. Computer simulations
over the top three factors there was disagreement about
the extent to which other factors were important deter- Table 6 shows the mean, mode, range and count of
minants of success. the frequency of mention for the methods, methodologies,
Interestingly, `Having a clear project boundary' was tools and techniques used. Between them respondents
regarded as the least important factor in relation to the indicated use of a total of 1210 methods, methodologies,
outcome of a project. This may add weight to the argu- tools and techniques. The maximum number of methods,
ment put forward earlier when discussing unexpected methodologies, tools or techniques used by any single
side-eects, that insucient account was taken of pro- respondent was 23, the mode was 3 and the mean was 5.
ject boundaries and environments. 2% (5) of respondents stated that they did not use any
methods, methodologies, tools or techniques whatsoever.
3.5. Methods, methodologies, tools and techniques Although 28% (66) of respondents did not use any
method or methodology over 95% (225) of respondents
The section on methods, methodologies, tools and used at least one project management tool. 52% (123) of
techniques presented respondents with a list of 44 respondents did not use any decision making techniques.
options and were asked to indicate which had been used 54% (128) of respondents used their own `in house'
in the project being considered. The options chosen for project management method. The most commonly used
inclusion in the list were those found in a selection of project management tool (77%, 182) was `o the shelf'

39
8 D. White, J. Fortune / International Journal of Project Management 20 (2002) 111

Table 6
Project management methods, methodologies, tools and techniques frequency of use

Project management method/methodology/tool/technique Count of frequency of use Total used Mean Mode Range

Project management methods/methodologies 206 0.87 1 3


Projects in controlled environments (PRINCE) 23
Projects in controlled environments 2 (PRINCE2) 14
Structured systems analysis and design methodology (SSADM) 17
The European risk management methodology (RISKMAN) 1
The RIBA plan of work 2
Other project management methods/methodologiesa 16
In house project management methods 128
In house similar to PRINCE 5

Project management tools 617 2.61 1 7


Critical path method (CPM) 70
Work breakdown structure (WBS) 75
Cash ow analysis (CFA) 43
Gantt bar charts 152
Graphical evaluation and review technique (GERT) 4
Programme evaluation and review technique (PERT) 24
Strengths weaknesses, opportunities and threats (SWOT) 41
Other project management toolsb 21
Project management software 182
In house project management tools 5

Decision making techniques 172 0.73 0 4


Cost benet analysis (CBA) 88
Decision analysis (DA) 9
Sensitivity analysis (SA) 19
Expressed preferences 23
Implied preferences 11
Revealed preferences 11
Other decision making techniques 9
In house decision making techniques 2

Risk assessment tools 147 0.62 0 10


Life-cycle cost analysis (LCCA) 25
Event tree analysis (ETA) 8
Fault tree analysis (FTA) 6
Probability analysis (PA) 34
Reliability analysis 13
Uncertainty analysis 3
Failure mode and eect analysis (FMEA) 10
Hazard analysis (HAZAN) 9
Hazard and operability studies (HAZOP) 9
Operation and maintenance risk analysis (OMRA) 4
Preliminary hazard analysis (PHA) 5
Other risk assessment tools 7
In house risk assessment tools 14

Computer models/databases/indexes 40 0.17 0 3


CRUNCH 1
Lessons learnt les (LLF) 23
Expert systems 4
In house computer models/databases/indexes 12

Computer simulations 11 0.05 0 2


Hertz 1
Monte Carlo 10

Other techniques 11 0.05 0 2


Other techniques 17

All methods, tools and techniques 1210 5.13 3 23


a
Includes other methods used in Information Systems Development Projects.
b
Includes tools used in Information Systems Development Projects.

40
D. White, J. Fortune / International Journal of Project Management 20 (2002) 111 9

software. 64% (152) of respondents used Gantt charts, A cross-tabulation of the methods, methodologies,
37% (88) of respondents used Cost Benet Analysis. tools and techniques with limitations and the descrip-
Looking at these ndings in relation to those reported tions of the limitations is given in Table 10. It is worth
elsewhere it is worth noting the similarity between this noting that of the 56 respondents who experienced pro-
study and that of Baldry [39]. In his report of a survey blems with `Project Management Software', 26 found
of practitioners' use of risk analysis and management the software `Inadequate for complex projects'. It is also
techniques, he commented on the large width of the worth noting that `In House Project Management
range of techniques used. As can be seen from Table 6, Methods', which was the fourth most frequently cited
risk analysis was the area where the number of dierent (expressed as a percentage of frequency of use) as pos-
tools used was highest. However, it is also worth noting sessing limitations, was criticised for a wide variety of
Humphreys' submission [40] that only a small number reasons.
of project managers use risk assessment tools because The most frequently described limitation was `Inade-
only a few of the tools provide support for the man- quate for complex projects' (32 cases). 26 of these cases
agement of risk. This may provide an explanation as to referred to `Project Management Software'. One factor
why 65% (154) of respondents to the survey did not use that may lie behind this nding has been identied by
any risk assessment tools. Cotterell [41]. He found that only a few software
packages included the facility to track shared resources.
3.6. Limitations and drawbacks of the methods, This must constrain the accurate modelling of complex
methodologies, tools and techniques used projects.

42% (99) of respondents said they had encountered


limitations or drawbacks with the methods, methodolo-
gies, tools or techniques they had used. Those they cited
(23 respondents cited two methods/tools/techniques Table 8
each) are listed in Table 7. In total, limitations were Methods and tools with the highest frequency of reported limitations
reported in connection with 10% (122) of the 1210 Methods/tools with limitations Frequency Frequency Frequency of
methods, methodologies, tools and techniques used. of use of reported limitations
`Project Management Software' was reported the high- limitations as a % of
est number of times. Table 8 shows the methods and frequency
of use
tools with the highest frequency of reported limitations
as a percentage of their frequency of use. Project management software 182 56 31
The respondents' descriptions of the limitations or Structured systems analysis and 17 5 29
drawbacks experienced have been grouped into twelve design methodology
PRINCE and PRINCE 2 37 9 24
categories and are presented in Table 9. In house project management 128 18 14
methods

Table 7
Methods, methodologies, tools or techniques with limitations
frequency of mention Table 9
Description of limitations or drawbacks frequency of mention
Method/methodology/tool/technique with limitations Count
Description of limitation/drawback to method/ Count
Project management software 56 methodology/tool/technique
In house project management method 18
Project management tool 10 Inadequate for complex projects 32
PRINCE 7 Dicult to model `real world' 18
Structured systems analysis and design methodology 5 Too heavy in documentation, too time consuming 12
Method/tool used in software development projects 5 Other 11
Other techniques 5 Failed to predict problems 9
Project management methods 4 Constrained activities, did not allow a holistic view 9
Risk assessment tool 4 Too unwieldy, not cost eective 9
PRINCE 2 2 Lack of training/ expertise, etc. 6
Decision making technique 2 Not suitable no suitable tools available 5
In house risk assessment tool 2 Too much emphasis on following the `standard' 5
In house project management tool 1 Not fully developed/ immature 4
In house computer model/database/index 1 Lessons learnt in past not carried forward 2
Total 122 Total 122

41
10 D. White, J. Fortune / International Journal of Project Management 20 (2002) 111

Con-strained Too unwieldy, Other Total


4. Conclusion

122
10
18
56
7
2
5
4

1
2
4
2
1
5

5
This paper reports on the results of a recently con-

11
1
1

1
4

1
1
1

1
ducted survey that attempted to capture the `real world'
experiences of project managers with a view to:
did not allow eective
not cost

. identifying any common criteria used for dening


1

1
6

9
project success;
a holistic view

. exploring the extent to which projects give rise to


activities,

unexpected side-eects;
. establishing a common list of `critical success fac-
tors';
1

4
3

9
the standard
emphasis on

. identifying the methods, methodologies, tools and


Too much

following

techniques in current use in the eld of project


management;
1

1
3

5
. identifying any limitations or drawbacks to the
developed/ documentation,

methods, methodologies, tools and techniques.


Not fully Too heavy in

consuming
immature too time

The three criteria used for judging project success


12

most cited in the literature (on time, to budget, to spe-


3

3
5

cication) were also the highest ranked success criteria


identied in the survey. However, they were not the sole
Cross tabulation methods, methodologies, tools and techniques with limitations and a description of the limitations

criteria by which project outcome was judged; the t


1

between the project and the organisation and the con-


Not suitable

sequences of the project for the performance of the


no suitable

business were also reported as important criteria.


available

A remarkably high proportion (41%) of projects were


tools

judged to be a complete success but 46% were described


1
1

1
1
1

5
Lessons learnt

as giving rise to unexpected side-eects but it should be


in past not

noted that 14% (34) of the projects that gave rise to


forward
carried

unexpected side-eects were among those considered to


Description of limitation to method/tool/technique

be a complete success.
1

2
1

Three of the four top ranking factors respondents


expertise,
for complex training/
Dicult to Failed to Inadequate Lack of

identied as critical to project success mirrored the suc-


etc.

cess criteria (realistic schedule, adequate funds/resour-


2

1
2

ces, clear goals/objectives). `Support from senior


problems projects

management' was the other most frequency mentioned


critical factor.
1

3
26

32

Between them, respondents used 1210 methods tools


and techniques. The most widely used were `Project
model `real predict

Management Software' and `Gantt Charts'. The mode


2
2
1

9
3

number of tools used was 3. 66 respondents did not use


any method or methodology, 123 respondents did not
world'

use any decision making techniques and 154 respon-


11

18
2
1

dents did not use any risk assessment tools. 128


Method/tool used in software development

respondents used their own `in house' project manage-


In house computer model/database/index
Method/tool/technique with limitation

In house project management method

ment method. 99 respondent said they had encountered


In house project management tool

limitations or drawback with the methods, tools or


techniques they had used. `Project Management Soft-
Project management software

In house risk assessment tool


Project management method

Decision making technique

ware' was reported as the tool with the most limitations


Project management tool

and was identied as being particularly unsuitable for


Risk assessment tool

use with complex projects.


Other techniques

Following this survey, work is in hand to carry out a


PRINCE 2

real-time study of two projects with a view gaining a


Table 10

PRINCE

SSADM

projects

greater understanding of the eects of the various


Total

interacting processes and decisions that take place

42
D. White, J. Fortune / International Journal of Project Management 20 (2002) 111 11

throughout the life of a project, thus building up a more [20] Charette RN. Large-scale project management is risk manage-
holistic view of project management. ment. IEEE Software 1996;13(4):1107.
[21] Hottenstein MP, Dean JW. Managing risk in advanced manu-
facturing technology. California Management Review
References 1992;34(4):11226.
[22] Stump EJ. Pre-emptive project risk management: part 1. Project
[1] Smart G. Don't be a systems victim. Internal Auditing, Manager Today, September 1996, 267.
1995;December:1820. [23] Willcocks L, Margetts H. Risk assessment and information sys-
[2] Copperdale J. Manage risk in product and process development tems. European Journal of Information Systems 1994;3(2):127
and avoid unpleasant surprises. Engineering Management Jour- 38.
nal, 1995;February:358. [24] Ansell J. Reliability: industrial risk assessment. In Ansell J,
[3] Willcocks L, Griths C. Predicting risk of failure in large-scale Wharton F, editors. Risk: analysis, assessment and management.
information technology projects. Technological Forecasting and UK: John Wiley & Sons, 1992.
Social Change 1994;47:20528. [25] Buss DM, Craik KH, Dake KM. Contemporary worldviews and
[4] Baker BN, Murphy D, Fisher D. Factors aecting project suc- perception of the technological system. In: Covello VT, Menkes
cess. In Cleland DI, King WI, editors. Project management J, Mumpower J, editors. Risk evaluation and management. USA:
handbook. Van Nostrand Reinhold Company, USA, 1983. Plenum Press, 1986.
[5] Belassi W, Tukel OI. A new framework for determining critical [26] Cassidy K. Assessment, regulation and the public. In: Risks to
success/failure factors in projects. International Journal of Pro- the public: the rules, the rulers and the ruled. Paper presented at
ject Management 1996;14(3):14151. Symposium, 14 December 1994, The Hazards Forum, UK.
[6] Cash C, Fox R. Elements of successful project management. [27] CCTA. Introduction to the management of risk. UK: CCTA
Journal of Systems Management 1992;43(9):1012. Library/HMSO, 1993.
[7] Couillard J. The role of project risk in determining project man- [28] Cleland DI, King WR. Systems analysis and project manage-
agement approach. Project Management Journal 1995;26(4):315. ment. 3rd ed. USA: McGraw-Hill Book Company, 1983.
[8] Gowan JA, Mathieu RG. Critical factors in information system [29] Clifton JJ. Hazard prediction. In: Keller AZ, Wilson HC, editors.
development for a exible manufacturing system. Computers in Disaster prevention, planning and limitation. Proceedings of the
Industry 1996;28(3):17383. First Conference University of Bradford, 1213 September 1989,
[9] Jang Y, Lee J. Factors inuencing the success of management The British Library, 1990.
consulting projects. International Journal of Project Management [30] Kerzner H. Project management: a systems approach to planning
1998;16(2):6772. scheduling and controlling. 6th ed. USA: Van Nostrand Rein-
[10] Morris WG. Research at Oxford into the preconditions of success hold, 1998.
and failure of major projects. In: Proceeding of 18th Annual [31] Klets TA. Critical aspects of safety and loss prevention. UK:
Seminar/Symposium of Project Management Inst, Montreal, Butterworth & Co, 1990.
Canada, 2025 September 1986. p. 5366. [32] Linnerooth-Bayer J, Wahlstrom B. Applications of probabilistic
[11] Morris PWG, Hough GH. The anatomy of major projects. UK: risk assessments: the selection of appropriate tools. Risk Analysis
John Wiley & Sons, 1987. 1991;11(2):23948.
[12] Pinto JK, Kharbanda OP. Lessons for an accidental profession. [33] Lucas D. Understanding the human factor in disasters. Inter-
IEEE Engineering Management Review 1995;23(4):1827. disciplinary Science Reviews 1992;17(2):18590.
[13] Pinto JK, Slevin DP. Critical factors in successful project imple- [34] Merkhofer MW. Comparative analysis of formal decision-mak-
mentation. IEEE Transactions on Engineering Management ing approaches. In: Covello VT, Menkes J, Mumpower J, editors.
1987;Em-34(1):227. Risk evaluation and management. Plenum Press, USA, 1986.
[14] Tan RR. Success criteria and success factors for external tech- [35] Partington D. The project management of organisational change.
nology transfer projects. Project Management Journal International Journal of Project Management 1996;14(1):1321.
1996;27(2):4556. [36] Reason J. Human error. UK: Cambridge University Press, 1990.
[15] Wateridge J. IT projects: a basis for success. International Jour- [37] Robins N. Changing perceptions of project management. Paper
nal of Project Management 1995;13(3):16972. presented to conference on Changing perception of risk: the
[16] Wideman RM. Criteria for a project-management body of implications for management, Bolton Business School, 27 Feb-
knowledge. International Journal of Project Management ruary1 March 1994. p. 118.
1995;13(2):715. [38] Weaver PL. Practical SSDM version 4. UK: Pitman Publishing,
[17] Yeo KT. Planning and learning in major infrastructure develop- 1994.
ment: systems perspectives. International Journal of Project [39] Baldry D. The evaluation of risk management in public sector
Management 1995;13(5):28793. capital projects. International Journal of Project Management
[18] Magal SR, Carr HH, Watson HJ. Critical success factors for 1998;16(1):3541.
information centre managers. Management Information Systems [40] Humphreys P. Risk analysis tools and techniques in project
Quarterly 1988;2:41326. management. Methods of Operations Research 1990;63:36987.
[19] Chapman CB, Ward S. Project risk management: processes, [41] Cotterell S. Annual software review 1998. Project Manager
techniques and insights. UK: John Wiley & Sons, 1997. Today 1998; August: 223.

43
Critical Success Factors in Enterprise Wide Information
Management Systems Projects
Mary Sumner
Southern Illinois University Edwardsville
Campus Box 1106
Edwardsville, IL 62026
618-650-2093
Msumner@ siue.edu

ABSTRACT engineering business processesbefore implementing IT


solutions (Hammer and Champy, 1993), and assuring close
In the past several years many organizations have initiated involvement between users and the IT organization
enterprise-wide information managementsystems projects, (Mumford, 1981).
using such packagesas SAP, Peoplesoft,and Oracle. These
projects often representthe single largest investment in an With new technology, it is often critical to acquire external
information systems project in the history of these expertise, including vendor support, to facilitate successful
companies,and in many casesthe largest single investment implementation. Also, the costs of training and support are
in any corporate-wideproject. often under-estimated,and these costs may be many times
greater than originally anticipated. Client-server
implementationsoften bring surprises with respectto cost,
These enterprise-wide information management systems because of the costs of decentralized servers, systems
projects bring about a host of new questions. Someof these integration software,technical support, and software updates
questionsand issuesare: and version control (Caldwell, 1996).
. How are theseprojectsjustified? The lessonslearned from systemsdevelopment projects can
. Do firms need to change their fundamental business also pose somechallengesfor large-scaleintegratedprojects.
processes, organizational structures, and business Someof the best practices in project managementinclude
strategiesto fit the package? effective external integration strategies,such as creation of a
. What are the technical challenges associated with user steering committee, user participation on the project
implementing a large-scalepackagesuch as SAP? team, and user responsibility for education and installation
. What is the best way to implement a large-scale (Cash, McFarlan, 1992). In addition, successful projects
package: To work through the organization unit-by- require the use of effective internal integration strategies,
unit? Or to use a cross-functionalapproach? such as use of outside technical expertise, selection of an
. What are the critical success factors in the experiencedproject manager,and selection of team members
implementation of a large-scalepackage? with significant previous work relationships.
. What are the implications of theseprojects for the skill
and knowledge requirementsof the IT workforce? Other systems development practices which contribute to
project success are effective project planning, effective
This paper will provide case studies of seven organizations changecontrol, businessjustification, compatibility of skills
implementing enterprise-wide information management with the skill set needed for project requirements, and
systemsprojects and will provide insight into each of these leadership by a champion who markets the project
questionsbasedupon their experiences. internally (Beath, 1991).

1. BACKGROUND: 2. CAUSES OF PROJECT FAILURES:


Some of the success factors associatedwith large-scale Much has been written about the causes of information
client-server implementation projects include securing the systemsproject failures. Poor technical methodsis only one
supportof top management(Beath, 1991),improving or re- of the causes, and this cause is relatively minor in
Permission to make digital or hard copies ofall or part ofthis work for comparison to larger issues, such as failures in
personal or classroom we is granted without fee probided that copies communicationsand ineffective leadership.
are not made or distrihutcd for protit or commercial advantage and that
copies bear this notice and the full citation on the first page. To Copy In Robert Blocks analysis, there are twelve categorieswhich
otherwise. to republish. to post on sewers or to redistribute to lists. classify most systemfailures. These are summarizedin the
requires prior specific permission and/or a fee. following table:
SIGCPR 99 New OrleansLA USA
Copyright ACM 1999 l-58113-063-5/99/04...$5.00

297
44
m with many changesin

structuredanalysis/design. maintenancecosts,sc

scheduledelays, and inconsistent

maintain.
10. Methodology failures Failure to perform the activities This type of failure can lead to any of
needed,while unnecessaryactivities the consequencesof systemfailure.
areperformed.
11. Planning and control failures Causedby vague assignments, Work assignmentsmay overlap,
inadequateproject managementand deliverablesmay be poorly defined,
tracking tools. and poor communication may result.
12. Personality failures Theseare causedby people clashes. Passivecooperationand covert
resistance,with possible actsof
vengeance.

In summary,Block points out that successfulprojectsare on- . How was the investment in the integrated system
time and within budget, reliable, maintainable, and meet the justified? What were the tangible and intangible
goals and requirements of users. Block points out that businessbenefits that were considered?
managerswho succeeddo an initial evaluation of a project. . What was the role and importance of top management
They evaluate the rules, the players, the goals, the support?
constraints, and the project managers responsibility and . How were businessprocessesaffectedby the software?
authority, and well as the feasibility of success (Block, . What investmentsin training, support, and maintenance
1983). Managersshould recognize and implement strategies were neededto assureproject success?
to minimize the risk of failure, asoutlined in the abovetable. . Was external vendor expertise used to accomplish
certain aspectsof the project?
3. RESEARCH QUESTIONS: . What was the role of end-usersin project management
and systemsdevelopment?
. What is the purpose and scope of the enterprise-wide
information managementsystemsproject?
. What are the project objectivesand outcomesto date?

298
45
4. FINDINGS: . Re-design of business processes. Rather than
attempting to modify the software, Monsanto re-
The findings describe seven case studies which have been engineered their business processes to be consistent
accomplished as a pilot study for this research. These case with the software. This has proved to be critical to the
studies will highlight the issues of project justification, projects success. Many companies go to war with
benefits, critical success factors, and factors associated with the package and try to make it meet their process
project failure. They deal with three SAP Projects, two requirements, only to lead the way to huge cost overruns
Peoplesoft Projects, and two Oracle Projects. and project failure in some cases.
. Training and re-skilling. Monsanto invested heavily in
training and re-skilling their developers in SAP software
5. ENTERPRISE-WIDE PROJECTS design and methodology.
USING SAP: . External consultants. When they didnt have needed
Monsanto: expertise internally, Monsanto brought in the
consultants they needed.
Monsanto is one of the worlds largest chemical and life . Management support. Without question, top
sciences companies. Since becoming chairman and chief management support is critical to the success of a
executive in 1995, Robert Shapiro has intensified the project, and the SAP project had approval by top
transformation of Monsanto from a pioneering chemical management.
company to a cutting-edge bioengineering company with a . Role of the champion. The project leader for the SAP
concentration on food and nutrition. The focus of project was clearly a champion for the project, and
Monsantos business is a $2 billion drug division, a $1.2 that role was critical to marketing the project throughout
billion food ingredients division, and a $3 billion maker of the organization.
agricultural products. This $6 billion group of companies is . Discipline and standardization. Another success
positioned to lead the biotech revolution of the future. factor which is closely associated with the software
itself is the need to adhere with the standardized
Project Justification: specifications that the software supports.
. Effective communications. Another important CSF is to
Beginning in 1996, Monsanto started a corporate-wide SAP tell everyone in advance what is happening, including
project. The business justification for the project was the scope, objectives, and activities of the project.
operational excellence, e.g. cutting the costs of core Admit that there will be change.
transactions-processing systems, such as order processing . Obtain business analysts. One of the critical
and inventory management. In addition, an integrated workforce requirements for the project was the ability to
package could support worldwide business operations and obtain analysts with both business and technology
replace division-level systems. Before SAP, Monsanto had knowledge. Instead of 200 programmers with
four purchasing packages--one for each business unit. SAP average skills, the SAP project demanded and could be
provided economies of scale in development, maintenance accomplished with 20 of the best and brightest
and operations. Its overall costs were divided by a much analysts. However, retaining these professionals was a
larger number of users. For example, buying a $100,000 significant problem because of their market value.
package to support 5000 users is less expensive than buying
a $25,000 package to support 100 users. In addition, the In terms of factors conducive to project failure, one of the
SAP project enabled Monsanto to reduce its information main factors associated with failure is lack of integration.
systems development staff from 500 to 50 people. The project needs to be based on an enterprise-wide design.
Some of the business drivers for the SAP implementation You cant start with pieces, and then try to integrate the
at Monsanto included: data integration, standardization, software components later on. Other problems are caused
access to timely and complete information, leverage gained by going to war with the package, and trying to modify the
in purchasing, and globalization. SAP cut the costs of code to conform with existing business processes. As
operational systems, improved the reliability of customer mentioned earlier, this is a sure formula for potentially huge
service, and assured timely delivery and follow-up. cost overruns and possible project failure.

Critical SuccessFactors: In terms of lessons learned, Monsantos experience


demonstrated the importance of using SAPs built-in best
Some of the critical success factors in the Monsanto practices, its systems development methodology, and a
project dealt with the management structure, the re-design of combination of systems analysts with both business sense
business processes, investments in re-skilling and and knowledge of information technology. Other lessons
professional development, and acquisition of external are: start with small, empowered teams; standardize data
expertise. early on; eliminate complexity; and avoid compromising the
system and its specifications. Standardization is key to
. Management structure: Monsanto put someone in success, and can create greater flexibility and changeability
charge and centralized the management structure of the down the line.
project in order to avoid duplication of effort.
Monsanto considered their project a success because of a
centralized management structure, adherence to design

299
46
specifications, and investments in training and support customers in a global marketplace. This meant that any
required to overcometechnical and procedural challengesin customer in the world could place orders using one
design and implementation. integrated order processing system, as opposed to using
many different systemsfor different product lines.
Anheuser Busch Companies:
Project Management:
Anheuser Busch Companiesinitiated a corporate-wide SAP
project in 1996. The project is perhapsthe largestIT project The project team at Sigma consistedof 30 individuals, with
ever undertaken by the company. Anheuser Busch is a 10 consultants and 20 internal people. Consultants were
worldwide manufacturer of beer and related food products, rotated in and out of the project, with some being weeded
headquarteredin St. Louis, Missouri. out. The project team consisted of both business and
system types.
Project Justification:
Technical Issues:
The original project justification was similar to the one at
Monsanto. There were extensive economies of scale One of the major technical issues faced at Sigma was the
associatedwith consolidating four MIS projectsinto one, and capacity of servers on a client-server network. Becauseof
SAP offered an integrated, corporate-wide solution. The capacity issues, SAP requires mainframe-like operating
business justification entailed major cost savings from system performance in a PC-network operating system
reducing the costs of operational level information systems. environment.
SAP provided hard-dollar savings,basedupon integration of
data and processes, a common database, and increased Critical SuccessFactors:
leveragein purchasingand buying.
Some of the major critical successfactors for the SAP
Critical SuccessFactors: Project included:

The critical successfactors mentioned by Anheuser Buschs Achieve the support of senior management for
senior project managerfor the SAP project included: accomplishingproject goals and objectives and aligning
thesewith strategicbusinessgoals.
Maintain excellent staffing, both by developing internal Re-design business processes to be consistent with
personneland by using external consultants. systemspecifications.
Place a business leader in charge, so that project Invest in training, re-skilling, and professional
leadershipcomesfrom the businessperspective. developmentof the IT workforce.
Avoid trying to modify the software; instead,re-design Use a mix of consultants and internal staff to work on
or re-engineerexisting businessprocessesto fit what the project team, so that internal staff members can
the softwarewill support. grow the necessarytechnical skills for SAP design
Obtain and maintain top managementsupport for the and implementation.
project. Make a major commitment to training end-users in
Obtain and retain team memberswith knowledge of the systemuses.
business processes,as well as an understandingof the
technical aspects. 6. ENTERPRISE-WIDE PROJECTS
Understandthe role of the champion in marketingthe USING PEOPLESOFT:
projects successthroughout the organization.

Three of the lessons learned in the Anheuser Busch case Boeing Company:
were to obtain the right people for the project, empower
the team, and take the training seriously. Recruiting and Project Justification:
retaining qualifed SAP systems developers was a critical
issue,as was the cost and availability of training. The Peoplesoft Project at Boeing was justified in terms of
better information, cost-reduction, and data integration.
Sigma Chemical: Between 70 and 80 systems were replaced by a single,
integrated system. While the original intent was to
Project Justification: implement an integrated human resources/payroll system
using Peoplesoft, the first phase of the project involved
The major sources of justification for the SAP project at completing the human resources (HR) component and
Sigma Chemical were the need to integrate a number of creating an interface to the existing McDonnell Douglas
different order processing systems,the need to improve and Corporation (MDC) payroll system. After the completion of
integrate financial systems, and the ability to reduce the the MDC-Boeing merger,the plan was to integrate both HR
workforce through systems integration. The major and payroll, using the Peoplesoftsoftware. As you will learn
motivation behind the project was to gain a competitive later, this phased-in approachcreatedsignificant problems
advantage by providing seamless order processing to in systemimplementation.

300
47
Project Management: MDC have created delays in establishing consistency
and coordination in platforms, database management
Originally, the Boeing Peoplesoft team consisted of 80 to systems, and operating system environments for the
100 members which migrated to the project from legacy Peoplesoft application. For example, the choice of
systems development units within the divisions. Each whether to implement Peoplesoft using Unix/Oracle as
project component utilized Peoplesoft consultants and an operating system/database environment or
consultantsfrom contractors. MVSIDB2 is still pending. While Unix/Oracle is the
standard environment at MDC, MVS/DB2 is the
Critical Success Factors: systemstandardat Boeing.

The CSFs in Boeings Peoplesoft project were similar to User resistance to change: Once data input was
those already mentioned for the various SAP projects. First decentralizedto the shop floor at McDonnell Douglas as
and foremost was the importance of using a vanilla part of the PeoplesoftHR systemimplementation, there
implementation,e.g. not changing the original software. In was major resistanceby end-users. This reinforces the
the Boeing case, a number of pieces of the Peoplesoft critical importanceof training.
software were customized. In its implementation, for Managementstructure: Another, more complex issue is
example, the HR piece was 70% vanilla, 30% custom. The relatedto the problem of having too many chiefs. At
payroll piece was 60% vanilla, and 40% custom; and the MDC, three different vice-presidents(including the HR
benefits piece was 50% vanilla, 50% custom. One of the head,the IT head,and the Finance VP) all had the same
most difficult and time-consumingaspectsof the project was authority and conflicts arose in establishing common
the creation of a bridge betweenthe HR and legacy payroll requirements. In implementing a centralized system,
application, and this resulted in extensive time and cost centralized authority must call the shots.
delays.
Finally, in terms of the lessons learned from the Boeing
To summarizesomeof theseCritical SuccessFactors: project, there were recurring themes: standardization, re-
. Use a federal approach;define what is neededat the skilling, addressing client-server technological
entreprise-level, and then apply it to the businessunit implementation issues up front, and effective team
level. management.Someof theselessonsinclude:
. Obtain strong top managementsponsorship.
. Attempts at customization will bring about unnecessary
Use a commondata model and commondatadefinitions
time and cost delays and should be avoided.
to drive commonbusinessprocesses.
. Training costsmay be higher than expected.
Obtain a full-time commitment of customers to
project managementand project activities. It may be difficult to get managersto commit to project
. managementroles, becausethey may be uncertain about
Use a vanilla approach;avoid customization.
what responsibilities will still be open to them once they
. If modifications are necessary, establish an up-front are transferredback to their functional areas.
agreementbetween IT and user managerswith respect
It is difficult to recruit and retain good technical people
to what is to be modified.
. becausemarketratesfor thesepeople are much higher.
Avoid technological bottlenecks (e.g. Boeing did not
It is important to prepare for client-server
have a client workstation configuration ready).
implementationwell in advance.
. Re-skill the end-users in new technologies and
It is important to implement a total integrated package
applications.
at one time, rather than in pieces. The building of a
. Supplementgeneralized user training with training in bridge between a Peoplesoft module and a legacy
the use of specific application modules. application was problematicand proved this point.
. Make a commitmentto training and re-skilling technical Manageteamexpectationseffectively.
professionals. Mandatechange;there will be change.
. Train, train, train.
. Be sensitive to user resistance.
. Establish disciplined, flexible programmanagement.
Edward Jones Company:

Boeings project encountered a number of difficulties and Project Justification:


challenges which were not within the projects control.
These included organizational re-structuring, user resistance The major justification for the PeoplesoftProject at Edward
to change,and managementstructure. To briefly summarize Joneswas dataintegration, a commonsystemsapproach,and
the impact and implications of eachof theseissues: hard dollar savingsthrough integration.

. Organizational re-structuring: The Boeing/McDonnell Critical Success Factors:


Douglas merger complicated the project and
necessitated the creation of a bridge between the The critical successfactors for this project, as with many of
Peoplesoft HR software and the MDC legacy system, the others,were:
resulting in extensive time and cost delays. In addition, . Obtaining top managementsponsorship;
different technology environments at Boeing and . Putting a steeringcommitteein charge;

301
48
. Recruiting and retaining key Peoplesoftpeople. . Obtain consultants who are specialists in specific
application modules.
Lessons Learned: . Understand and appreciate the criticality of high-tech
worker turnover, recruitment, and retention issues.
The lessons learned included other strategiescommon to . Emphasize reporting, including custom report
many of the other projects, including: avoiding development, the use of report generators, and user
customization and getting the business areas to dedicate training in reporting applications.
peopleto the managementof the project. . Emphasizeeffective user training.

7. ENTERPRISE-WIDE PROJECTS Emerson Electric Company:


USING ORACLE:
Project Characteristics and Justification:
Ralston PurinaCompany:
The major purpose of the project was to implement Oracle
Project Characteristics: financial, distribution, and manufacturing systems. The
business justification included: inventory reduction,
headcountsavings, and reduced lead times through on-time
The Oracle Financials Project at Ralston Purina supporteda delivery.
modular approach, with the implementation of add-on
modules in manufacturing. Since Ralston is a divisionalized
company, the Oracle projects were accomplished on a Project Management:
division-by-division basis.
Each Oracle project was accomplishedwith an internal team
of 25, augmentedby 50 to 60 external consultants and an
Project Justification: additional 50 users. The extensive use of consultants was
necessary because of project completion dates, which
The Oracle project was justified in terms of data integration dictated bringing in high-demand skills, rather than
and cost-reduction through the re-engineering of business developing these skills in-house. This meant considerably
processes. higher costs than the costs that would have been incurred
through the re-skilling of internal IT professionals.
Project Management:
Technical issues:
The project included over 100 people on teamswith 8 to 9
team membersand 2-3 external consultants composingeach The major technical issues dealt with migrating toward a
team. The consultant/internal staff mix facilitated cross- server-based architecture. Avoiding client workstation
training and the ability to grow internal staff memberswith implementationshelped to facilitate completion of technical
the needed technical skills. It was absolutely critical to activities.
obtain consultants with technical expertise in the
implementation of specific modules, such as A/R and G/L. Critical Success Factors:
Understanding how to use consultants as subject-area
specialistsis critical to project success. Some of the CSFs in the project re-iterated strategies
expressed in the interviews with many of the other
Critical Success Factors: companies,and included;

Someof the CSFs for the Oracle project at Ralston included . Obtain top managementsupport for the project.
similar messages,such as the need for strong management . Obtain IT top managementsupport for the project.
sponsorship, experienced technical consultants, and . Re-designbusinessprocessesto support the software.
experienced project managers. Some new strategies . When software does not meet requirements, use bolt-
included: ons, or add-onpackageswhich are offered by Oracle or
third-party vendors.
. Address scope expansion requests with information . In some cases, when software does not meet user
on the time, cost, and business impacts of these requirements, create manual workarounds, including
changes; such things asthe use of bar-coding.
. Addresstough issuessquarely. Surpriseshurt trust. . Make a commitmentto user training.
. Bring all related projectstogetherand managethem. . Maximize use of consultants.
. Avoid customization (e.g. dont touch the software,
dont touch the Oracle tables, dont touch the Oracle Lessons Learned:
interfaces).
. Dont add people to a project, if it is already behind . Break the project into smaller phasesif possible. A
schedule. phased-inapproachis superior to the big-bang, all-at-
once approach.

302
49
. Limit the number of bolt-ons, or add-ons, to those projects requires centralized control, strict discipline, and
which are absolutely critical to accomplishing project extensive monitoring of project outcomes. Compared with
activities. traditional MIS projects, less emphasis is placed upon
defining user requirementsand organizing system processes
8. SUMMARY AND CONCLUSIONS: to support these needs. The risk is that using a large-scale
packagesuch as SAP to support the businesscreatesa more
These large integrated information systems projects pose centrally controlled, consistent organizational structure and
new opportunities and significant challenges. Based upon moves away from supporting unique business requirements
these preliminary case studies, it seems that the risk of at the businessunit level.
project failure is being containedprimarily through a number
of strategies. 9. REFERENCES:
Justify the enterprise-wide projects based upon cost- Beath, C. Supporting the information technology
justification and economiesof scale. champion, MIS Ouarterlv, V. 15, No. 3, 1991,pp. 355-373.
Re-engineer business processesto fit the package,
rather than trying to modify the software to fit the Block, Robert. The Politics of Proiects, Yourdon Press,
organizations current businessprocesses. Prentice-Hall, 1983.
Identify and implement strategiesto re-skill the existing
IT workforce and acquire external expertise through Caldwell, B. Client-Server: can it be saved? Information
vendorsand consultantswhen needed. u, V. 584,1996, pp. 36-44.
Utilize business analysts, with both business
knowledge and technology knowledge. Hammer, M. and Champy, J., Re-engineering the
Obtain top managementsupport for the project and a Cornoration: A Manifesto for BusinessRevolution, Nicholas
commitment to establishing and supporting project Brearley Publishing, London, 1993.
leadership.
Make a commitment to training ene-usersin custom Cash, J, McFarlan, F.W., McKenney, J, and Applegate, L,
A Portfolio Approach to IT Development, Cornorate
report development.
Information SvstemsManagement, Irwin Publishing, Third
Without question, the effective managementof these huge Edition, 1992.
Mumford, E., Participative systemsdesign: structure and
projectsis a new and unique challengewhich requiresthe use
method. Svstems,Obiectives, Solutions, V. 1, No. 1, 1981,
of project managementand control methods that have not
pp. 5-19.
been used extensively in the past. The sheer size of these

303
50
One Size does not Fit All Projects: Exploring Classical Contingency Domains
Author(s): Aaron J. Shenhar
Source: Management Science, Vol. 47, No. 3 (Mar., 2001), pp. 394-414
Published by: INFORMS
Stable URL: http://www.jstor.org/stable/2661507 .
Accessed: 26/04/2011 04:53

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .
http://www.jstor.org/action/showPublisher?publisherCode=informs. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

INFORMS is collaborating with JSTOR to digitize, preserve and extend access to Management Science.

http://www.jstor.org
51
One Size Does Not Fit All Projects:
Exploring Classical Contingency Domains

Aaron J. Shenhar
Wesley J. Howe School of TechnologyManagement, Stevens Institute of Technology,
Castle Point on the Hudson, Hoboken,New Jersey 07030
ashenhar@stevens-tech.edu

Not many authors have attempted to classify projects according to any specific scheme,
and those who have tried rarely offered extensive empirical evidence. From a theo-
retical perspective, a traditional distinction between radical and incremental innovation has
often been used in the literature of innovation, and has created the basis for many classi-
cal contingency studies. Similar concepts, however, did not become standard in the litera-
ture of projects, and it seems that theory development in project management is still in its
early years. As a result, most project management literature still assumes that all projects are
fundamentally similar and that "one size fits all." The purpose of this exploratory research
is to show how different types of projects are managed in different ways, and to explore
the domain of traditional contingency theory in the more modern world of projects. This
two-step research is using a combination of qualitative and quantitative methods and two
data sets to suggest a conceptual, two-dimensional construct model for the classification of
technical projects and for the investigation of project contingencies. Within this framework,
projects are classified into four levels of technological uncertainty, and into three levels of
system complexity, according to a hierarchy of systems and subsystems. The study provides
two types of implications. For project leadership it shows why and how management should
adapt a more project-specific style. For theory development, it offers a collection of insights
that seem relevant to the world of projects as temporary organizations, but are, at times,
different from classical structural contingency theory paradigms in enduring organizations.
While still exploratory in nature, this study attempts to suggest new inroads to the future
study of modern project domains.
(Project Management; Contingency Theory; Project Types; Project Classification; Technological
Uncertainty;System Complexity)

Introduction ically defined as a temporary organization that has


As an organized activity of mankind, projects could been established to complete a specific goal (Cleland
probably be found in all civilizations. However, as and King 1983).
a formal managerial discipline, project management The wide deployment of projects today illuminates,
is usually traced back to the precedence network in a rather paradoxical way, that as an organizational
diagramming techniques developed for the Polaris concept project management is quite new, probably
Submarine project in the 1950s and early 1960s not well understood, and clearly understudied. Most
(Fondahl 1987). Today, however, virtually all construc- research literature on the management of projects is
tion, product development, and engineering efforts are relatively young and still suffers from a scanty the-
using some formal project management structure, typ- oretical basis and lack of concepts. The goal of this

MANAGEMENT SCIENCE ? 2001 INFORMS 0025-1909/01/4703/0394$5.00


Vol. 47, No. 3, March 2001 pp. 394-414 1526-5501 electronic ISSN
52
SHENHAR
One Size Does Not Fit All Projects

exploratory research is to contribute to theory build- in contrast, was characterized as being informal,
ing of project management in two ways: first, to show decentralized, having just a few authority levels, hav-
how different types of projects are managed in differ- ing a breadth view (rather than a specialized one),
ent ways, and second, to explore the domain of tradi- and typically using extensive levels of communi-
tional contingency theory in the more modern world cation. According to the classical theorists, organic
of projects. This study was conducted in two steps organizations would better cope with uncertain and
while using a combination of qualitative and quan- complex environments while mechanistic organiza-
titative methods, and a two-dimensional model for tions predominate in simple, stable, and more certain
the classification of projects. The first step involved a environments. Mechanistic and organic organizations
qualitative study of 26 case projects, and was followed also differ in their capacity to deal with informa-
by the second, quantitative, part, which involved sta- tion, suggesting that organic organizations provide
tistical data on 127 projects. more capacity. Later scholars have similarly hypoth-
The paper is structured as follows. Following the esized that organizations that perform more innova-
theoretical background, the paper suggests a con- tive tasks would be different from organizations that
ceptual model for distinction among projects. The develop more routine products (e.g., Perrow 1967,
methodology section describes, respectively, the cases Thompson 1967, Mansfield 1968, Zaltman et al. 1973,
as well as the statistical data sets for the quali- Moch and Morse 1977, Blake 1978, Abernathy
tative and quantitative parts. The qualitative find- and Utterback 1978, Freeman 1982, Galbraith 1982,
ings then show how project management styles are Burgelman 1983, Ettlie et al. 1984, Drazin and van
typically clustered according to the project's techno- de Ven 1985, Dewar and Dutton 1986, Bart 1988,
logical uncertainty and system complexity. The quan- Pennings 1992).
titative second-part findings are then used to support While correlates of structural and environmental
the qualitative results and to test significant emerging attributes have been well studied when the orga-
trends. Finally, the discussion and implication sections nization is the unit of analysis, they have been
illustrate the theoretical differences between existing much less investigated in the project context. The
and temporary organizations (projects) and suggest project management literature has often ignored the
implications for management and further research. importance of project contingencies, assuming that
all projects share a universal set of managerial char-
acteristics (Pinto and Covin 1989, Shenhar 1993,
Theoretical Background and Yap and Souder 1994). Yet, projects can be seen
Basic Proposition as "temporary organizations within organizations,"
Classical contingency theory asserts that different and may exhibit variations in structure when com-
external conditions might require different organiza- pared to their mother organizations. Indeed, sev-
tional characteristics, and that the effectiveness of the eral authors have recently expressed disappointment
organization is contingent upon the amount of con- in the universal "one-size-fits-all" idea, and recom-
gruence or goodness of fit between structural and mended a more contingent approach to the study
environmental variables (Lawrence and Lorsch 1967, of projects (Yap and Souder 1994, Eisenhardt and
Drazin and van de Ven 1985, Pennings 1992). The the- Tabrizi 1995, Balachandra and Friar 1997, Brown
ory was introduced by Burns and Stalker (1961), who and Eisenhardt 1997, Souder and Song 1997, Song
were among the first to suggest the traditional dis- et al. 1997). As argued, by utilizing traditional con-
tinction between incremental and radical innovation, cepts in a new domain, new insights will most likely
and between organic and mechanistic organizations. emerge in this evolving and dynamic field (Brown
A mechanistic organization was described as for- and Eisenhardt 1997).
mal, centralized, specialized, and bureaucratic; having But how would classical contingency arguments
many authority levels; and maintaining only a mini- hold, and what are the dimensions of structure and
mal level of communication. An organic organization, variations in the dynamic, temporary, and changing

MANAGEMENTSCIENCE/Vol. 47, No. 3, March 2001 395


53
SHENHAR
One Size Does Not Fit All Projects

world of projects? As the coming discussion demon- More recent studies have looked at management of
strates, a careful review of the classical, as well as innovation and associated it with change. For exam-
the more recent, literature suggests the emergence of ple, Tushman and Anderson (1986) discussed the
two major dimensions-uncertainty and complexity. interplay between radical and incremental innova-
Thus, we have made them the focus of this research. tion and the cyclical model of technological change.
In selecting these dimensions, we are not suggest- Henderson and Clark (1990) linked different types
ing that these are the only variables that might be of technological change to product class and dif-
found to be different in various projects, but only ferent organizational consequences. Also, Burkhardt
that they seemed to be relevant dimensions which and Brass (1990) have conceptualized technological
our own observations and those of earlier researchers change as a source of uncertainty, and discussed the
have suggested might be important. We start by dis- relations between social structure and power and the
cussing the role of uncertainty and complexity in the diffusion and adoption of technological change.
classical, as well as modern, literature of organiza- Although traditional contingency studies in the
tions, and then observe their function in the project management of innovation have had only a lim-
and product management literature. ited impact on the literature of project manage-
Three influential works that were published inde- ment, some exceptions exist. Most have similarly
pendently in 1967 have had a significant impact focused on the impact of uncertainty and change on
on contingency theory. Lawrence and Lorsch (1967) the way organizations are conducting their project
operations. For example, Blake (1978) has sug-
focused on how different rates of change in technol-
gested a normative distinction between minor change
ogy, science, and markets impact the organization's
(alpha) projects, and major change (beta) projects,
ability to cope with these changes. Specifically, they
and Wheelwright and Clark (1992) have mapped
asked how such changes might influence the orga-
in-house product development projects according to
nization's orientation toward differentiation (division
the degree of change in product portfolio. Some
of labor and acquiring specific skills and practices)
have adapted the radical versus incremental dis-
and integration of the complex organization (collab-
tinction (e.g., Yap and Souder 1994, Eisenhardt and
oration and unity of effort). Using one integrated
Tabrizi 1995, Brown and Eisenhardt 1997, Souder and
score of uncertainty, they concluded that in a more
Song 1997, Song et al. 1997), while others suggested
diverse and dynamic field, effective organizations
more refined frameworks (e.g., Steele 1975, Ahituv
have to be highly differentiated and integrated, while and Neumann 1984, Cash et al. 1988, Pearson 1990).
in a more stable and less diverse environment, effec- While almost all of these studies used a distinction
tive organizations can be less differentiated, but must based on technical uncertainty, none of their typolo-
still achieve a high degree of integration. Thomp- gies has developed so far into a standard, empirically
son (1967) suggested that coping with uncertainty is based theoretical framework that is used to analyze
the central problem for complex organizations, and the full range of today's projects.
that technology and environments are major sources As for project complexity, the hierarchical nature of
of uncertainty. To deal with contingencies, he showed systems and their subsystems has long been at the
how rational organizations would use different strate- cornerstone of general systems theory (Boulding 1956,
gies for interaction and organizational design. Finally, Van Gigch 1978, Shenhar 1991). Boulding (1956), for
Perrow (1967) used an integrated viewpoint on tech- example, suggested a hierarchical classification of sys-
nology and complex organizations, while treating tems which includes nine levels, starting with the
technology as the independent variable and struc- lowest type of static structures and going up to tran-
ture as the dependent variable. Using technology scendental systems. This concept has often been men-
to distinguish between analyzable and unanalyzable tioned in the design literature to distinguish between
problems, he identified four types of industries- a product as a whole and a product in its parts
craft, routine, nonroutine, and engineering. (Marples 1961, Alexander 1964). Obviously, since

396 MANAGEMENTSCIENCE/Vol. 47, No. 3, March 2001


54
SHENHAR
One Size Does Not Fit All Projects

products are composed of components, and systems The Technological Uncertainty Dimension
of subsystems, hierarchies in products are almost The classification presented below is based on lev-
always addressed in practitioners' books and mono- els of technological uncertainty at the time of project
graphs that deal with engineering design problems initiation (Shenhar 1993, Shenhar and Dvir 1996).
(e.g. Pahl and Beitz 1984, Lewis and Samuel 1989, In general, we associated such uncertainty with
Rechtin 1991). However, as has been noted, a great the degree of using new (to the company) versus
deal of the existing knowledge of the design concept mature technology within the product or process
is still anecdotal and diverse; its theoretical basis is produced. Such association is based on previous
quite scant, and applicable design principles are only studies, equating "high-tech" with extensive use of
beginning to appear. Consequently, a clear taxonomy new technologies, and technological maturity with
of product levels and their design domains is difficult low uncertainty (Shanklin and Ryans 1984, Roussel
to construct (Hoover and Jones 1991); thus, no clear et al. 1991, Eisenhardt and Tabrizi 1995). Since most
classification of project hierarchies and their manage- projects employ a mixture of technologies, our clas-
ment styles has so far been suggested. sification is related to the share of new technology
Based on the rich foundations of structural within the product. The four project types are defined
contingency theory for existing organizations, the as follows (see Table la).
main proposition of this research is that, in Type A-Low Technological Uncertainty Projects
projects too, "one-size-does-not-fit-all" (Balachandra (Low-Tech). This type of project involves implemen-
and Friar 1997, Souder and Song 1997). Further, we tation of familiar technologies. Such projects rely
contend that modern projects exhibit a richer varia- only on mature technologies to which all industry
tion than can be captured by a simple dichotomy such players have equal access. All technologies are well
as the radical versus incremental distinction, or the known, well established, and considered base tech-
traditional organic versus mechanistic model (Burns nologies, namely, they offer little potential for com-
and Stalker 1961, Eisenhardt and Tabrizi 1995, Brown petitive advantage (Little 1981, Roussel et al. 1991).
and Eisenhardt 1997). The conceptual model used in Although the effort may be very large in scale, tech-
our study is discussed in the next section. nology is easily obtained and does not carry any dif-
ficulty or uncertainty in execution. Typical projects in
this category are construction, road building, bridges,
The Conceptual Model and utility installation. Another example is "build to
At this exploratory stage, we have decided to focus on print" projects in which one contractor is required
the study of technical and engineering-based projects to build a product previously developed by some-
typically resulting in a new product, process or ser- one else.
vice. This choice was based on two reasons. First, Type B-Medium Technological Uncertainty
while there are other types of projects, technically Projects (Medium-Tech). These are the most com-
based tasks capture a significant portion of today's mon industrial projects. Such projects rest mainly
activity in modern organizations. Second, much of the on existing and mature technologies; however, they
classical, as well as the later, literature has focused on may involve a limited amount of new technology
technology-based (enduring) organizations. Extend- (often one or two, but never more than 50% of the
ing the theory to projects (i.e., temporary organiza- technologies embodied). In some cases, such projects
tions) seems to be a natural evolutionary step at this incorporate a new feature which has not been tried
time. Based on our observations and earlier research, before. The new technology or feature is what usually
this paper is using task technological uncertainty and provides the competitive advantage of the product,
complexity as the main dimensions, and suggests a and thus serves as its key technology (Little 1981,
framework of four levels of uncertainty and three lev- Roussel et al. 1991). Typical projects in this category
els of complexity (Dvir et al. 1998). may include the development of a new model in a

MANAGEMENTSCIENCE/Vol. 47, No. 3, March 2001 397


55
SHENHAR
One Size Does Not Fit All Projects

Table 1 Definitionof DifferentProjectTypes

(a) Fourtypes of technologicaluncertainty

ProjectType A B C D

Name Low-Tech Medium-Tech High-Tech SuperHigh-Tech

Definition Using existingtechnologies Adaptationof familiar Integratingmanynew, Integratingkey


technologies;some new but existing,technologies technologies that do not
technologyor a new feature exist at the time of project
initiation
TypicalProjects Construction,roadbuilding, Derivativesor improvements New systems in a fast-moving New nonproven
and Examples utilities,"builtto print" of existing products; industry,e.g., computers, concepts, beyondthe
new models in a well- new militarysystems currentstate of the art,
established,stable industry e.g., Apollo,moon-
e.g., automobiles,consumer landingproject
electronics

(b) Threelevels of system scope

Scope Level 1 2 3
Name Assembly System Array
Definition Buildingor developinga collectionof Buildingor developinga collection Building,developing,or addingto
components and modules combined of subsystems and interactive a largewidespreadcollectionof
into a single unit,eitheras a elements that performa wide systems functioningtogetherto
subsystem of a largersystem, or rangeof functionsor activities achievea common purpose
a stand-aloneproductperforming
a single function
Examples A powersupply,an antenna, Computers,radar,buildings,aircraft Nationalair defense system, building
householdappliancessuch as, a city, a neighborhood,or the city's
CDplayersor washing machines publictransportationsystem

well-established industry (e.g., automobile or con- this category, as well as projects in high-tech or high-
sumer electronics), or improvements, modifications, velocity industries (Bourgeois and Eisenhardt 1988).
derivatives, and upgrades of existing products.
Type C-High Technological Uncertainty Projects Type D-Super High Technological Uncertainty
(High-Tech). These projects constitute the first use Projects (Super-High-Tech). Such projects require the
of new, but existing, technologies. Specifically, in development of new technologies that do not exist
such projects, more than 50O/% of the technologies at the time of project initiation. Some of these tech-
employed are pacing new technologies. As defined, nologies are emerging (Little 1981); others are still
such technologies have the potential to change the unknown and have to be developed during the period
basis of competition (Little 1981, Roussel et al. 1991,
of project execution. This kind of project is very risky
Eisenhardt and Tabrizi 1995). Although not yet
and relatively rare. Although it provides enormous
embodied in a product or process, these technolo-
opportunity for competitive advantage, it is usually
gies have been developed prior to the actual project
effort. Incorporating existing, but new, technologies carried out by few and probably large organizations
for the first time typically leads to products that did or government agencies. Typical known examples
not exist in the past, or are even "new to the indus- of this type are the Apollo moon-landing program
try." Many defense development projects that employ (Pellegrino and Stoff 1985) or the Hubble Space Tele-
recently developed technologies would be included in scope (Villard 1989).

398 MANAGEMENTSCIENCE/Vol. 47, No. 3, March 2001


56
SHENHAR
One Size Does Not Fit All Projects

The Complexity Dimension-System Scope Scope 3-An Array Project (or Program). An array
The notion that there are different hierarchies within is defined as a dispersed collection of systems that
a product or a system with different levels of function together to achieve a common purpose. Such
design and managerial implications is used as the systems are never placed in a single site; rather, they
second dimension for distinction among projects are spread over a wide geographical area. An array
(Table lb). We chose to conceptualize complexity can be considered a "super-system," expressing its
by a hierarchical framework of systems and sub- nature as a conjunction or conglomeration of systems.
systems (Boulding 1956, Lewis and Samuel 1989, A national air defense system with early warning
Rechtin 1991). We labeled this dimension system radar, command, and control centers, combat aircraft,
scope. Its three levels are defined as follows: and ground-to-air missiles is a good example of
such a super-system. Well-known examples of array
Scope 1-An Assembly Project. Such a project efforts are New York City's Transit Authority Cap-
deals with a single component or with a complete ital Program of modernizing its subway infrastruc-
assembly-defined as a collection of components and ture (Manne and Collins 1990), the English Channel
modules combined into a single unit. An assembly Tunnel (Lemley 1992), and the U.S. Strategic Defense
can be a subsystem performing a well-defined func- Initiative, or, as it is often called, "Stars Wars"
tion within a larger system, or it can be an inde- (Lawrence 1987). Within the other dimension, uncer-
pendent stand-alone product that performs a single tainty, these array programs can be classified as low-
function of a limited scale. A radar receiver or a com- tech, medium-tech, and super-high-tech respectively.
puter's hard drive are common examples of assem-
blies (subsystems) within larger systems. Compact Research Focus
disk players, television sets, washing machines, and In the present study we concentrated on an individ-
other household appliances are independent assem- ual project as the fundamental unit of analysis and
blies of the second kind. Using dimensions both of examined the relationship between project classifica-
uncertainty and scope, the first VCR developed in the tion and project-specific characteristics. Any project
mid-1970s (Rosenbloom and Cusumano 1987) would effort involves linking two different, though not dis-
be considered a high-tech assembly project in our jointed, processes along the project life cycle (Clark
framework. and Fujimoto 1989). The first process-the techni-
cal process-involves the reduction of technological
Scope 2-A System Project. A system is defined uncertainty, while assembling external and/or inter-
as a collection of interactive elements functioning nal pieces of technological knowledge. Essentially, it
together within a single product. However, unlike an consists of all technical activities that lead to the cre-
assembly, a system consists of many subsystems and ation and shaping of the project's final outcome. The
is capable of performing a wide range of functions second process-the managerial process-consists of
to address an operational need or mission. Projects at the management activities that are performed to com-
this level are dealing with systems such as radar, com- plete the project task within a given time frame and
puters, missiles, or communication; yet they may also other constraints. This proc6ss involves allocating,
involve a higher level of system, which consists of utilizing, and monitoring resources; coordinating the
entire platforms, such as aircraft, vessels, automobiles, parties involved; managing the communication and
or buildings. The first Macintosh computer developed information flow; and supporting the technical pro-
by Apple in the 1980s was a typical example of a high- cess via decision making and data management. In
tech system project (Guterl 1984), and building the our search for project contingencies, we have asked
famous SR-71 "Blackbird" reconnaissance aircraft by how these processes would be affected with different
Lockheed can be classified as a super-high-tech sys- levels of technological uncertainty and system scope.
tem project (Johnson and Smith 1985). However, given the early nature of our research, our

MANAGEMENTSCIENCE/Vol. 47, No. 3, March 2001 399


57
SHENHAR
One Size Does Not Fit All Projects

main goal was to validate the two-dimensional frame- studied ranged in budget from $40,000 to $2.5 billion,
work described above and suggest additional contin- and in duration from 3 months to 12 years. Out of
gency insights for future research. our sample of 127 projects, 62% were projects of new
product development, 15% were product modification
projects, and 23% were construction projects; 18% of
Methodology the projects were for the consumer market, 21% for
Research Design: Two Databases the industrial market, and 61% for the government.
Because of the exploratory nature of our research and Caution should be exercised in generalizing the
the complexity of the research problem, we performed results of this study because the projects studied here
a two-stage study that involved a combination of were not randomly selected and may not be represen-
qualitative and quantitative methods. The first stage tative of all projects in general, or in other parts of the
involved 26 projects on which we applied a multiple world. However, Israeli industry is closely coupled to
case study approach, focusing on the dynamics within Western culture, either in Europe or the United States,
single settings (Yin 1984). Specifically, we subscribed and many of the organizations involved in our study
to the process of case study research as suggested are subsidiaries or partners of American companies.
by Eisenhardt (1989). This process is particularly use- Projects for this study were mainly chosen because
ful in cases such as ours, when an a priori construct data were available to record characteristics and man-
is triangulated by multiple investigators, within-case agerial practices in real, or almost real, time. How-
and cross-case analysis, and combined with the role ever, no project was dropped because it did not fit
of literature (Glaser and Strauss 1967, Strauss 1987, the model, and there is no reason to suspect that the
Eisenhardt 1989, Kirk and Miller 1986). For this por- sample is biased in any particular way.
tion we initially approached 29 projects in 16 com- The two-dimensional typology described above
panies. The final set of projects was selected based was presented to all managers who participated in
on the clarity and detail of the data obtained. The our study. They were asked to classify their projects
second form of data collection involved the distribu- on 4-by-3-level scales, according, respectively, to our
tion of questionnaires among project managers and defined levels of technological uncertainty and sys-
the collection of detailed quantitative data on each tem scope. Almost all respondents were comfortable
project. For this portion we obtained information on with this classification, and easily placed their task
127 projects (in 76 companies) out of a total number in the appropriate categories. Less than 5% expressed
of 182 managers who were approached (70% response some doubt as to where a specific project should be
rate). placed in the uncertainty dimension. Their doubts
Data collection was performed in Israel, in firms were promptly resolved, however, after they were
operating in the military or commercial market. Out asked to do a two-step classification-first into radical
of these firms, 30 were electronics companies, 18 versus incremental change, and then into one of the
aerospace, 12 construction, 4 computer, 2 mechani- four types. Classifying scope was even easier, since
cal, and the rest represented a variety of industries almost all managers immediately acknowledged the
such as chemical, pharmaceutical, biochemical, etc. hierarchical nature of the defined scope. Our research
The largest firms in our study included an aerospace design resulted in a widespread distribution of the
company that provided information on 26 projects, surveyed projects in the two-dimensional space (see
two defense development contractors that provided Figure 1).
information on 23 and 10 projects respectively, an
electronics and communication company (6 projects), The Case Research
a computer company (5 projects), and a construc- Data collection for the first part was multifaceted
tion contractor (5 projects). The rest of the companies (Kirk and Miller 1986), and included in-depth inter-
were mostly commercial, and each provided infor- views, observations, questionnaires, documents, and
mation on less than three projects. The projects we archives. Interviews were conducted by teams of two

400 MANAGEMENT SCIENCE/Vol. 47, No. 3, March 2001


58
SHENHAR
One Size Does Not Fit All Projects

Figure1 The Two-DimensionalModel and the Distributionof Data Most data for this portion were obtained through
Projects:Case Projects/AllProjects Studied interviews in the form of open questions listed in a
System structured document that was used by all investiga-
Scope
tors. Questions were asked about the project mission
and objectives, and the motivation of the various par-
3 ties involved: the contractor, customer, and user. Data
Array were also obtained on the managerial procedures and
1/6 2/2 -/1
tools used, such as organization, planning and con-
2 trol methods, engineering design practices, computer-
System
3/17 4/23 3/36 2/8 aided and software packages, and documentation.
Finally, data were also obtained on decision-making
Assembly processes, information flow, and communication
-/5 3/19 4/8 1/2 patterns.
A B C D Technological The qualitative case data of this study were pro-
Low-Tech Medium High-Tech Super Uncertainty cessed through a method of cross-case compara-
Tech High-Tech
tive analysis (Glaser and Strauss 1967, Miles and
Huberman 1984). Multiple tables were created to clus-
or three, and they interviewed a total of 115 peo- ter typical project characteristics for various vari-
ple over a period of more than two years. In addi- ables tested. As this strategy requires, the process
tion to the project managers, interviewees included was highly iterative, with continuous comparison of
members of the project management team, functional data and theory, until patterns clearly emerged and
team members who were involved in the project, additional data no longer added to the refinement of
project managers' supervisors, and customer repre- the concepts (Eisenhardt 1989, Kirk and Miller 1986).
sentatives. At least three people were interviewed Clusters of behavior were clearly converging and they
from each project. All investigators in this portion of were summarized in a set of typical characteristics
our study were graduate students in Management of (Eisenhardt 1989) according to the underlying twofold
Technology who received, prior to its execution, at typology.
least 20 hours of training in organizational research.
To strengthen our research validity, and as is often The Quantitative Research
required by qualitative studies (Kirk and Miller 1986), During the case data collection of our study, we pre-
we insisted that investigators interact with their sub- pared a preliminary draft of the questionnaire for our
jects on their own turf, namely, at the project site. second research stage. This draft was distributed to
Notes were taken during all encounters and they were a convenience sample of 17 projects before it was
promptly summarized in writing after each interview. refined to form the final questionnaire version. Data
Following an initial phase of data collection, a draft obtained included information about the kind of work
report was prepared for each project according to a that was done. It identified the project as involving
common set of guidelines. After an intrateam reli- a new product development, a product modification,
ability test, based on thoroughness and detail, and or a construction or production effort. It also identi-
an initial integration stage of these drafts, teams fied the type of user (consumer, industrial, or gov-
were usually asked to obtain additional data to dis- ernment), and the type of industry. In addition, the
cover new facts before a final report was prepared questionnaire included several theoretical constructs
(Kirk and Miller 1986). The lengths of these reports using seven-point multi-item scales, ranging from "To
were between 40 and 120 typewritten pages. In some no extent" to "A great extent" or from "Very low"
cases the author and several of the field investigators to "Very high." These constructs related to the engi-
returned to a project to clarify additional questions neering and design practices that were used in the
and cross-check relevant data. project and to various managerial and administrative

MANAGEMENTSCIENCE/Vol. 47, No. 3, March 2001 401


59
SHENHAR
One Size Does Not Fit All Projects

variables such as extent of planning, control, modifi- additional iterations of design, building, and test-
cations, replanning, and computer utilization. Finally, ing. These iterations were part of the development
we collected data on the project's budget and dura- activities (Hoover and Jones 1991, Eisenhardt and
tion, number of personnel occupied, percentage of Tabrizi 1995), and they were defined in our study as
workers holding academic degrees, number of design design cycles. A technical project can therefore be seen
cycles, and the design freeze quartile. as a multistage logical process of design cycles, per-
Analysis of the quantitative data included test- formed to reduce uncertainty (Weick 1979). This pro-
ing the consistency of all multiscale items using cess tended to be greatly influenced by the initial level
Cronbach's alpha values. We then calculated descrip- of technological uncertainty.
tive statistics of all scale variables for each level of The data in our study indicate also that the com-
technological uncertainty and system scope. ANOVA pletion of the sequence of design cycles was marked
tests and correlation calculations were performed by an important event called design freeze. This event
for all single- and multiscale representative variables did not mean that no further changes were made; it
with our two dimensions of uncertainty and scope. did indicate, however, that the product had reached
Finally, we performed regression analyses to deter- its final projected form and that additional changes
mine the linear trends of variables and the interaction would be made only if essential. The transition
effects between our two main dimensions. from pre- to post-design freeze was characterized by
an abrupt change in the project managers' attitude
towards change. A high level of flexibility and toler-
Qualitative Findings ance for change characterized the project during the
The comparative analysis of the 26 case projects initial stage, followed by low or almost no flexibility
allowed the identification of distinct patterns of once the design was frozen. The specific differences
project management strategies, clustered according between the various types of projects will be demon-
to our grounded classification model. At the tacti- strated by the following discussion.
cal level, the study identified the various manage-
Type A Projects
ment tools and practices used in different projects
All Type A case projects in our study employed
and at different levels of uncertainty and scope. As
well-known and existing technologies, and almost all
we looked along the uncertainty dimension we found
involved some type of construction. Product archi-
typical activities for reducing technical uncertainty.
tecture, engineering design, and resources planning
Distinctions along the second dimension, system
were carried out during the conceptual and planning
scope, related to typical organizational and adminis-
phases and were usually performed by engineering
trative practices that were employed for different lev-
consulting firms. Those served as a basis for price
els of scopes. We present these patterns separately
quotation and contract negotiation with potential con-
for each dimension, followed by additional observa-
tractors who were responsible for project execution.
tions relating to a joint advancement along the two
In each one of these projects, the product was entirely
axes. Later we will use our quantitative data analysis
shaped and the design completely frozen prior to the
to support our qualitative findings and suggest addi-
execution phase.
tional insights. Projects were executed after a formal contract was
signed, and from there on, they were managed in
Reducing Uncertainty Though Design Cycles and a very formal and rigid style. Managers' main con-
Design Freeze cern was to finish the project on time and within
Shaping the product's configuration and setting its the expected budget, and in general, no changes
specifications involves the execution of many tech- were introduced.' None of these projects entailed any
nical activities such as engineering design, build-
ing, assembling, testing, and approving. In some 1 As one manager put it: "We are in this business to make money
of the cases, the completion of the design required when finishing our projects on time. To do so, I must be firm and

402 MANAGEMENT SCIENCE/Vol. 47, No. 3, March 2001


60
SHENHAR
One Size Does Not Fit All Projects

development, testing, or redesign. Communication in the past. Some of these products were even "new
between management teams and subcontractors was to the industry" and constituted a new product line
typically conducted through formal channels, docu- for the company. The companies involved in these
ments, existing forms, and regular meetings on a low- projects were all in the aerospace, electronics, and
rate basis, usually once every month or two weeks. computer industries. In half of the cases, projects were
based on technical feasibility rather than market need
Type B Projects (Marquis 1969), and were initiated by the contractor.3
Substantial differences were found between this Type C projects were characterized by long periods
group and the previous, Type A projects. The projects of development, testing, and redesign. Design freeze
observed in our study included building a new was scheduled, in most cases, to be in the second
product in a well-established industry, developing a or even the third quarter of the project's duration,
derivative or modification of a previous design to and it was not concluded until two or even three
achieve better performance, increased reliability, or design cycles were performed. During this period
extended operational life. In contrast to the Type A many changes were made before the product's spec-
projects, the contractors undertaking these projects ifications were finalized. In comparison to Types A
were responsible for the entire range of activities, and B projects, managers of Type C projects had to
from engineering design, to resources planning, to employ a much more flexible attitude, and they had to
execution. Although the technologies employed were make extensive trade-offs. In at least two of the cases,
not entirely new, almost all projects in this category the initial requirements could not be met without a
involved some development and testing. However, substantial addition of time and budget. In these cases
only limited changes were added to the initial design. customers were asked and agreed to waive some of
Management's policy in these cases was usually to their requirements.4
resist change, and managers were highly aware of the Formal and informal communication among project
need to avoid excessive costs.2 Design was usually teams, as well as with the customer, was usually
frozen early, no later than the first or second quarter intensive. It included written information in the form
of the project's execution period, after one or at most of status reports, computer printouts, minutes, mes-
two design cycles, and none of these projects had uti-
sages, and memos.5 However, the major flow of infor-
lized a formal risk-management procedure.
mation was oral, and was conducted during meetings
The communication pattern in our Type B projects
for problem solving and information sharing. These
was more intense than in those categorized as Type A.
There were regular weekly or biweekly meetings
3One project manager described this process: "We had this idea for
of the management team, as well as biweekly or
years. We knew it could be done; the problem was to sell it to the
monthly meetings with major subcontractors. Addi- customer (in this case the military). We were able to get a contract
tional in-between communications were conducted only after four years and numerous technological demonstrations
through ad hoc meetings, telephone discussions, and that proved the validity of our new concept to various management
e-mail correspondence to resolve occasional problems. levels within the customer's organization."
'In one project all specifications could be achieved except one high-
Type C Projects end requirement. It became clear to all parties that this requirement
would involve enormous additional resources. The project man-
Most high-tech projects in our study produced com-
ager recalls: "Our customer understood our problem. To him this
pletely new products or systems that did not exist requirement was really marginal compared to the additional time
needed. However, he required that we make sure two other speci-
inflexible. I resist any changes or new ideas. If the customer wants fications would be met completely."
a change he must pay for it." 5Since some of these projects were in aerospace and electronics
2
One project manager expressed his strategy by saying: "Our pol- industries and had early access to electronic mail systems, they
icy is to add value to the product without adding cost. We will have used this medium as an additional form of project com-
therefore use the previous product as much as we can; we are not munication, thus increasing their "richness of media" (Daft and
trying to be perfect and we do not need too many improvements." Lengel 1986).

MANAGEMENTSCIENCE/Vol. 47, No. 3, March 2001 403


61
SHENHAR
One Size Does Not Fit All Projects

consisted of internal meetings of the project team, required enormous amounts of information exchange
meetings with all subcontactors (both together and and extensive communication. All team members
separately) and meetings with the customers. In gen- were expected to share immediate information, and
eral, the atmosphere in these projects was one of open no one waited for the formal meetings and documents
communication and continuous discussion. In some to report problems and difficulties.7
cases, managers have initiated social events, like par-
ties, barbecues, and field trips, to increase interaction Addressing System Scope: Managing Resources
among team members and to reinforce group cohe- and Project Administration
siveness and spirit. Project planning typically starts by breaking the work
into a "Work Breakdown Structure" (WBS) in a tree-
Type D Projects like form and identifying all product subunits and
The three Type D case projects in our data were support activities (Lavold 1988). Each activity is then
defense projects, envisioned to respond to some far- budgeted and its projected length is estimated. This
reaching needs. As defined, no adequate technology process results in a project schedule and budget,
was available at project initiation. The major concern which are often set as constraints for project manage-
was the extremely high level of uncertainty as to ment. As we observed, when system scope increased,
what technology should be used, and how to resolve this process became more intense, more detailed,
it. Customers' decision to commit themselves to the and more formal. Another major difference that was
project was typically marked by a great deal of hes- found, however, was the project organizational struc-
itation because of the unknown technologies and the ture. The following discussion summarizes the main
risk involved. distinctions that were observed among projects for
All projects in this category used a similar tech- different scope levels.
nique to resolve the issue of unknown technologies.
They involved an intermediate program in which Scope 1 Projects
an experimental, scaled-down prototype model was Several of the projects in our study were set up to
developed and built. These intermediate programs build a unit or a module that would become part of
were instituted to prove the validity of the system's a larger system. Others involved stand-alone prod-
concept and to test unknown technologies.6 The deci- ucts designed to be used as is. Most of the project
sion to freeze product design and to set its specifica- was done in-house, and the responsibility for the
tions was therefore scheduled for a late moment, often project completion was usually within one functional
during the third quarter of the project life cycle, and group, engaging, at times, additional disciplines from
the typical number of design cycles was two or three, other functions. Team members knew each other very
with one exception of five. well and the atmosphere, in general, was casual and
Management styles required high levels of flexibil- informal.
ity and tolerance for change, and high awareness of Resources planning and scheduling was relatively
potential problems. The atmosphere can be character- simple, done either manually, or with a personal com-
ized as: "Look for trouble-it must be there; if you puter software package designed to handle no more
don't see it, you have a problem." The high level than a few hundred activities. Control was also sim-
of uncertainty and the continuous flow of changes ple, consisting mainly of budget and milestone mon-
itoring. Most of the documents were technical, with
some supplemental managerial documents, includ-
6One project manager described the intermediate program: "Like
in the moon-landing program, this served as our less ambitious ing a general milestone and work plan, financial
'Gemini' program before the full-scale launch of 'Apollo.' Through
this model we learned a lot about the new concept. We could 'As one project manager explained: "I require that any major prob-
test our algorithms for solution, simulate the behavior of various lem in the project be brought to my attention within half an hour
parameters and decide which of the possible technologies should from its emergence. If I am not available, everyone in the chain of
be integrated into the system." command must know about the problem."

404 MANAGEMENTSCIENCE/Vol. 47, No. 3, March 2001


62
SHENHAR
One Size Does Not Fit All Projects

and man-hour reports, and purchasing documents. and reviews.9 In most cases there were at least ten
Finally, communication and decision making within formal documents relating to various technical and
the project team were informal too. Although the gen- managerial aspects of the project. Some projects had
eral informal attitude usually contributed to techni- to develop their own format to fit their specific needs
cal success, it sometimes impaired managerial aspects, and organizations.10
resulting in less order, which led to excessive costs.8
Scope 3 Projects
Scope 2 Projects All the array projects we studied consisted of siz-
Our fifteen case projects in this category included able programs that simultaneously managed many
development or construction of complex systems that other projects. However, unlike the previous Scope 2
were designed either to function independently or to projects, they posed much less of an integration
be installed on another, larger system, such as an air- problem. They were not required to build a single
craft or vessel. Managerial styles and related prac- integrated product that had to be delivered to the
tices were the same, however, for both kinds. Each of customer at one time. Arrays were built in an evolu-
the system projects we studied had a main contrac- tionary manner, and their various components were
tor who was responsible for the final product. The finalized and supplied at different times and even
entire effort was divided among several subcontrac- in different places. Typically, the program structure
tors, either in-house or external. The main contrac- involved a large "umbrella" organization whose pri-
tor was in charge, though, of the final integration of mary mission was to set goals, direct, and coordinate
the product, and was responsible for meeting perfor- the efforts of many subprojects. In two cases the array
mance, quality, time, and budget goals. Work within organization was even established as an ongoing pro-
the leading organization was usually done in a matrix gram to which more projects could be continuously
form, and was led by a project management office added.
interacting with various functional departments and The dispersed nature of the end product and the
dealing with outside organizations through separate extent of subcontracting made it necessary to manage
contracts. these programs in a very formal way and to put a
Dividing the work among separate subcontrac- lot of effort into the legal aspects of numerous con-
tors, and performing the coordination among them, tracts. In addition, program managers were mainly
required considerable managerial efforts: defining occupied with the direction and mission of the pro-
and analyzing customer needs, planning the pro- gram as a whole and with financial and budgetary
gram resources, negotiating with all subcontractors, controls, while they were less concerned with techni-
and instituting a complicated system of coordination, cal aspects, which were usually left to the managers
control, decision, and information gathering. Man- of subprojects.
agement, in general, tended to 'bureaucratize' the
project by installing a system of procedures, doc- 'In three of the case projects in this category we found, in addition
uments, management tools, meetings, reviews, and to the usual status and cost reports, a "Cost Performance Index"
organizational structure. Project control was just as report or an "Earned Value" report that expressed the project's
combined status of money spent and the actual work achieved in
complex, and required extensive reports, meetings,
terms of financial figures.
'0For example, in one of the projects we found a "Termination
8'ne of the team members explained: "The only problem with Price Report" to continuously assess the cost in case the customer
this form of communication was that the contents of these discus- decides to terminate the project. Another project has used a "Level
sions and even the decisions reached were not always recorded. of Effort" (LoE) document that classified all activities into those
We finally overcame this question by creating a 'master docu- that have a direct impact on the program and those that do not.
ment' that contained all the technical decisions that were reached This enabled management to concentrate its efforts on activities that
and was continuously maintained and updated by one of the directly affect the probability of success, rather than waste time on
team members." non-LoE activities.

MANAGEMENTSCIENCE/Vol. 47, No. 3, March 2001 405


63
SHENHAR
One Size Does Not Fit All Projects

Managing the Combination of High Uncertainty of technological uncertainty and system scope. As
and High Scope mentioned, the goal of the quantitative portion of
Managing the execution of high-scope, high- our study was to strengthen the validity of the two-
uncertainty projects required more than just detailed dimensional model by testing the statistical differ-
planning and technical skills. These projects were ences in managerial variables among different project
designed to produce large multidisciplinary systems, types and identifying various contingency trends. To
which involved numerous new technologies and do so, we used all scale assessment variables as they
many subsystems and components. The case data appeared in the questionnaire and have quantified all
indicate that most of these projects found it neces- other variables as described hereafter.
sary to incorporate the tools of system engineering to Table 2 contains information about the resources
optimally harmonize an ensemble of subsystems and consumed by different project types. It includes the
components (Booton and Ramo 1984). descriptive statistics for various levels of uncertainty
Because many of the components of large system and scope. The scale value associated with budget
projects were developed by external subcontractors, was coded into: 1-less than $100,000; 2-$100,000 to
the main contractor faced a difficult problem of sys- $1M; 3-$1 to $10M; 4-$10 to $100M; 5-$100M to
tem integration. Problems of interfaces, energy dissi- $1B; and 6-more than $1B. The project's duration
pation, and even lack of space, required a long and scale value was coded into: 1-less than 6 months;
tedious process of assembly, testing, and necessary 2-between 6 months and two years; 3-two to four
trade-off,11 and resulted, in some cases, in more than years; 4-four to eight years; and 6-more than eight
one design cycle of the entire system. Finally, there years. The other two variables were the average num-
were problems of configuration and risk manage- ber of employees during execution and the percentage
ment. Special software was used in high-uncertainty, of people holding academic degrees. Table 2 also con-
high-scope projects to keep track of all the decisions tains the results of ANOVA tests for each variable. It
and changes and to identify potential interactions that also contains Pearson correlation coefficients between
would occur with each change. As for risks, higher- these variables and the two dimensions of uncertainty
scope, higher-tech projects were more receptive to the and scope, where uncertainty was quantified into 1-
need for systematic risk analysis and management. Type A, 2-Type B, 3-Type C, and 4-Type D, and
The objective of the risk-management program was scope into 1-1, 2-2, and 3-3.
not to eliminate risk, but to balance it across the As we found, there seems to be an association
project, so as to avoid investing excessive resources in between scope and size. Both budget and duration
the resolution of a given risk while neglecting another. were significantly increased with scope. Yet, a simi-
lar trend was also observed for uncertainty; namely,
higher uncertainty required increased budgets and
Quantitative Analysis longer projects. Still, while the average number of
In the qualitative part of our research we found con- workers employed on the project was found to be
sistent patterns of behavior for different categories positively associated with scope, this number did not
increase with technological uncertainty. The increase
" One of the project managers of a high-tech system in our research in budget and time of higher-tech projects can be
admitted that no integration problems were anticipated when the attributed to the project's complexity and not to the
project was initiated. All key members of the project manage- need to employ more people. However, higher-tech
ment team were relatively young, with previous experience only in projects employ more academicians than lower-tech
smaller-scale, though high-tech, projects. Once all subsystems were projects. This trend was positively correlated and sig-
ready, they expected a very short integration period. It resulted,
nificant. In contrast, we found that this percentage
however, in numerous problems of combining units of various dis-
ciplines and poor functioning of the system as a whole. External tends to decrease with scope, probably because build-
experts had to be called in, and it took another year before all the ing high-scope projects requires a large number of
integration problems were resolved. builders and craft workers and a smaller portion

406 MANAGEMENTSCIENCE/Vol. 47, No. 3, March 2001

64
SHENHAR
One Size Does Not Fit All Projects

Table2 ProjectResources for VariousLevels of Uncertaintyand Scope

TechnologicalUncertainty System Scope

A B C D 1 2 3
ANOVA Corr. ANOVA Corr.
Mean Mean Mean Mean Mean Mean Mean
Variables (S.D.) (S.D.) (S.D.) (S.D.) df F (S.D.) (S.D.) (S.D.) df F

1. Projectbudget 3.03 3.11 3.51 3.70 3,123 2.04 .218 2.47 3.47 4.55 2,124 31.45 .548
- scale level (1.17) (1.16) (0.78) (0.67) * (0.99) (0.81) (0.88) *** ***
2. Projectduration 2.39 2.88 3.15 3.40 3,123 6.12 .318 2.50 3.00 3.66 2,124 8.12 .308
- scale level (1.10) (0.78) (0.95) (0.96) *** *** (0.96) (0.90) (1.11) ** ***
3. Averagelabor 142 45 38 80 3,123 1.99 -.142 11 54 393 2,124 18.26 .359
employed (382) (90) (36) (126) (17) (79) (623) *** ***
4. Percentageof 20.2 55.1 59.3 66.6 3,123 15.2 .452 57.2 49.6 23.3 2,124 4.52 -.229
acad. degrees (29.9) (27.2) (25.5) (17) *** *** (28.0) (31.3) (23.1) * **

*p<0.05
**p< 0.01
***p< 0.001

Table3 Results for VariousLevels of TechnologicalUncertaintiesand System Scope: Engineering-and


DescriptiveStatistics and ANOVA
Design-RelatedVariables

TechnologicalUncertainty System Scope

A B C D 1 2 3
ANOVA Corr ANOVA Corr
Mean Mean Mean Mean Mean Mean Mean
Variables (S.D.) (S.D.) (S.D.) (S.D.) df F (S.D.) (S.D.) (S.D.) df F
(Alpha)
Design Cycles 1.03 2.07 2.60 2.70 3,121 28.7 .608 1.97 2.20 1.44 2,122 3.1 -.022
(0.33) (0.60) (0.95) (1.2) *** * ** (0.77) (1.1) (0.53) *
Design Freeze .25 1.95 2.3 2.6 3,121 37.65 .581 1.75 1.83 0.77 2,122 3.39 -.125
(0.52) (1.0) (0.83) (0.96) *** * ** (1.1) (1.1) (1.2) *

Design Considerations 4.65 5.20 5.26 5.26 3,114 1.09 .134 4.75 5.27 5.26 2,115 1.74 .150
(0.91) (2.1) (1.0) (1.20) (1.1) (1.7) (1.1) (1.5)
Design Reviews 3.13 5.29 5.97 5.6 3,117 12.66 .416 4.63 5.57 3.65 2,118 5.12 .050
(0.78) (2.3) (1.9) (1.3) (1.8) *** * ** (2.2) (1.8) (2.0) **
RiskManagement 1.87 2.38 2.8 3.25 3, 89 2.07 .255 2.22 2.81 2.2 2, 90 1.80 .107
(0.83) (1.7) (1.5) (1.4) (0.94) * (1.4) (1.5) (1.3)
Systems Engineering 2.74 3.95 4.99 4.58 3, 92 6.31 .364 3.20 4.84 3.81 2, 93 8.55 .264
(0.86) (2.2) (1.8) (1.5) (1.8) ** * ** (1.9) (1.6) (2.5) *** ***
QualityManagement 3.59 3.87 4.72 4.85 3, 96 2.35 .247 3.52 4.63 4.28 2, 97 3.71 .214
(0.87) (2.3) (1.8) (1.7) (1.5) * (1.9) (1.8) (1.6) * *

*p < 0.05
**p< 0.01
***p< 0.001

MANAGEMENT SCIENCE/Vol. 47, No. 3, March 2001 407


65
SHENHAR
One Size Does Not Fit All Projects

Table4 Descriptive Statistics and ANOVAResults for Various Levels of Technological Uncertaintyand System Scope Managerialand
AdministrativeVariables

TechnologicalUncertainty System Scope

A B C D 1 2 3
ANOVA Corr ANOVA Corr
Mean Mean Mean Mean Mean Mean Mean
Variables (S.D.) (S.D.) (S.D.) (S.D.) df F (S.D.) (S.D.) (S.D.) df F
(Alpha)
Activities 1.50 1.66 1.93 2.30 3,121 12.16 0.301 1.18 1.89 2.66 2,122 46.2 0.524
(0.83) (0.72) (0.69) (0.48) *** *** (0.47) (0.68) (0.86) *** ***
WorkBreakdown 4.45 4.62 5.06 4.83 3,104 0.51 0.097 3.75 5.09 5.91 2,105 7.07 0.340
Structure(.94) (2.1) (2.1) (1.6) (1.8) (2.2) (1.6) (0.91) ** ***

Planning 3.92 4.83 5.20 6.06 3,120 5.95 0.351 3.99 5.26 4.66 2,121 7.95 0.247
(0.72) (1.9) (1.4) (1.4) (1.1) *** *** (1.7) (1.4) (1.5) ** **
Control 4.12 4.48 4.89 5.12 3,123 2.15 0.170 3.90 4.82 5.14 2,124 5.88 0.282
(0.82) (1.8) (1.5) (1.1) (1.5) (1.5) (1.3) (1.2) ** **
Documentation 4.95 5.21 5.47 5.73 3,122 1.30 0.081 4.75 5.45 5.91 2,123 4.49 0.168
(0.83) (1.5) (1.3) (1.2) (0.97) (1.6) (1.1) (0.96) *
Contracting 4.90 5.43 5.59 5.50 3,122 1.19 0.147 4.49 5.66 6.00 2,123 8.24 0.236
(0.71) (1.9) (1.6) (1.5) (1.5) (1.9) (1.3) (1.2) *** **
Consultation 4.01 5.10 5.49 4.90 3,117 3.85 0.227 4.60 5.27 3.62 2,118 4.07 0.011
(0.81) (2.3) (1.6) (1.5) (1.5) * * (1.9) (1.7) (1.9) *

p < 0.05
p < 0.01
***p< 0.001

of academic personnel who are usually engaged in for serviceability, etc. Together, these variables did
design, planning, analysis, and testing. not demonstrate significant association with techno-
A major distinction between projects was observed, logical uncertainty. The risk management measure
however, for several managerial and technical vari- represents five scale variables assessing issues such
ables (Tables 3 and 4). Table 3 includes the descrip- as initial identification of project risks, probabilistic
tive statistics, analysis of variance, and correlation assessment of risks, and a detailed plan for risk mit-
coefficients obtained for variables which relate to igation. The systems engineering measure included
engineering design and activities performed to reduce four variables such as usage of structured systems
uncertainty. The first two variables describe the num- engineering procedures, configuration management,
ber of design cycles performed before the design was and usage of various types of software. Also, qual-
frozen and the quarter in which the design freeze ity management represents four variables measuring
took place (zero means the design was frozen prior the extent to which a total quality plan was prepared,
to the project's initiation). Both measures were sig- quality goals were selected, and statistical control was
nificantly associated with technological uncertainty. performed in the project. All three variables increased
The rest of Table 3 describes several combined vari- with technological uncertainty, but only systems engi-
ables of the project management process, together neering showed significant association.
with their alpha reliability measures. For example, the As observed in the case research part, the quan-
design consideration measure represents eight seven- titative results show a clear (and often significant)
scale variables assessing the extent to which man- increase in almost all variables with the level of
agers were concerned with issues such as design for technological uncertainty. Higher-technology projects
manufacturability, design for maintainability, design required more design cycles, later design freeze,

408 MANAGEMENTSCIENCE/Vol. 47, No. 3, March 2001

66
SHENHAR
One Size Does Not Fit All Projects

and increased attention to design considerations, risk project scope increases. Our results seem to support
management, systems engineering, and quality man- the trends observed in the qualitative part. As can be
agement (Eisenhardt and Tabrizi 1995). Fewer of these seen, all variables in Table 4 are significantly associ-
trends, however, were seen to characterize an increase ated with system scope. Similarly, some of these vari-
in system scope. In fact, increase in scope demon- ables were positively associated with technological
strated in almost all cases a curvilinear pattern in uncertainty, indicating, for example, the need for bet-
engineering and design-related variables. While many ter planning and control in high- and super high-tech
values increased from assembly to system projects, projects.
they then declined at the array level. To explain this
trend, one needs to look at the nature of system and
array projects. Compared to assembly projects, system
efforts are typically associated with extensive engi- Interactive Effects
neering tasks which require careful technical design, Tables 5 and 6 include the results of a two-step hierar-
numerous testing and reviews, detailed risk and qual- chical regression analysis. A regression equation was
ity management, and extensive systems engineering obtained for our two dimensions of uncertainty and
activities (Rechtin 1991, Jansiti 1997). Array projects, scope for each one of our quantitative variables. The
however, are much less technical in nature, and thus, R square levels obtained for this phase were between
these tasks will get less emphasis on the entire array 0.043 and 0.375. The multiplication effect of uncer-
effort. These projects would typically employ fewer tainty and scope was added during the second step.
quality and systems engineers, and would be less con- Several variables exhibit an interactive effect demon-
cerned with design reviews or design cycles of the strated by the additional variance, thus supporting
entire program. Together, this difference may explain our qualitative observations. The main interaction
the curvilinear change of variables. appeared in project resources such as budget, dura-
Table 4 includes the results obtained for various tion, and labor, with some additional interaction in
managerial and administrative variables. The first project control, documentation, contracting, risk man-
variable, the number of activities included in the agement, and design considerations.
project's planning network, was coded into: 1-less
As mentioned in the qualitative part, high lev-
than 100; 2-between 100 and 1,000; 3-between 1,000
els of technological uncertainty and system scope
and 10,000; and 4-more than 10,000. The rest of
characterize projects that are designed to produce
the variables in Table 4 were seven-point scale com-
large, multidisciplinary systems. Managing such sig-
bined measures representing the extent to which for-
nificant efforts requires a delicate balance between
mal methods were utilized in each of these groups
two challenges-technical and managerial. Managers
of variables. For example, work breakdown struc-
must pay attention to numerous design considera-
ture represents nine variables, measuring the extent to
which such structure was built for the system, prod- tions, work carefully to reduce project uncertainty and
uct, development, testing, logistics, and management. risk, and make technical trade-off decisions at the
Planning includes three variables, assessing comput- interface between scientific disciplines. At the same
erized planning methods, detailed milestones, and time, they must be aware of multiple managerial and
integrative planning of budget and schedule. Consul- administrative issues, such as planning and control-
tation represents three variables, measuring the extent ling the large effort, formalizing the process through
of customer involvement and consultation for artic- detailed documentation, and carefully preparing and
ulating customer need, concept selection, specifica- monitoring contracting engagements. The interaction
tions, and ongoing problems. effects demonstrate clearly that a combined increase
Unlike the previous group, here system scope was in scope and uncertainty amplifies project complexity
the dominating dimension. This clearly indicates the and requires additional resources and more detailed
need to resort to more formal procedures when managerial attention.

MANAGEMENTSCIENCE/Vol. 47, No. 3, March 2001 409


67
SHENHAR
One Size Does Not Fit All Projects

Table5 Results of Two-StepRegression Analysisa:Resources and Design-RelatedVariables

Uncertainty Scope Uncertaintyx Scope


Variable b b R2 b AR2 F df F(AR2)
Budget 0.298** 1.060*** 0.368 -0.429** 0.035 27.70*** 3,123 7.21***
Duration 0.371*** 0.593*** 0.212 -0.423** 0.039 13.70*** 3,123 6.40***
AverageNumberof -24.900 123.400*** 0.142 -109.500** 0.065 10.73*** 3,123 10.08***
Workers
Percentageof 14.970*** -11.020** 0.242 14.940** 0.048 16.77*** 3,123 8.31***
AcademicDegrees
Design Cycles 0.660*** 0.046 0.370 0.084 0.002 23.86*** 3,121 0.38
Design Freeze 0.785*** -0.150 0.375 0.021 0.000 24.22*** 3,121 0
Design 0.222 0.383 0.043 _0.439t 0.023 2.69* 3,114 2.80*
Considerations
RiskManagement 0.441* 0.240 0.073 -0.352 0.015 2.81* 3, 89 1.46
Systems Engineering 0.803*** 0.883** 0.202 -0.025 0.000 7.78*** 3, 92 0
QualityManagement 0.518** 0.693** 0.104 -0.113 0.001 3.75** 3, 96 0.11

aUnstandardized coefficientsare shown; F is for the final regressionequation.


tp < 0.10
*p<0.05
**p< 0.01
***p< 0.001

Table 6 Results of Two-StepRegression Analysisa:Managerialand AdministrativeVariables


Uncertainty Scope Uncertaintyx Scope

Variable b b R2 b AR2 F df F(A R2)


Activities 0.290*** 0.766*** 0.396 -0.100 0.004 26.62*** 3,120 0.80
WorkBreakdown 0.265 1.230*** 0.129 -0.421 0.009 5.53** 3,104 1.08
Structure
Planning 0.679*** 0.815*** 0.198 0.003 0.000 9.86*** 3,120 0
Control 0.398** 0.801*** 0.139 -0.595 0.34 8.56*** 3,123 5.05**
Documentation 0.281** 0.659** 0.103 -0.461 t 0.24 5.89** 3,122 3.35*
Design Reviews 0.976** 0.232 0.177 -0.195 0.002 8.49*** 3,117 0.28
Contracting 0.296** 0.971** 0.135 -0.514 0.27 7.84*** 3,122 3.93*
Consultation 1.681** 0.409 0.062 -1.233 0.008 3.01** 3,120 1.03

aUnstandardized coefficientsare shown; F is for the final regressionequation.


tp < 0.10
*p<0.05
**p< 0.01
***p< 0.001

410 MANAGEMENTSCIENCE/Vol. 47, No. 3, March 2001

68
SHENHAR
One Size Does Not Fit All Projects

Discussion and Insights array projects deal mainly with administrative and
Given the scanty theoretical basis of project manage- legal issues and leave the technical details to their
ment, this research may be seen as an early step subunits.
toward building a theory of project management. It The two-dimensional model and the distinction
proves that projects indeed have a wide range of between four levels of technological uncertainty and
variations, and that "one size does not fit all." The three levels of system scope provides, to some extent,
two dimensions, technological uncertainty and sys- an orthogonal framework for looking at engineering
tem scope, seem to be dominant factors which affect projects. Furthermore, the distinction between four
project characteristics and managerial styles, and they levels of technological uncertainty and three levels of
provide several important insights for the study of system scope provides more than just a classification
projects. First, this study demonstrates that the tra- system of technical projects. Rather, this classification
ditional low-high (or incremental-radical) dichotomy seems to meet the criteria for a typological theory of
seems inadequate in describing the wide spectrum organizations (Doty and Glick 1994), and the distinct
of today's projects. Both Type A and B projects on levels of uncertainty and scope can be seen as theo-
the uncertainty scale can be considered incremental retical "ideal types" (Shenhar and Dvir 1996). Finally,
innovations. However, Type A projects are mostly there also seems to be a notable interaction between
construction or installation efforts, while Type B are the two dimensions, and many variables such as
development projects, which involve building new project resources, project documentation and control,
models within a well-established product line, or project contracting, and various design considerations
improvement of existing products. A similar distinc- are impacted by a simultaneous increase in uncer-
tion can be made between C and D types, which tainty and scope.
both represent a radical innovation in the traditional New insights may also be gained from this study
view. Type C projects are based on existing, state-of- on some of the classical domains of structural con-
the-art technologies symbolizing the existing frontiers tingency theory. From an organization theory per-
of technological advancement, while the more risky spective, our findings seem to challenge the classical
(and thus rare) Type D projects are set up to achieve distinction between mechanistic and organic organi-
an even higher goal-one which is beyond existing zations, and the traditional link between organic pro-
state-of-the-art. cesses and uncertain situations. As we learned, while
As we have seen, moving along the uncertainty project organizations may manifest numerous forms,
dimension is mainly associated with the way techni- none is simply identified with the traditional modes,
cal problems are resolved. It affects number of design nor do they vary in accordance with the classical dis-
cycles, time committed to design changes, the need tinctions. It seems that while management of highly
for prototype building, the extent of testing, and the uncertain projects is indeed more flexible and less
frequency and complexity of trade-off decisions. It formal, a central portion of the organic organization
seems that the second dimension, system scope, is is still missing-the breadth aspect. High-tech and
mainly associated with extent of administrative issues super high-tech projects must rely on the knowledge
and degree of formality of mangerial processes. As and depth of highly educated and experienced peo-
scope increases, projects are managed with additional ple in specific narrow fields. Breadth and integration
attention to planning, control, and coordination; they are only added with the second dimension-system
usually resort to a larger number of external sub- scope, namely, when highly uncertain projects are also
contractors, often use additional legal help, and are becoming large and complex.
generally characterized by increased bureaucracy and Our qualitative and quantitative observations seem
documentation. Assembly projects are typically con- also to be challenging the classical trade-off between
ducted within one internal group, in a rather informal rich and less rich media of communication (Daft
way. System projects use a central office to coordinate and Lengel 1986). It seems that in most projects,
the close integration of numerous subcontractors, and the choice is not between rich and nonrich media;

MANAGEMENTSCIENCE/Vol. 47, No. 3, March 2001 411


69
SHENHAR
One Size Does Not Fit All Projects

namely, they do not use one form or the other. To possess exceptional technical skills, as well as the
some extent, all projects use the lower end of the capability to assess potential value and risk in new, or
richness spectrum. They employ written reports and not yet developed, technology. Similarly, while assem-
quantitative data for purposes of monitoring and bly projects do not require extensive managerial skills,
status reporting. However, more rich media forms managers of system efforts need a wealth of admin-
are typically being added with increased technolog- istrative and organizational capabilities. They should
ical uncertainty, in most cases by additional team be mature and experienced, able to see the system
meetings and fostering an informal climate within as a whole and to understand the collective effect of
the project. When increasing the second dimension- its separate components. When dealing with arrays,
system scope-additional, more formal, and less rich managers must be ready to back off from technical
media are also employed. However, high-uncertainty, matters, developing instead a broader view of the
high-complexity projects use both formal and infor- industry, legal, environmental, and political issues.
mal means of communication. The second area for adapting project management
Finally, perhaps the most important insight is the practices to project type is project organization and
extent and speed with which decisions are made on processes. Assembly projects will use a small, usu-
project sites. Projects, particularly those in the higher- ally functional, organization with simple processes
tech categories, can be seen as "decision intense and tools. In contrast, system projects must establish a
environments." When strategic decisions made by project office that will handle the subcontracting and
top management reach the project floor, they typ- integration efforts and use a more formal process; and
ically perpetuate a constant stream of subsequent array projects will have to build an umbrella orga-
decisions with much higher velocity and density nization for coordination and for handling legal and
(Bourgeois and Eisenhardt 1988). As we observed,
external connections, and use a much more "hands-
project managers are required to maintain a balance
off" approach. Project communication, in turn, will
between inward and outward attention, planning and
also be determined by technological uncertainty. At
action, differentiation and integration (Lawrence and
the low and medium levels, communication will be
Lorsch 1967, lansiti 1997), and formal and informal
less intense and frequent than at the high and super
conduct (Brown and Eisenhardt 1998).
high levels. At these levels, managers of projects must
establish numerous formal and informal communica-
Implications and Conclusions tion channels for interaction among team members.
In addition to its theoretical insights, our study offers Finally, since different projects are associated with
a handful of implications for management. First, it various outcomes on the one hand, and with various
demonstrates that management and organizations at levels of risk on the other hand, organizations may
large should deliberately adopt a more project-specific use the framework of this research for a more rigor-
approach to project management. Although, as we ous process of weighing risk and opportunities and
learned, most organizations are implicitly using dif- for selecting a balanced portfolio of projects. Typi-
ferent strategies for different projects, there is usu- cally, an organization would concentrate on one (or at
ally no clear identification of project type prior to most two) type of projects, while engaging at times in
project initiation and no conscious adaptation of man- more risky and complex ones. For example, a defense
agement styles. Organizations should add a formal contractor would typically execute Type C, high-
step of project classification to the traditional plan- tech projects, but might attempt, at times, a super
ning phase, and follow it by a myriad of organiza- high-tech project with an objective of leapfrogging
tional implications. The specific project type should competition. And a consumer electronics firm might
affect the selection of project leaders, project team commonly launch medium-tech assembly projects,
members, and skill development needs. For example, but would sometimes move into a system, and even
leaders of high-tech or super high-tech projects must one or two high-tech projects.

412 MANAGEMENT SCIENCE/Vol. 47, No. 3, March 2001


70
SHENHAR
One Size Does Not Fit All Projects

As an exploratory step in the evolving process Booton, R. C., S. Ramo. 1984. The development of systems engi-
of theory building, more research seems appropri- neering. IEEE Trans. on Aerospaceand Electr. Systems AES-20(4)
306-309.
ate to establish additional validity of contingencies in
Boulding, K. 1956. General systems theory: The skeleton of science.
projects and to further explore the "one size does not Management Sci. 2 (April) 197-208.
fit all" paradigm. For example, further studies may Bourgeois, III, L. J., K. M. Eisenhardt. 1988. Strategic decision pro-
use this scheme to build a theory that explains how cesses in high velocity environments: Four cases in the micro-
uncertainty is related to project success and addresses computer industry. Management Sci. 34(7) 816-835.
the issue of fit between project classification, project Brown, S. L., K. M. Eisenhardt. 1997. The art of continuous change:
Linking complexity theory and time-paced evolution in relent-
management style, and project effectiveness (Drazin
lessly shifting organizations. Admin. Sci. Quart. 42 1-34.
and van de Ven 1985). And if our levels of uncer- .,I . 1998. Competingon the Edge: Strategy as Structured
tainty and scope are seen as "ideal types" (Doty and Chaos. Harvard Business School Press, Boston, MA.
Glick 1994), then how will a deviation of actual man- Burgelman, R. A. 1983. A process model of internal corporate ven-
agement styles from these types affect the project suc- turing in the diversified major firm. Admin. Sci. Quart. 28
223-244.
cess and its effectiveness? Finally, more research is
Burkhardt, M. E., D. J. Brass. 1990. Changing patterns or patterns of
clearly needed to explore additional dimensions of change: The effect of a change in technology on social network
project contingency. The next candidates, in our view, structure and power. Admin. Sci. Quart. 35 104-127.
would be market uncertainty (Wheelwright and Clark Burns, T., G. Stalker. 1961. The Management of Innovation. Tavistock,
1992) and project pace (Brown and Eisenhardt 1998). London, U.K.
Cash, J. I. Jr., W. F. McFarlan, J. L. McKenney. 1988. CorporateInfor-
Additional studies of traditional contingency theo-
mation Systems Management. Irwin, Homewood, IL.
ries in new domains may evolve into new organiza- Clark, K. B., T. Fujimoto. 1989. Lead time in automobile product
tional paradigms that are more applicable to today's development: Explaining the Japanese advantage. J. Engrg. and
dynamic and continuously changing organizations. Tech.Management 6 25-58.
Cleland, D. I., W. R. King. 1983. Systeins Analysis and Project Man-
agement. McGraw-Hill, New York.
Acknowledgment
Daft, R. L., R. H. Lengel. 1986. Organizational information require-
The author wishes to thank Dov Dvir, Dov Eden, Avi Nir, and
ments, media richness and structural design. Management Sci.
Max Wideman for their comments on earlier versions of this paper.
32 554-571.
The author would also like to thank the department editor and the
Dewar, R. D., J. E. Dutton. 1986. The adoption of radical and incre-
referees for their ideas and comments on the earlier draft of this
mental innovations: An empirical analysis. Management Sci. 32
paper. Support for this study was provided by the Israel Institute
1422-1433.
of Business Research; the Ministry of Defense, Israel; the Center for
Doty, H. D., W. H. Glick. 1994. Typologies as a unique form of the-
the Development of Technological Leadership, The University of
ory building: Toward improved understanding and modeling.
Minnesota; and by the Center for Technology Management
Acad. Management Rev. 19(2) 230-251.
Research at Stevens Institute of Technology.
Drazin, R., A. H. van de Ven. 1985. Alternative forms of fit in con-
tingency theory. Admin. Sci. Quart. 30 514-539.
Dvir, D., S. Lipovetsky, A. Shenhar, A. Tishler. 1998. In search of
References project classification: A non-universal approach to project suc-
Abernathy, W. J., J. M. Utterback. 1978. Patterns of industrial inno- cess factors. Res. Policy 27 915-935.
vation. Tech.Rev. (June-July) 40-47. Eisenhardt, K. M. 1989. Building theories from case study research.
Ahituv, N., S. Neumann. 1984. A flexible approach to information Acad. Management Rev. 14 532-550.
system development. MIS Quart. (June) 69-78. ., B. N. Tabrizi. 1995. Accelerating adaptive processes: Prod-
Alexander, C. 1964. Notes on the Synthesis of Form. Harvard Univer- uct innovation in the global computer industry. Admin. Sci.
sity Press, Cambridge, MA. Quart. 40 84-110.
Balachandra, R., J. H. Friar. 1997. Factors for success in R&D project Ettlie, J. E., W. P. Bridges, R. D. O'Keffe. 1984. Organizational strat-
and new product innovation: A contextual framework. IEEE egy and structural differences for radical vs. incremental inno-
Trans.on Engrg. Management 44 276-287. vation. Management Sci. 30 682-695.
Bart, C. 1998. Organizing for new product development. J. Bus. Fondahl, J. W. 1987. The history of modern project management.
Strategy 9 (July/Aug.) 34-39. ProjectManagementJ. 28(2) 33-36.
Blake, S. B. 1978. Managingfor Responsive Researchand Development. Freeman, C. 1982. The Economics of Industrial Innovaiton, 2nd ed.
Freeman and Co., San Francisco, CA. MIT Press, Cambridge, MA.

MANAGEMENT SCIENCE/Vol. 47, No. 3, March 2001 413


71
SHENHAR
One Size Does Not Fit All Projects

Galbraith, J. R. 1982. Designing the innovating organization. Organ. Perrow, C. C. 1967. A framework for the comparative analysis of
Dynam. 10 (Winter) 5-25. organizations. Amer. Soc. Rev. 32 194-208.
Glaser, B. G., A. L. Strauss. 1967. The Discovery of GroundedTheory: Pinto, J. K., J. G. Covin. 1989. Critical factors in project implemen-
Strategiesfor Qualitative Research.Aldine, Chicago, IL. tation: A comparison of construction and R&D projects. Tech-
Guterl, F. 1984. Design case history: Apple's Macintosh. IEEE Spec- novation 9 49-62.
trum 21 (December) 34-43. Rechtin, E. 1991. Systems Architecting. Prentice Hall, Englewood
Henderson, R. M., K. B. Clark. 1990. Architectural innovation: The Cliffs, NJ.
reconfiguration of existing product technologies and the failure Rosenbloom, R. S., M. Cusumano. 1987. Technological pioneering
of established firms. Admiin.Sci. Quart. 35 9-30. and competitive advantage: The birth of the VCR Industry.
Hoover, C. W., J. B. Jones. 1991. Report of the Committee on Engi- CaliforniaManagement Rev. 24(4) 51-73.
neering Design Theory and Methodology. Improving Engineer- Roussel, P. A., K. N. Saad, T.J.Erickson. 1991. Third GenerationR&D.
ing Design. National Research Council, New York. Harvard Business School Press, Boston, MA.
Iansiti, M. 1997. TechnologyIntegration:Making Critical Choices in a
Shanklin, W. L., J. K. Ryans, Jr. 1984. Marketing High Technology.
Dynamic World.Harvard Business Review Press, Boston, MA.
Lexington Books, New York.
Johnson, C. L. "Kelly", M. Smith. 1985. Kelly: More Than My Share
Shenhar, A. J. 1991. On system properties and systemhood. Internat.
of It All. Smithsonian, Washington, D.C.
J. General Systems 18(2) 167-174.
Kirk, J., M. L. Miller. 1986. Reliability and Validity in Qualitative
. 1993. From low to high-tech project management. R&D
Research.Sage University Series on Qualitative Research Meth-
Management 23(3) 199-214.
ods, Vol. 1. Sage, Beverly Hills, CA.
Lavold, G. D. 1988. Developing and using the work breakdown ., D. Dvir. 1996. Toward a typological theory of project man-
structure. D. I. Cleland, W. R. King, eds. Project Management agement. Res. Policy 25 607-632.
Handbook,2nd ed. Van Nostrand, New York. 302-323. Song, X. M., W. E. Souder, B. Dyer. 1997. A casual model of the
Lawrence, P. R., J. W. Lorsch. 1967. Organization and Environment: impact of skills, synergy, and design sensitivity on new prod-
Managing Differentiation and Integration. Harvard University, uct performance. J. Innovation Management 14 88-101.
Boston, MA. Souder, W. E., X. M. Song. 1997. Contingent product design and
Lawrence, R. M. 1987. Strategic Defense Initiative. Westview Press, marketing strategies influencing new product success and fail-
Boulder, CO. ure in U.S. and Japanese electronics firms. J. Product Innovation
Lemley, J. K. 1992. The chainel tunnel: Creating a modern wonder Management 14 21-34.
of the world. PmNetwork 6 (July) 14-22. Steele, L. W. 1975. Innovation in Big Business. Elsevier Publishing
Lewis, W., A. Samuel. 1989. Fundamentals of Engineering Design. Company, New York.
Prentice Hall, New York. Strauss, A. L. 1987. Qualitative Analysis for Social Scientists.
Little, A. D. 1981. The Strategic Management of Technology.European Cambridge University Press, Cambridge, U.K.
Management Forum, London, U.K. Thompson, J. D. 1967. Organizations in Action. McGraw-Hill,
Manne, S. J., L. Collins. 1990. Reconstructuring an aging infrastruc- New York.
ture. ProjectManagementJ. (April) 9-24. Tushman, M. L., P. Anderson. 1986. Technological discontinu-
Mansfield, E. 1968. Industrial Research and Technical Innovation. ities and organizational environments. Admin. Sci. Quart. 31
Norton, New York. 439-465.
Marples, D. L. 1961. The decisions of engineering design. IEEE Van Gigch, J. P. 1978. Applied GeneralSystems Theory,2nd ed. Harper
Trans. on Engrg. Management EM-8 (June) 55-71. and Row, New York.
Marquis, D. G. 1969. The anatomy of successful innovations. Villard, R. 1989. From idea to observation: The space telescope at
National Science Foundation, Washington, D.C. work. Astronomy (June) 38-44.
Miles, M. B., M. A. Huberman. 1984. Qualitative Data Analysis. Sage Weick, K. E. 1979. The Social Psychology of Organizing, 2nd ed.
Publishing, Beverly Hills, CA. Addison-Wesley Publishing, Reading, MA.
Moch, M., E. V. Morse. 1977. Size, centralization and organizational Wheelwright, S. C., K. B. Clark. 1992. Revolutionizing Product Devel-
adoptation of innovations. Amer. Soc. Rev. 42 716-725. opment. The Free Press, New York.
Pahl, G., W. Beitz. 1984. Engineering Design. Springer-Verlag, Yap, C. M., W. E. Souder. 1994. Factors influencing new prod-
New York. uct success and failure in small enterpreneurial high-technol-
Pearson, A. W. 1990. Innovation strategy. Technovation 10(3) ogy electronics firms. J. Product Innovation Management 11
185-192. 418-432.
Pellegrino, C. R., J. Stoff. 1985. Chariots for Apollo. Athenaeum, Yin, R. K. 1984. Case Study Research:Design and Methods. Sage Pub-
New York. lishing, Beverly Hills, CA.
Pennings, J. M. 1992. Structural contingency theory: A reappraisal. Zaltman, G., R. L. Duncan, J. Holbek. 1973. Innovations and Organi-
Res. in Organ. Behavior14 267-309. zations. Wiley, New York.

Accepted by Ralph Katz; receivedMay 1999. This paper was with the authors 9 monthsfor 2 revisions.

414 MANAGEMENTSCIENCE/Vol. 47, No. 3, March 2001

72
Project Management Quarterly. Vol. VIII, No.1, March 1977. Reprinted with permission

Organizational Alternatives for Project Managers


Robert Youker

There is no single, perfect organizational structure for managing projects and similar
temporary organizations. But you can -- and should--assess the feasibility of the
various alternatives.

In the past ten years interest has grown in techniques and approaches for management of temporary
projects (in contrast to ongoing operations). There has been an explosion of literature dealing with these
techniques and strategies, and more recently, we have seen the beginnings of organized academic
research on various aspects of project management.

However, in discussions and in the literature we still seem to be confused about the exact meaning of
some terms. This is particularly true in the area of alternative organizational approaches for the
management of projects.

Functional organizations

The most prevalent organizational structure in the world today is the basic hierarchical structure (Figure
1). This is the standard pyramid with top management at the top of the chart and middle and lower
management spreading out down the pyramid. The organization is usually broken down into different
functional units, such as engineering, research, accounting, and administration.

Figure 1. Functional Organization

The hierarchical structure was originally based on such management theories as specialization, line and
staff relations, authority and responsibility, and span of control. According to the doctrine of
specialization, the major functional subunits are staffed by such disciplines as engineering and
accounting. It is considered easier to manage specialists if they are grouped together and if the
department head has training and experience in that particular discipline.

The strength of the functional organization is in its centralization of similar resources. For example, the
engineering department provides a secure and comfortable organizational arrangement with well-defined
career paths for a young engineer. Mutual support is provided by physical proximity.

1
73
Project Management Quarterly. Vol. VIII, No.1, March 1977. Reprinted with permission

The functional organization also has a number of weaknesses. When it is involved in multiple projects,
conflicts invariably arise over the relative priorities of these projects in the competition for resources.
Also, the functional department based on a technical specialty often places more emphasis on its own
specialty than on the goals of the project. Lack of motivation and inertia are other problems.

However, many companies use the functional organization for their project work as well as their standard
operations. The world is a complicated place. In addition to discipline and function, other nuclei for
organizational structures include products, technologies, customers, and geographic location.

Project organizations

The opposite of the hierarchical, functional organization is the single-purpose project or vertical
organization. In a projectized organization, all the resources necessary to attain a specific objective are
separated from the regular functional structure and set up as a self-contained unit headed by a project
manager. The project manager is given considerable authority over the project and may acquire
resources from either inside or outside the overall organization. All personnel on the project are under
the direct authority of the project manager for the duration of the project.

In effect, a large organization sets up a smaller, temporary, special-purpose structure with a specific
objective. It is interesting to note that the internal structure of the project organization is usually
functional, that is, that the project team is divided into various functional areas (Figure 2).

Figure 2. Projectized Organization

Note that the term here is project organization, not project management. You can manage projects
with all three types of organizational structure. The advantages of the project organization come from the
singleness of purpose and the unity of command. An esprit de corps is developed through the clear
understanding of, and focus on, the single objective. Informal communication is effective in a close-knit
team, and the project manager has all the necessary resources under his direct control.

The project organization, however, is not a perfect solution to all project management problems, as some
have suggested. Setting up a new, highly visible temporary structure upsets the regular organization.
Facilities are duplicated and resources are used inefficiently. Another serious problem is the question of
job security upon termination of the temporary project. Personnel often lose their home in the functional
structure while they are off working on a project.

2
74
Project Management Quarterly. Vol. VIII, No.1, March 1977. Reprinted with permission

The functional, hierarchical organization is organized around technical inputs, such as engineering and
marketing. The project organization is a single-purpose structure organized around project outputs, such
as a new dam or a new product. Both of these are unidimensional structures in a multidimensional world.
The problem in each is to get a proper balance between the long-term objective of functional
departments in building technical expertise and the short-term objectives of the project.

Matrix organizations

The matrix organization is a multidimensional structure that tries to maximize the strengths and minimize
the weaknesses of both the project and the functional structures. It combines the standard vertical
hierarchical structure with a superimposed lateral or horizontal structure of a project coordinator (Figure
3).

Figure 3. The Matrix Organization

The major benefits of the matrix organization are the balancing of objectives, the coordination across
functional department lines, and the visibility of the project objectives through the project coordinators
office. The major disadvantage is that the man in the middle is working for two bosses. Vertically, he
reports to his functional department head. Horizontally, he reports to the project coordinator or project
manager. In a conflict situation he can be caught in the middle.

The project manager often feels that he has little authority with regard to the functional departments. On
the other hand, the functional department head often feels that the project coordinator is interfering in his
territory.

The solution to this problem is to define the roles, responsibility, and authority of each of the actors
clearly in a project charter. The project coordinator specifies what is to be done and the functional
department is responsible for how it is done (Figure 4).

3
75
Project Management Quarterly. Vol. VIII, No.1, March 1977. Reprinted with permission

Figure 4. Matrix Organization Relationship of Project Management to Functional Management

Criteria for selecting an organizational structure In the field of management, zealots like to say that their
particular model is best. Neophytes want a simple and unambiguous answer. Experienced and thoughtful
observers, however, know that no one particular approach is perfect for all situations. The current vogue
in management literature is the contingency model. This theory states that the best solution is contingent
upon the key factors in the environment in which the solution will have to operate.

The same is true for the choice of an organizational structure. What we need, then, is a list of key factors
that will help us to choose the right organizational structure for the given conditions on a specific project
with a given organization and a particular environment. A set of such factors is listed in Figure 5.

Figure 5. Criteria for Organization Design Decisions

4
76
Project Management Quarterly. Vol. VIII, No.1, March 1977. Reprinted with permission

For example, an organization developing many new but small projects with standard technology would
most likely find a functional structure best. On the other hand, a company with a long, large, complex,
and important project should favor the project organizational structure. A firm in the pharmaceutical
business with many complicated technologies would probably go to a matrix structure.

It is possible to use all three structures in the same company on different projects. All three structures
might also be used on the same project at different levels--for example, an overall matrix structure for
the project with a functional substructure in engineering and a project organization in another functional
sub-area.

Before we can make a final choice, however, we must consider the following additional factors:

1. What is the relationship between organizational design, the skills of the project manager, and the
project planning and reporting system?

2. Are there ways we can improve coordination and commitment in the functional structure without
moving to a project or matrix structure?

3. What variations exist in the matrix structure and what are the advantages of each variation?

Project managers and organizational design

It is not possible to decide on the organizational design without also deciding whom to select as the
project manager and what kind of design you want for the planning and reporting systems. These
decisions are closely interrelated. For example, a successful project organization requires a project
manager with the broad skills of a general manager. He must combine technical knowledge of the
subject matter with management abilities before he can lead the entire project team. It makes no sense
to select a project organization form if such a project manager is not available.

The planning and reporting system in a project organization can be fairly simple because the team is in
close proximity. The opposite is true in the management of projects through a functional organization.
Information in the form of plans, schedules, budgets, and reports is the key medium for integrating a
functional organization. Therefore, a more sophisticated planning and reporting system is required in a
functional organization than in a project organization.

Improving lateral communications in the functional structure Organizations typically turn to a project
organization or a matrix organization because the normal functional structure has failed on a series of
projects. It is not necessary, however, to throw the baby out with the bath water. Before giving up on the
functional organization, analyze the real problems and see if steps can be taken short of reorganization.
Some results of a reorganization may be favorable, but other unintended but logical consequences are
certain to be unfavorable.

Methods of lateral or horizontal communication need to be developed across functional department


boundaries. Alternative approaches for lateral communication include:

1. Such procedures as plans, budgets, schedules, and review meetings.

2. Direct contact between managers.

3. Informal liaison roles.

4. Teams.

These are integrating mechanisms short of the establishment of a matrix organization. They help to
break down the barriers that seem to separate different disciplines, departments, and geographic
locations.

Weak to strong matrix--a continuum

5
77
Project Management Quarterly. Vol. VIII, No.1, March 1977. Reprinted with permission

The three major organizational forms--functional, matrix, and project--may be presented as a continuum
ranging from functional at one end to project at the other end (Figure 6). The matrix form falls in between
and includes a wide variety of structures, from a weak matrix near functional to a strong matrix near
project. The continuum in Figure 6 is based on the percentage of personnel who work in their own
functional department versus the percentage of personnel who are full-time members of the project
team. Note that in a functional organization the project team has no personnel of its own. The dividing
line between functional and matrix is the point at which an individual is appointed with part-time
responsibility for coordination across functional department lines.

The bottom line of Figure 6 shows that a weak matrix has a part-time coordinator. The matrix gets
stronger as you move from full-time coordinator to full-time project manager and finally to a project office
that includes such personnel as systems engineers, cost analysts, and schedule analysts. The difference
between a coordinator and a manager is the difference between mere integration and actual decision-
making.

Figure 6. Organizational Continuum

On the far right we have the project organization. Ordinarily, there is a clear distinction between a strong
matrix in which most of the work is still being performed in the functional departments and a project
organization in which the majority of the personnel are on the project team.

It is rare for a project organization to have all the personnel on its team. Usually some functions, such as
accounting or maintenance, would still be performed by the functional structure.

Some persons have taken issue with the use of the term strong matrix. They say that a strong matrix
comes from an even balance of power between the functional departments and the project office. That
may be true in some instances, but not always. Strong and weak are not used in the sense of good and
bad. Rather, they refer to the relative size and power of the integrative function in the matrix.

Measuring authority: Functional vs. project staff

Another way to differentiate between a strong matrix and a weak matrix is to analyze the relative degree
of power between the functional departments and the project staff. We could construct another
continuum with function on the left and project on the right. For a given project we would decide where
the power rests on the continuum for decisions over project objectives, budgets, cost control, quality,
time schedule, resources, personnel selection, and liaison with top management. On any given project
the power will be strongly functional for some factors and strongly project for others. However, a profile
line can be drawn from top to bottom that would indicate whether the trend is to the left (weak) or to the
right (strong).

Making matrix management work Matrix management is a controversial concept. Some people have had
bad experiences operating in a matrix. Others have had a great deal of success. It does require careful
definition of authority and responsibility as well as strenuous efforts toward coordination and diplomacy.

6
78
Project Management Quarterly. Vol. VIII, No.1, March 1977. Reprinted with permission

The matrix is basically a balance of power between the goals of the functional structure and of a specific
project.

7
79
Project Management Quarterly. Vol. VIII, No.1, March 1977. Reprinted with permission

Overloaded functional departments

One key problem with matrix organizations is that they tend to overload the functional departments with
work. If a functional department makes a commitment to work more man-hours on projects than it has
available, conflicts over priorities between projects are inevitable. This problem can be alleviated, if not
solved, by better planning.

A matrix organization will not work effectively unless a matrix strategic plan setting priorities on
objectives and a matrix budget allocating resources also exist. For example, in Figure 7, the project
manager for Project A will add horizontally across functional departments to get his total budget. In a
similar manner, the vice-president of manufacturing must add vertically across all the projects for which
he has committed funds and resources as well as his strictly departmental efforts. The matrix budget
must add up to 100 percent in both directions. The usual picture is that the functional departments are
overcommitted and show required man-hours of perhaps 120 percent of actual man-hours available.
When this happens, politics and disappointment become inevitable.

Figure 7. Matrix Organization and Matrix Budget

The golden rule in matrix management states, He who has the gold makes the rules. If a project
manager does not control the budget, he can only beg for handouts from the functional departments. A
matrix budget assigns resources to the project manager for purchases from the functional departments.
Making up such a budget takes careful work during long-range and annual planning. Regular updating of
the matrix plan and budget are also necessary.

Survival techniques in the matrix

A common picture of the project coordinator in a matrix organization is of a frustrated diplomat


struggling to cajole the functional departments into performing the work on schedule and within budget.
His position is difficult, but the following approaches can help:

1. It is important to have a charter from top management defining responsibilities and authority for the
project manager as well as the role of the functional departments.

2. The project coordinator or manager must anticipate conflicts in the matrix. Conflict is inevitable with
dual authority, but it can be constructively channeled.

3. Since conflict is inevitable, it is important to take positive steps to develop teamwork. Regular lunches

8
80
Project Management Quarterly. Vol. VIII, No.1, March 1977. Reprinted with permission

or social gatherings help to foster a team spirit. In recent years the behavioral sciences have developed
a number of specific techniques for alleviating or using conflict effectively. Training programs for matrix
managers should include experiences with such techniques.

4. The project coordinators main power comes from the approved objectives, plans, and budgets for the
project. Use these documents to hold departments to their commitments.

5. It is vital that the functional department heads be committed to the plans and schedules for the project
as well as the lower-level task leaders. Functional managers should review and sign off on these
documents.

6. It is usually best to avoid direct conflict with the functional department heads. The matrix manager
should use his boss when a situation threatens to get out of hand.

7. It is important to remember that the project coordinator is concerned with what is to be done, not
how. Use a management-by-objectives approach and do not supervise the functional departments
(Figure 4) too closely.

8. Many of the problems of matrix management flow from the uncertainty inherent in the project
environment. By definition, a project is, to some extent, a new effort. Careful and continuous planning
can help reduce uncertainty.

No one perfect organizational structure for managing projects exists. The functional, the project, and
the different matrix structures all have strengths and weaknesses. The final choice should come after
weighing various factors in the nature of the task, the needs of the organization, and the environment of
the project.

The functional structure will work for many projects in many organizations, especially if lateral
communications can be improved through integrating mechanisms and procedures short of hiring a
matrix coordinator.

When a matrix approach is chosen, the entire organization must put a good deal of effort into it to make
it work. In particular, the project coordinator or project manager in the matrix must be carefully chosen
and trained. His interpersonal skills are more important than his technical knowledge.

In many situations, a project organization may appear to be the simplest solution from the viewpoint of
the project manager. However, the functional managers or top management may not find it to be the
best long-range or most strategic decision.

Robert Youker

9
81
International Journal of Project Management 21 (2003) 18
www.elsevier.com/locate/ijproman

On the nature of the project as a temporary organization


J. Rodney Turnera,*, Ralf Mullerb
a
Department of Marketing and Organization, Faculty of Economics, Room H15-3, Burgemeester Oudlaan 50, Erasmus University Rotterdam,
3062 PA Rotterdam, The Netherlands
b
Henley Management College and NCR Teradata, Global Program Management Oce, Sjobogatan 10, 212 28 Malmo, Sweden

Received 5 September 2001; accepted 6 March 2002

Abstract
The nature of the project as a temporary organization is analysed from the perspective of organizational theory. This leads to a
reassessment of the denition of a project. It is suggested that classical denitions of projects are not wrong, just incomplete. The
project as a temporary organization is viewed here as a production function, as an agency for assigning resources to the manage-
ment of change within the functional organization, and as an agency for managing uncertainty. The role of the project manager is
also considered. The project manager is chief executive of the temporary organization, and thus their roles in objective setting and
motivating team members are emphasized over their role in planning and executing work. Second, as manager of the agency, they
are the agent of the owner (principal) and so a second hierarchy of management and control must be put in place to monitor their
performance. These agency costs add to the cost of the project, but may also explain why professional recognition is so important
to project managers. # 2002 Elsevier Science Ltd and IPMA. All rights reserved.
Keywords: Agency; Principal-agent; Resouce; Change; Uncertainty; Executive management; Professionalism

1. Introduction 1. It is unique: no project before or after will be


exactly the same.
It is 12 years since Turner [1] wrote a paper entitled 2. It is undertaken using novel processes: no project
What are projects and project management, which before or after will use exactly the same
has been taken as his denitive statement on the subject, approach.
and formed the basis of the denitions used in his books 3. It is transient: it has a beginning and an end.
[2]. A project was dened as:
These features create three pressures:
An endeavour in which human, material and
nancial resources are organized in a novel way, to 1. Projects are subject to uncertainty: we cannot be
undertake a unique scope of work, of given speci- certain that our plans will deliver the required
cation, within constraints of cost and time, so as project outcomes or desired benecial change.
to achieve benecial change dened by quantitative 2. They create a need for integration: of the resour-
and qualitative objectives. ces to do the project, between dierent parts of
the project, and of the project into the business.
The second edition [3] retained the denition, but also 3. They are undertaken subject to urgency: of deli-
concentrated more on features of projects, indicating a vering the desired outcomes within the desired
range of features shared by projects, Table 1. A project timescales.
is undertaken to deliver benecial change, and thus has
three essential features: Turner [3] suggested it is these three pressures that are
special to project management, not the management of
time, cost and quality, which is shared with routine
* Corresponding author. Present address: Wildwood, Manor Close,
operations management. He further suggested that an
East Horseley, Surrey KT24 6SA, UK. Tel.: +44-1483-282-344; fax:
+44-1483-284-884. endeavour that had many of those features would be
E-mail addresses: rodneyturner@europrojex.com, turner@few. better managed as a project, but that one that had only
eur.nl (J.R. Turner), ralf.mueller@ncr.com (R. Muller). a few would be better managed as a routine operation.
0263-7863/02/$22.00 # 2002 Elsevier Science Ltd and IPMA. All rights reserved.
PII: S0263-7863(02)00020-0
82
2 J.R. Turner, R. Muller / International Journal of Project Management 21 (2003) 18

Table 1 address the nature of project management, standing by


Features of projects the earlier denition that project management is the
Aim Features Pressures Processes process by which projects are successfully delivered, and
their objectives successfully achieved (whatever is meant
To deliver Unique Uncertainty Flexible by successfully [13]). However, we do address the role
Benecial Novel Integration Goal Oriented
Change Transient Transience Staged
of the project manager, considering:

 the project manager as chief executive of the


For endeavours that have some of the features, you may temporary organization and
adopt some elements of both project management and  as the agent of the principal
routine operations management. Table 1 shows that the
project management processes need to be exible, goal
oriented and staged, in direct contrast to (truly) routine 2. The project as a production function
operations management, where the processes need to be
stable, activity oriented and continuous. Many of the classical denitions of projects emphasize
However, applying the concepts of organization the- the role of a project as a production function, just as the
ory [4] to the project as a temporary organization has earliest denitions of the rm in classical economics
lead us to question the denition; not to suggest it is [5,7]. For instance, the earlier denition starts by say-
wrong, just incomplete. It is adequate for some pur- ing the project is an endeavour, and some of the
poses; but does not dene fully the roles, functions and classical denitions, many given in Turner [2,3], do
limits of projects. It has not been possible for classical likewise:
economics to produce a denitive theory of the rm [5].
For instance, the rm as a production function or prin- Something which has a beginning and an end, [8]
cipal-agent theory cannot answer the make-or-buy
question; should a rm make all its intermediate pro- A human endeavour which creates change, is limited
ducts in the hierarchy, or buy them in the market. in time and scope, has mixed goals and objectives,
Either would predict that the entire capitalist system involves a variety of resources and is unique, [9]
should be organized as a single rm, or that the rm
should procure all its intermediate products in the mar- A complex eort to achieve a specic objective
ket, acting only as a design and marketing company within a schedule and budget target, which typically
(and manager of the supply chain), as with Benetton. cuts across organizational lines, is unique and is
Most classical denitions of projects (including the usually not repetitive within the organization, [10]
aforementioned) do not answer similar questions about
projects. Should a rm organize all its change activity as A one time unique endeavour to do something that
one large project or programme, or should it break it has not been done that way before, [11]
into a myriad of tiny projects? (According to the earlier
denition, the making of a cup of tea can be considered A temporary endeavour undertaken to create a
to be a project, which is not useful for most purposes.) unique product or service, [12].
Turner and Keegan [6] addressed the question of whe-
ther organizations should manage all their work as Even in Barness denition, one tends to interpret the
projects, showing that fundamentally routine work something as an endeavour, activity or task, rather
should be managed as such since there are higher than an organization or agency as we shall see shortly.
transaction costs associated with managing work as What follows, paraphrases Hart [5], but refers to
projects. But what is the limit on the size of a project as projects rather than rms. According to these deni-
a temporary organization? tions, the project is a collection of plans, presided over
In this paper, we re-address the nature of projects as by a manager, who buys and sells the projects inputs
temporary organizations. We consider: and outputs on the open market, and tries to maximize
the benet to the owner. The benet is the net present
 the project as a production function; value of the project, discounted for risk. This is a car-
 as a temporary organization; icature of the project. It is popular because:
 as an agency for change;
 as an agency for resource utilization; and  it lends itself to easy mathematical formulation,
 as an agency for uncertainty management (critical path analysis, earned value analysis,
etc.);
and show how some of the classic denitions of pro-  it is useful for examining how the project plan
jects do or do not address these issues. We do not responds to risks; and

83
J.R. Turner, R. Muller / International Journal of Project Management 21 (2003) 18 3

 it is useful for analysing the interaction between cracy [14]. It will have its proponents, who will try to
the project and other projects and routine oper- create structures to ensure its success, while its oppo-
ations. nents will try to undermine it. The principal will need to
appoint a manager (the agent) to manage the project on
However, this perspective has weaknesses. It does not their behalf. The principal will need to create structures,
explain: including information channels, to monitor the man-
agers decisions to ensure they are aligned with the
 how resources are assigned to the project; owners objectives of prot maximization. Turner and
 how conicts between stakeholders are resolved; Keegan [6] have suggested that this is one of the roles of
 or even how benet is maximized; the broker and the steward, and of conguration man-
 and whether it is better to undertake two projects agement.
as one, two or four projects. Viewing the project as a temporary organization
introduces many of the elements of project manage-
The theory tells us nothing of the structure of pro- ment, including:
jects. That does not mean the denitions above are
wrong. They are just not enough.  the conict of interest between the various sta-
keholders;
 the role of the manager (agent), and of the bro-
3. The project as a temporary organization ker and steward;
 the need to put in place information and
The rst layer of texture we can add is to view the communication systems to monitor delivery of
project as a temporary organization, which leads us to the project, to monitor achievement of the own-
the equivalent of principle/agent theory for the rm. ers objectives, and to avoid self-interest and
The concept of projects as temporary organizations is opportunism by the projects participants, espe-
fairly recent. Cleland and Kerzner [13] dene a project cially the agent.
as:
However, it still does not tell us whether two projects
A combination of human and non-human resour- should be managed as one, two or four. To be able to
ces pulled together into a temporary organization reach that conclusion we need to look on the project as
to achieve a specied purpose. an agency for change and as an agency for resource
utilization.
This overtly denes the project both as a temporary
organization, and as a production function and an
agency for assigning resources, (see later). Most of the 4. The project as an agency for change
other denitions refer only obliquely to projects being
temporary organizations. Turners [1] denition says The denition of projects by Andersen et al. [9] and
directly that the project is a vehicle (or agency) for Turner [1] emphasize that projects deliver change.
organizing resources (see later) and therefore implies Traditional organizations adopt projects as a vehicle (or
that it is an organization. The denition also implies agency) for change. They create the temporary organi-
quite overtly that the endeavour is unique, novel and zation to deliver a coherent set of change objectives,
transient, which are the three main features referred to because projects are better suited for managing change
in the second edition [1]. Barness denition says that than the functional organization. There are several rea-
the project has a beginning and end, but as I said earlier, sons for this:
the something sounds more like a production func-
tion than an organization. The others refer more or less 1. Functional organizations have high inertia to
directly to the projects being temporary, but most change [15]. Projects can provide an impetus to
emphasize the production function rather than the overcome the inertia, they can be set up sepa-
organization. Others, such as Cleland and King [10], rately from the functional organization so the
also refer to the role of the project as an agency for small, temporary organization has little or no
assigning resources. inertia, allowing the change to build up momen-
As a temporary organization, the project is an agency tum, and they can be used to prototype the
established by a parent organization (the principal) to change. Briner et al. [16] describe such a change
achieve specic objectives. On the time scale on which project with Norwich Union. The establishment
the project exists, the parent organization can be taken of the Sheeld Oce at rst sight appears
to be stable, its structures given. Thus, a project has unsuccessful since the oce closed soon after
much in common with an agency in a political bureau- being opened. However, it did enable the

84
4 J.R. Turner, R. Muller / International Journal of Project Management 21 (2003) 18

company to initiate and prototype a successful change, and if well managed they can be an ecient way
change process that ultimately enabled the com- of achieving that change. Unfortunately, in North
pany to be demutalized. America, project management is becoming the latest
2. Projects being small temporary organizations are management fad, like Total Quality Management or
more exible and better able to respond to Business Process Reengineering previously [19]. When
uncertainties in the change process and change people nd it is not the universal panacea, to cure all
objectives, as we shall see later. their ills, they may reject it in spite of its obvious
3. The functional organization is designed for the advantages.
management of the routine, and so is not suited We can answer the question about the limit of the
for managing change [3]. project: a project is dened by a coherent set of change
objectives to which it is sensible to assign a set of
resources for its management. If one set of resources is
This view of the project as an agency for change responsible for several unrelated sets of change objec-
begins to dene the boundaries of a project. It is dened tives, it could lead to their ineective or inecient utili-
by a coherent set of change objectives. zation. The role of the project as an agency for resource
It also supports the usual distinction between a pro- allocation also implies something about the structure of
gramme and a project [17]. A project is a temporary the project and the conicting goals of principle and
organization designed to deliver a specic set of change agent. While the agents goal is eciency in project
objectives. A programme is a framework to provide delivery through employment of the best and most
strategic direction to a group of projects so that they eective resources, the principals objective is the
can combine to provide higher order strategic or devel- balanced utilization of the entire sta and with it the
opmental change for the organization. If a project is a simultaneous allocation of eective and less-eective
temporary rm, a programme is a temporary group of resources to projects. While the most sought after
rms (which ceases to exist when the last project ceases resources are automatically booked-out, the principals
to exist). role becomes one of selling the less eective resources
to the agents. This conict is widened through the
imbalance in detailed knowledge about the project. It
5. The project as an agency for resource utilization puts the agent at an advantage, weakening the princi-
pals authority over project related decisions. Turner
Several denitions emphasize the role of the project as and Keegan [6] have suggested the sourcing of resources
a vehicle (or agency) for assigning (or organizing) should not be the role of the project manager, but of the
resources for completion of the endeavour or task. broker and steward.
Turners [1] denition includes this as a central element, From a principals perspective the boundaries of a
and Cleland and Kings [10] and Cleland and Kerzners project are less clear than from an agents (project
[13] denitions says the project is a focus for assigning managers). Organizations adapt their operational pro-
resources from across the rm. cesses to the needs of their projects to economize on
Carroll [15] says the success of dierent organiza- similarities in the support needs of projects. These
tional forms is dependent on their ability to attract background processes are aligned for the achievement
resources. Projects have been used as an organizational of project goals of the temporary organization while
form to provide a vehicle for assigning resources to the being mandated to achieve the permanent organiza-
delivery of change in organizations since the 1950s [18], tions objectives. Within this conict, the limit of a pro-
and this can be taken as a measure of its success. Clearly jects impact on these changes is not dened. Project
as temporary organizations, projects need to negotiate management maturity models, like the Organizational
afresh for resources as each is started. However, as we Project Management Maturity Model from the Project
saw in the last section, functional organizations have Management Institute [20], try to provide guidelines for
high inertia to change, and so projects do seem to be the optimization of processes at the interface between
perceived as an eective way of assigning resources to permanent and temporary organization. To that end the
change in organizations. boundary of a project is not as clear as from an agents
Carroll [15] also says that the only measure of an perspective.
organizations eciency is its longevity. Clearly this is Research on organizational support [21] showed
wrong in the case of projects which are established as extension of project requirements beyond the project
temporary organizations, although its success as an team and into the functional organization is lacking,
organizational form of rst choice for assigning resour- and mechanisms for sharing and resolving resource
ces to change is long lived. They can be successful at issues are seldom in place. The principals performance
attracting resources (although not always), they can be evaluation is seldom tight to their degree of support of
a way for organizations to set aside resources to achieve cross-functional project teams.

85
J.R. Turner, R. Muller / International Journal of Project Management 21 (2003) 18 5

To reinforce this point, it is common to dene a Table 1 suggests that there are three key pressures in
portfolio of projects as a set of projects which share a projects, not met in the functional organization: the
common resource pool, which are managed together for uncertainty, the need for integration and the urgency.
the ecient utilization of the resource pool. The port- Treating urgency as a risk, we see that these three pres-
folio itself is not dened as a single project. Reasons for sures have now been addressed. The agency itself man-
this include: ages the need for integration, both of the resources that
are assigned to the agency, and of the agency with the
1. The portfolio itself is more permanent in nature. parent organization and its context. And the agency
The projects come and go, but the portfolio itself is a vehicle for managing the uncertainty and
itself can exist as a permanent organization. urgency.
Each project is assigned to the portfolio, and
then resources assigned to the project from the
resource pool. Note that this contradicts the 7. The project manager as chief executive of the temporary
denition of Cleland and King [10], since in this organization
case the project does not cut across organi-
zational lines. As we said at the start, it is not our intention to rede-
2. Each project in itself delivers a coherent set of ne project management. Turner [1] dened project
change objectives. The portfolio is a permanent management as the process by which a project is suc-
organization created for the purpose of ecient cessfully completed, and its objectives successfully
utilization of resources across the projects that it delivered (whatever is meant by successfully). How-
comprises. The projects are temporary organiza- ever, we do wish to consider two issues relating to the
tions created for the eective delivery of the project manager arising from the earlier discussion.
change objectives. The rst is that if the project is a temporary organi-
zation, the project manager is chief executive of the
Thus we see that unlike the rm in classical econom- temporary organization. Scott [23], quoting Barnard
ics, it is quite easy to delineate a project, and clearly [24], says that the executives role is not mainly one of
dierentiate it from a programme and a portfolio of management, administration and of supervision, but:
projects.
to formulate purposes, objectives and ends of the
organization. . .This function of formulating grand
6. The project as an agency for uncertainty management purposes and providing for their redenition is one
which needs systems of communication, experience,
Almost none of the classic denitions of projects imagination, interpretation and delegation of
address their riskiness directly, although when discuss- responsibility.
ing the features of projects, Turner [3] says that uncer-
tainty of the product and process is one of the key This view of the project manager is perhaps at odds
consequences of the features. Turner and Cochrane [22] with many peoples view of them as a non commissioned
developed a taxonomy for projects based on uncertainty ocer, planning plans using their PC based software,
of product and process. and distributing those plans. The view of the project
Turner and Keegan [6] suggest that the need to man- manager as chief executive sees their role as one of for-
age conguration and the reduction of uncertainty is the mulating objectives and strategy for the project, and
main (additional) transaction cost associated with pro- through the purpose of the project, linking those objec-
jects. Thus the project as an agency for uncertainty tives and strategy to the objectives and strategy of the
management implies something about both the scope parent organization [25]. The project manager should
and structure of the project. Turner and Keegan [6] learn to delegate the planning and reporting, and most of
proposed the roles of broker and steward to act as the work. The managers role is further to interpret the
communication channels between the client and project plans and progress reports, to interpret them and rede-
manager (principal and agent) during delivery of the ne them to achieve the projects objectives.
project. Further, where uncertainty management of one Scott further says the executive has cognitive and
element of work has no impact on that of another, there cathectic roles. The cognitive roles emphasize the
is no benet from an uncertainty perspective to include rational ways in which the executive tries to guide and
them in the same project. Where such impacts exist it limit the choices of subordinates; the executive delegates
may well help economize on transaction costs to include to subordinates, but then seeks to guide, direct and
them in the same project, although on occasion such constrain their options. This is the traditional view of
impacts may be handled at the programme level, or at project management of rational planning, but one of the
the broker/steward interface. guider rather than the doer. The cathectic roles are more

86
6 J.R. Turner, R. Muller / International Journal of Project Management 21 (2003) 18

non-rational, the motivational and emotional aspects of owner organization establishes the project, the tempor-
goal setting. Project team members must be motivated ary organization, to achieve certain ends. The principal
to act, and they must develop faith in and commitment (owner) then needs to appoint an agent (chief executive)
to a larger moral purpose. Several of the authors in to manage the work of the temporary organization to
Williamsons book, including Williamson himself, achieve those ends. First the principal needs to recruit a
emphasize this moralistic role to avoid sub-goal pursuit project manager of appropriate competence.
and opportunism. Covey [26] emphasizes the cathectic But once he or she is appointed, the principal needs to
roles in his pyramid of inuence, saying it is necessary impose a second level of hierarchy and control above
to build belief in the executives purpose and to build the project to control the activities of the agent. Even
relationships, before trying to overtly inuence sub- though the project was created by the principal, the
ordinates through rational argument. agent being more closely involved knows more about it,
There is growing evidence that competence in the tra- and so there is a power shift from principal to agent.
ditional areas of the project management body of The principal tries to win back power through this sec-
knowledge [12,27] are essential entry tickets to the game ond hierarchy. Turner and Keegan [6] have described
of project management, but they do not lead to superior this second hierarchy of control in project-based orga-
performance [28,29]. They are hygiene factors, necessary nization, and identied the roles of the broker and
conditions for project management performance, but steward as two key roles in this structure. However,
they are not competitive factors for which improved even still the principal faces two classic problems in
competence leads to superior project performance. The their relationship with the agent. The rst is the princi-
two elements from PMIs Body of Knowledge which pal knows less about the project than the agent and so
lead to superior project performance are Managing will nd it dicult to question their decisions. This is
Communication and Managing Risk [30,31]. Other the adverse selection problem [14]. Second, the agent
competitive factors are those described earlier by Scott. has their own objectives to pursue. They may have a
Thus viewing the role of the project manager as chief dierent idea about what is right for the project, or may
executive of the temporary organization gives us per- even have objectives that are not aligned with the prin-
haps a dierent view of their role, as the purely rational cipal. This is the moral hazard problem [14]. Thus, the
one many traditional views of project management principal needs to put in place incentives to align the
encourage. This perspective identies commonalities in agents objectives with their own, and control mechan-
the role of the chief executive and the project manager, isms to stop opportunism. Communication between the
but it also shows apparent dierences in existing control principals organization and the agency must be kept
instruments and measures for success. Project managers redundant through multiple channels and boundary
are measured and remunerated on a very limited set of spanning individuals [32,33] to ensure a holistic view of
success criteria, such as time, scope, budget, and con- the project on the side of the principal.
trolled by a small number of people, like the principle or These agency costs increases the cost of a project as
broker. The chief executive has to cope with a variety of an organizational form compared to the functional
(often conicting) goals and measures and is controlled hierarchy. It might be said they also exist in the func-
by a board of directors and the community of stake- tional hierarchy. The bureaucratic costs of the func-
holders. This leads to a more balanced management tional hierarchy do make it more expensive than the
towards the sum of objectives, including the interface to market. But it is easier for the managers behaviour to
the organizations environment. Accepting the project be monitored in a routine environment, and if they
manager as a chief executive of a project, with a similar indulge in opportunistic behaviour they can be dis-
set of measurements for success, identies the project missed. The managers desire to keep their job, and their
managers additional responsibilities for managing the nancial incentives in the organization reduces their
interface to neighbouring projects, the community of rewards from opportunism.
users and the contribution to the clients overall objec- Because the project is a temporary organization, the
tives of the project. It questions the current, constrained rewards of opportunistic behaviour are greater, and the
and inward looking denition of the role of project incentive not to indulge less (the project might be n-
manager. ished by the time the manager is found out). Thus there
is a need for greater control and so the agency costs are
higher. I am not suggesting that project managers are
8. The project manager as the agent of the principal less trustworthy than other managers; it is just less easy
to be found out before the project is nished, and so the
A complimentary view sees the project manager as the principal will require greater monitoring.
agent of the principal. In the section headings we refer- The role of project manager as agent may be why
red to the project as an agency for the achievement of professionalism is so important to project managers.
certain purposes, not as a vehicle, to emphasize that the When the principal appoints an accountant, lawyer or

87
J.R. Turner, R. Muller / International Journal of Project Management 21 (2003) 18 7

engineer, their membership of a profession is an indi- We would also propose a denition of a programme
cator of both of peer-reviewed competence, and of of projects as:
trustworthiness, and so can ease the appointment pro-
cess. The desire to maintain professional status out- A programme of projects is a temporary organiza-
weighs the desire for short-term gain arising from tion in which a group of projects are managed
opportunism, and so increases the trustworthiness of together to deliver higher order strategic objectives
the professional. Professional membership will reduce not delivered by any of the projects on their own.
the agency costs associated with a given project man-
ager compared to one who is not. However, we realize that the denition of a project
Turner and Simister [34] have described how dierent given earlier applies equally well to programmes as
contract forms can be used to incentivize project man- described by Murray-Webster and Thiry [17]. The dif-
agers to share the owners (principals) objectives from ference then between a project and a programme is that
the project. the objectives of a programme are then less specic and
longer term. The objectives of a project are SMART
(specic, measurable, achievable, realistic and time-
9. Conclusion lined), whereas those of a programme are less SMART
(less specic, less accurately measurable, and less accu-
We have considered the perspectives organization rately timelined, but still achievable and realistic).
theory brings on the project as a temporary organiza- This view of the project as a temporary organization
tion, and have found it easier to develop a theory of highlights the role of the project manager as chief
the project than it is to develop a theory of the rm. If executive of the temporary organization, and additional
the project is viewed as a temporary production func- competences that implies over and above those laid out
tion, for assigning resources to the implementation of in the project management bodies of knowledge. In this
change objectives, it delineates each project, and sug- role, the project manager has cathectic as well as cogni-
gests why it should be of a certain size. Further it dif- tive roles. He or she is responsible for dening the pur-
ferentiates projects from programmes and portfolios of pose and objectives of the project, motivating the
projects. project team to achieve them, and manage the project
We would now propose a denition of a project. relative to internal and external objectives. Planning,
We opened by saying that we think that the former execution and control of the work should be delegated,
denitions are not wrong, just incomplete. However, but the project manager should limit the options of the
we might now propose this revised denition of a team to ensure objectives are delivered as required by
project: the owner (principal), and to interpret progress to dene
and reformulate the work as required.
A project is a temporary organization to which The role of the project manager as agent of the prin-
resources are assigned to undertake a unique, novel cipal points to the need for the owner as principal to put
and transient endeavour managing the inherent in place a higher tier of hierarchy and control above the
uncertainty and need for integration in order to project to monitor the performance of the project man-
deliver benecial objectives of change. ager as agent. This is to ensure that the project managers
objectives are aligned with the owners (the moral hazard
We have not qualied the resources to say that they problem) in a situation where the project manager knows
are human, material and nancial. We now take that as more about the project than the owner (the adverse
given. We have also not said that they are a novel selection problem). These risks are more acute in a tem-
grouping, nor that they are drawn from across the porary organization than a permanent one. This leads to
organization. Since the project is temporary, it is likely higher transaction costs associated with the management
(but not necessary) that the grouping will be novel, and of an endeavour as a project. If project management were
that they will be drawn from across the organization, recognized as a profession, and the use of the title Pro-
but they can come from just one function, including the ject Manager restricted to those individuals in posses-
project function. sion of related professional certication, it would help in
We would also oer this denition of a portfolio of the selection of project managers, giving greater con-
projects: dence in their competence and trustworthiness.

A portfolio of projects is an organization, (tem-


porary or permanent) in which a group of projects References
are managed together to coordinate interfaces and [1] Turner JR. What are projects and project management [Henley
prioritize resources between them and thereby Working Paper 9002]. Henley-on-Thames: Henley Management
reduce uncertainty. College; 1990.

88
8 J.R. Turner, R. Muller / International Journal of Project Management 21 (2003) 18

[2] Turner JR. The handbook of project based management. Lon- [21] Johns TG. On creating organizational support for the project
don: McGraw-Hill; 1993. management method. International Journal of Project Manage-
[3] Turner JR. The handbook of project based management. 2nd ed. ment 1999;17(1):4753.
London: McGraw-Hill; 1999. [22] Turner JR, Cochrane RA. The goals and methods matrix: coping
[4] Williamson OE, editor, Organization theory: from Chester Bar- with projects with ill-dened goals and/or methods of achieving
nard to the present and beyond, expanded edition. New York: them. International Journal of Project Management 1993;11(2):
Oxford University Press, 1995. 93102.
[5] Hart O. An economists perspective on the theory of the rm. In: [23] Scott WR. Symbols and organizations: from Barnard to the
Williamson OE, editor. Organization theory: from Chester Bar- institutionalists. In: Williamson OE, editor. Organization theory:
nard to the present and beyond, expanded edition. New York: from Chester Barnard to the present and beyond, expanded edi-
Oxford University Press; 1995. p. 15471. tion. New York: Oxford University Press; 1995. p. 3855.
[6] Turner JR, Keegan AE. Mechanisms of governance in the pro- [24] Barnard CI. The functions of the executive. Cambridge, MA:
ject-based organization: a transaction cost perspective. European Harvard University Press; 1938.
Management Journal 2001;19(3):25467. [25] Thomas J, Delisle CL, Jugdev K, Buckle P. Mission possible:
[7] Varian H. Microeconomic analysis. New York: W.W Norton; selling project management to senior executives. PM Network
1978. 2001;15(1):5962.
[8] Barnes, NM. Have projects, will manage. BBC2, London; 1989. [26] Covey SR. Principle centred leadership. London: Simon &
[9] Andersen ES, Grude KV, Haug T, Turner JR. Goal directed project Schuster; 1992.
management. London: Kogan Page/Coopers & Lybrand; 1987. [27] APM. The project management body of knowledge. 4th ed. High
[10] Cleland DI, King WR. Systems analysis and project manage- Wycombe: Association for Project Management; 2000.
ment. New York: Mc Graw-Hill; 1983. [28] Crawford L. Project management competence: the value of stan-
[11] Smith B. Project concepts. In: Eective project administration. dards. DBA thesis, Henley Management College, Henley-on-
London: Institution of Mechanical Engineers; 1985. Thames.
[12] PMI. A guide to the project management body of knowledge. 4th [29] Hobbs B, Pettersen N, Guerette H. Building, validating and
ed. Newtown Square, PA: Project Management Institute; 2000. implementing a PM competency model: the experience of one
[13] Cleland DI, Kerzner H. A project management dictionary of aerospace company. In: First to the future, Proceedings of the
terms. New York: Van Nostrand Reinhold; 1985. PMI Symposium, 2001, Nashville, Tennessee. Newtown Square,
[14] Moe TM. The politics of structural choice: towards a theory of PA: Project Management Institute; 2001.
public bureaucracy. In: Williamson OE, editor. Organization [30] Muller R, Turner JR. The impact of performance in project
theory: from Chester Barnard to the present and beyond, expan- management knowledge areas on earned value results in infor-
ded edition. New York: Oxford University Press; 1995. p. 11653. mation technology projects. Project Management: International
[15] Carroll GR. On the organizational ecology of Chester, I Barnard. Project Management Journal, Project Management Association
In: Williamson OE, editor. Organization theory: from Chester Finland, Norwegian Project Management Forum 2001;7(1):44
Barnard to the present and beyond, expanded edition. New York: 51.
Oxford University Press; 1995. p. 5671. [31] Cooke Davies, T. The real success factors in projects. In: A pro-
[16] Briner B, Grundy A, Tyrrell F, Turner JR. Case studies from the ject management odyssey, Proceedings of the PMI Europe 2001,
insuranmce service industry. In: Turner JR, Grude KV, Thurlo- the 4th PMI European Project Management Conference, Lon-
way L, editors. The project manager as change agent. London: don, June. Marlow: Marlow Events; 2001.
McGraw-Hill; 1996. p. 22743. [32] Tushman ML, Katz R. External communication and project
[17] Murray-Webster R, Thiry M. Managing programmes of projects. performance: an investigation into the role of gatekeepers. Man-
In: Turner JR, Simister SJ, editors. The Gower handbook of agement Science 1980;26(11):107185.
project management. 3rd ed. Aldershot: Gower; 2000. p. 4664. [33] Tushman ML, Scanlan TJ. Boundary spanning individuals: their
[18] Morris PWG. The management of projects. 2nd ed. London: role in information transfer and their antecedents. Academy of
Thomas Telford; 1997. Management Journal 1981;24(2):289305.
[19] Hobbs B. Private communication. 2001. [34] Turner JR, Simister SJ. Project contract management and a the-
[20] Schlichter J. Organizing for project management maturity. PMI ory of organization. International Journal of Project Manage-
Today:15. ment 2001;19(8):45764.

89
ISMFall2004Book Page 45 Tuesday, August 17, 2004 9:37 AM

IS GOVERNANCE

EVOLVING THE PROJECT


MANAGEMENT OFFICE:
A COMPETENCY
CONTINUUM
Gerard M. Hill

Many organizations today have recognized the need for a project management office (PMO) to
achieve project management oversight, control, and support. The PMOs role is to help both the
project manager and the relevant organization to not only understand and apply modern
project management practices, but also to adapt and integrate business interests into the orga-
nizations project management efforts. This article describes a five-stage competency model
for the PMO.

URING THE PAST DECADE, MODERN Each PMO stage suggests a particular level of
D project management precepts have
emerged to instill a vitalized profession-
functional capability that the PMO will have
achieved if functions are fully implemented.
al approach to project management The five PMO stages are also indicative of an or-
across countless industries. Individual capabili- ganizations maturity in project management,
ties in project management have been strength- with the PMOs role and and responsibilities
ened and enlarged through a combination of advancing from project management oversight
developments in project management process and control at the lower end of the competen-
and techniques, the implementation of training cy continuum to strategic business alignment
GERARD M. HILL is
programs, and automated tools that use ad- at the higher competency stages.
Vice President of vanced design concepts and technology.
Technical Services and Many organizations today have therefore
implemented an organizational entity, the INTRODUCTION TO THE PMO
Support for ESI
International. He is a project management office (PMO), to achieve COMPETENCY CONTINUUM
frequent speaker and project management oversight, control, sup- The PMO competency continuum provides a
author on the topic of port, and alignment. The PMOs role is to help vehicle that defines a series of PMO stages that
project management both the project manager and the relevant or- can be examined for application in an organiza-
and has worked ganization (whether an entire enterprise, a tion. The naming convention is relatively sim-
extensively with ple and somewhat consistent with PMO
business unit, or a department) to understand
Fortune 100 executives
and apply professional practices of project implementation efforts across most industries.
and managers at all
management, as well as to adapt and integrate However, these names provide only a frame of
levels to design and
deploy project business interests into the project management reference; other names can be applied as ap-
management practices. efforts. propriate to the nature of PMO responsibili-
He can be contacted at This article describes five stages of PMO ca- ties and the business environment in which it
JHill@esi-intl.com. pabilities along a competency continuum. operates.

I N F O R M A T I O N S Y S T E M S
F A L L 2 0 0 4
M A N A G E M E N T
45
90
ISMFall2004Book Page 46 Tuesday, August 17, 2004 9:37 AM

IS GOVERNANCE

FIGURE 1 Overview of PMO Capabilities across the PMO Competency Continuum

STRATEGIC ALIGNMENT
BUSINESS MATURITY Stage 5
PROCESS SUPPORT CENTER OF
Stage 4 EXCELLENCE
PROCESS CONTROL ADVANCED
Stage 3 PMO Manage
PROJECT OVERSIGHT STANDARD continuous
Stage 2 PMO Apply an improvement and
BASIC PMO integrated and cross-department
Establish
Stage 1 capability and comprehensive collaboration to
Provide a
PROJECT standard and infrastructure to project achieve strategic
OFFICE support and govern management business goals
repeatable PM
Achieve project a cohesive project capability to
methodology for
deliverables and use across all environment achieve business Multiple programs
objectives
objectives for projects Multiple projects Vice President or
cost, schedule, Multiple PMs Multiple projects Director of
and resource Multiple projects Managers
Program Multiple PMs Project
utilization Program Management
Multiple PMs Managers
1 or more projects Program
Director/Senior
Dedicated PMO
Program Manager Manager PMO Director technical staff
1 Project Manager Part-time PMO Full-time and Dedicated PMO
Enterprisewide
part-time technical and
support staff support staff support staff
PMO staff

Five general stages of PMO competency are domain of the project manager, who is respon-
prescribed. Figure 1 presents an overview of sible for the successful performance of one or
the PMO competency continuum and a de- more projects. It provides the capability to en-
scription of each of its stages. These five PMO sure professionalism and excellence in apply-
stages represent a progressive competency and ing widely accepted principles and preferred
advancement of functionality that can be at- project management practices to each project
tained to meet the needs of the project man- effort.
agement environment and the associated However, more than one project office may
business objectives of the relevant organiza- exist within an organization.When this occurs,
tion. It is presumed that a higher-stage PMO an obvious challenge lies in ensuring that each
has already achieved the competencies pre- project office pursues a common approach to
scribed for any lower-stage PMOs. Thus, if an project management. Ideally, senior members
organization wants to establish a Stage 3 stan- of the project management staff will collabo-
dard PMO, it will also have to ensure it has first rate in their design and implementation of
realized the competencies prescribed for Stage project office capability. Alternatively, a higher-
1 and Stage 2 PMOs. It is also suggested that a level PMO can be established to guide and sup-
PMO at any stage can pursue activities at any port project office activities.
level to address the needs within the relevant The inclusion of the project office in the
organization, which is far more important than PMO competency continuum is arguably an
stepping through levels of competency in se- uncertain fit. By definition, it does not influ-
quence. Moreover, it is critical to discern the ence actions and activities of more than one
approximate level of PMO competency that project manager; it has no program-level au-
the relevant organization needs. Not every or- thority or direct strategic business relevance;
ganization needs to have a PMO at Stage 5. In and it does not fulfill the traditional role of a
fact, for most organizations, the Stage 3 stan- PMO. However, the project office implements
dard PMO is probably more than adequate. and monitors the rules of project perfor-
The following subsections provide a de- mance at the project team level, and that over-
scriptive overview of each stage in the PMO sight in itself is a responsibility of PMOs at all
competency continuum. levels.Thus, the placement of the project office
at the beginning of the PMO competency con-
STAGE 1: THE PROJECT OFFICE tinuum ensures that effective project manage-
The Stage 1 PMO is the fundamental unit of ment oversight at the project level is
project oversight in the project management considered and implemented in the context of
environment. The project office is created as a PMO responsibility.
46 W W W . I S M - J O U R N A L . C O M
F A L L 2 0 0 4

91
ISMFall2004Book Page 47 Tuesday, August 17, 2004 9:37 AM

IS GOVERNANCE

The project office performs a variety of es- its responsibility for project and project team
sential project management activities, includ- performance.
ing:
Applying principles and techniques of mod- STAGE 2: THE BASIC PMO
ern project management, through the skill The Stage 2 or basic PMO is the first PMO level

T
and knowledge of the project manager, to that deals with multiple project oversight and
he project ensure that successful project performance control. It furnishes the capability to provide
is achieved. The project office concentrates aggregate oversight and control of multiple
office acts as on producing deliverables associated with projects relative to the performance of multi-
the frontline project objectives, and it manages the vital ple project managers.
point of signs of each project effort: cost, schedule, In some industries, this stage is traditionally
and resource utilization. Managing these known as the program office and represents
supervision for details invariably enhances examination of the domain of the program manager. It is possi-
implementing project performance and facilitates the appli- ble that there could be more than one basic
and cation of corrective actions to any problems PMO in the relevant organization, one for each
that are identified. program manager. However, it is not practical
integrating Serving as the direct interface to project for every program manager to independently
business team performance management. Because build the comprehensive capability that is pre-
processes in most project teams likely have a technical scribed here. Therefore, in the context of this
performance focus, the project office will model, the basic PMO is presumed to be the
the project introduce the elements of project manage- highest centralized entity of project manage-
management ment. Accordingly, the project office pro- ment that pursues its mission under the leader-
environment. vides for dif ferentiation between the ship and guidance of one designated program
technical methods, which are prescribed to
manager.
create an excellent technical product, and
The basic PMO will likely have minimal
the project management methods, which are
staff, in some cases just one individual assigned
prescribed to ensure project and business
to build the PMOs capability. Presumably, this
success.
person will be assigned full time to the PMO ef-
Applying organizational guidance in the form
fort and have access to at least a few additional
of policies, standards, executive decisions,
part-time support resources. An initiative that
and so on to each project effort. The project
is fully supported financially and appropriately
office also acts as the frontline point of super-
resourced typically is able to achieve basic
vision for implementing and integrating busi-
ness processes in the project management PMO capability and prescribed functionality
environment. within one year. This time, however, may vary
Serving as the first level of project oversight based on the business commitment and culture
and, often, the highest level of technical of the relevant organization.
oversight. Whereas higher-stage PMOs may With an emphasis on establishing control in
mandate and introduce technical methods the project management environment, the ba-
and procedures, it is the project office that sic PMO performs a variety of centralized
implements them in the project manage- project management activities, including:
ment environment. Indeed, at this level, Having primary responsibility for establish-
there is probably less emphasis on business ing a standard approach to how project man-
issues, unless the project manager has the agement is conducted in the relevant
double duty of serving also as program man- organization. This includes the introduction
ager. of common tools, repeatable processes, and
The project offices role is that of imple- preferred practices, ideally represented by
menter. It carries the policies, practices, and implementation of a comprehensive project
guidance prescribed by higher authority management methodology.
possibly higher-stage PMOs above it into the Providing the means to compile aggregate
project management environment for project results and analyses of project status and
team implementation. Yet the project office project progress as a basis for identifying and
does not have to achieve advanced levels of responding to project variations, evaluating
functionality beyond the one or several project and project manager performance,
projects it supports. Rather, a project office can and ensuring the achievement of project
exist formally in name or informally by virtue of objectives.
I N F O R M A T I O N S Y S T E M S
F A L L 2 0 0 4
M A N A G E M E N T
47
92
ISMFall2004Book Page 48 Tuesday, August 17, 2004 9:37 AM

IS GOVERNANCE

practice of project management. This often re-


TABLE 1 Twenty Project Management
quires the PMO to be proactive in planning its
Office Function
functionality and prepared to defend its busi-
ness position.
Practice Management
1. Project management methodology STAGE 3: THE STANDARD PMO
2. Project management tools
3. Standards and metrics The Stage 3 PMO is central to the PMO compe-
4. Project knowledge management tency continuum, representing the essence of
Infrastructure Management
a complete and comprehensive PMO capabili-
ty. Although it continues to address project
5. Project governance
6. Assessment management oversight and control, the Stage 3
7. Organization and structure PMO introduces a new focus on support that
8. Facilities and equipment support optimizes individual and project performance
Resource Integration in the project management environment. Its
9. Resource management purview ranges from managing multiple
10. Training and education projects and multiple project managers and
11. Career development may even include overseeing or otherwise
12. Team development aligning with one or more program managers.
Technical Support The standard PMO can evolve from earlier
13. Mentoring efforts to construct a basic Stage 2 PMO. It can
14. Planning support also be designed and implemented as the initial
15. Project auditing from scratch effort to introduce centralized
16. Project recovery
oversight, control, and support in the project
Business Alignment management environment. If a new PMO at the
17. Project portfolio management standard PMO level is to be pursued from the
18. Customer relationships outset, the designers and developers must en-
19. Vendor/contractor relationships sure that functionality prescribed for the basic
20. Business performance
Stage 2 PMO is incorporated into their PMO im-
plementation plans.
Introducing project management as a profes- Stage 3 PMO functionality is the solution for
sional discipline in the relevant organization organizations seeking to implement project
through its prescription of applicable stan- management as a core business competency or
dards, designation of qualified project man- otherwise looking to improve project manage-
agers, training and empowerment of project ment capability or increase project manage-
teams, and specification of roles and respon- ment maturity. The new Stage 3 PMO
sibilities of stakeholders in the project man- necessitates minimal staffing of a full-time PMO
agement environment. manager or director and at least two additional
The basic PMO has responsibility for estab- full- and part-time staff members qualified to
lishing the foundation of a viable project man- perform and facilitate PMO functionality de-
agement environment. This involves sign and implementation. Furthermore, the ex-
implementing capability across all PMO func- tent of standard PMO functionality may
tions. Table 1 lists 20 functions associated with warrant some part-time, possibly extended in-
a mature PMO, which are referred to later in volvement from other participants in the
this article. project environment, as well as potential par-
The new Stage 2 PMO will likely be tread- ticipation of business units in the relevant orga-
ing new ground within the relevant organiza- nization. As functionality is established, it is
tion; functional capability advancements may likely that a few more full-time staff members
initially be slow as business units become ac- will be needed to fulfill professional specialty
customed to the PMOs presence, accept tran- positions. As this PMO grows, additional full-
sition of certain responsibilities to the PMO, and part-time administrative support personnel
and evolve with greater reliance on PMO man- also will be required.
agement capability to achieve business inter- The assignment of these resources, along
ests associated with project oversight and with distinct executive business commitment
control. Conversely, the new basic PMO must to the effort, should enable complete Stage 3
demonstrate its full alignment within the rele- PMO functionality to be achieved within a two-
vant organization and its professionalism in the to three-year timeframe. Of course, timely
48 W W W . I S M - J O U R N A L . C O M
F A L L 2 0 0 4

93
ISMFall2004Book Page 49 Tuesday, August 17, 2004 9:37 AM

IS GOVERNANCE

deliberation and planning of operational needs The standard PMO has responsibility for im-
and priorities, along with assignment of ade- plementing a complete capability across all
quate initial resources, will enable significant designated PMO functions. It should examine
PMO functionality to be implemented within a the needs of the project management environ-
matter of months. Initial standard PMO operat- ment in each of the 20 PMO function models
ing capability normally can be realized within presented in Table 1.The challenge therefore is

T he
the first year of the implementation initiative.
The standard PMO performs complete cen-
to adapt each function model for optimized op-
erational fit and maximized business benefit
standard PMO tralized project management oversight and within the relevant organization. It should be re-
can be the control activities, with an added emphasis on iterated that not every PMO needs to develop full
introducing process and practice support in capability in all 20 function areas. However, the
relevant the project management environment. These PMO established at the standard level should at
organizations activities include: least consider every option for functionality.1
project Serving as the centerpiece of project man-
management agement support in the relevant organiza- STAGE 4: THE ADVANCED PMO
tion: a project management resource for The Stage 4 PMO evolves from an existing,
representative
business units, a professional practice facili- complete PMO capability and therefore is the
to business tator for project managers and project team big brother of the standard (Stage 3) PMO. Its
and industry members, and a coordinator and collabora- focus is on integrating business interests and
affiliates, tor for project stakeholders (resource man- objectives into the project management envi-
agers, customers, and vendors) activity and ronment. This implies introducing common
partners, and involvement. practices to be applied to both project manage-
professional Functioning as the interface between the ment processes and business processes. To use
institutions. business environment and the project man- a term familiar to many professional project
agement environment. The standard PMO managers, the advanced PMO helps create a
translates, as appropriate, policy and execu- projectized business environment.
tive guidance for project performance and Thus, by definition, the advanced PMO can-
implements actions and activities associated not be new. Rather, standard PMO functionality
with business interests and objectives in the must be established before an advanced PMO
project management environment. capability can be implemented. Of course, this
Acting as the facilitator of project manage- should not limit PMO designers and developers
ment environment process design and as a from incorporating advanced PMO consider-
catalyst for project management excellence. ations into their PMO implementation plans.
This extends from attending to project man- Establishing the functionality and capability of
agement methodology and practices used to the advanced PMO can be the next phase in
ensure project success; to introducing plans for PMO fulfillment within the relevant
project reporting tools and collaboration organization. It is anticipated that this stage in
techniques; to providing executive support the PMO competency continuum can be
processes regarding matters of project gover- achieved within one to two years following es-
nance, project portfolio management, and tablishment of the standard PMO capability.
business performance. The Stage 4 PMO has increased staffing and
Serving as the representative of the project the potential for direct alignment of resources.
management environment to the senior In particular, the PMO staff is enhanced to in-
executive of the relevant organization, and clude the professional and administrative re-
participating in or possibly convening and sources needed to develop, implement, and
leading associated control boards compris- manage expanded processes, programs, and
ing executives and senior managers. As such, functionality. The PMO director will have ex-
the standard PMO can be the relevant organi- panded authority to address business interests
zations project management representative in the project management environment. As-
to business and industry affiliates, partners, signed PMO resources may be aligned with a
and professional institutions. few key functional units within the PMO that
Operating as the recognized organizational provide the means to integrate business and
entity that directly or indirectly influences project management practices.
resource participation on projects, to The advanced PMO performs comprehen-
include addressing such matters as qualifica- sive, centralized project management over-
tion, training, assignment, and evaluation. sight, control, and support activities, together
I N F O R M A T I O N S Y S T E M S
F A L L 2 0 0 4
M A N A G E M E N T
49
94
ISMFall2004Book Page 50 Tuesday, August 17, 2004 9:37 AM

IS GOVERNANCE

with expanded functionality that represents a also have a business alignment or reporting af-
mature and business-oriented project manage- filiation with the center of excellence.
ment organization.These activities include: Normally, there is an executive in charge of
the center of excellence, and that individual
Appearing more and more like a separate
should either report to or have direct access to
business unit. If a PMO budget has not
the chief executive officer or any other top ex-

T he
already been established at an earlier PMO
stage, the advanced PMO normally prepares
and manages its own budget as a means of
ecutive in the relevant organization. To that
end, the center of excellence can be estab-
advanced PMO pursuing development and implementation
lished within the timeframe it takes an organi-
staff can zation to establish a new business unit, which
of advanced project management practices
generally takes from one to two years to create
include and business integration activities. a viable presence.
Collaborating with business units within the
business Although it appears at the top of the PMO
relevant organization and participating in the competency continuum, the center of excel-
analysts and development or adaptation of practices and lence is a unique project management entity
specialists processes that are common to both the busi- that can be established in one of two distinct
ness environment and the project manage-
from diverse ways. First, it could be created as a result of the
ment environment. growth and expansion of a lower-stage PMO,
professional Providing distinct expertise in state-of-the-art which would normally be the case in a small- to
disciplines. project management practices and proce- medium-sized organization. Second, it could be
dures. Senior staff members are assigned full established independently of any existing
time and represent highly skilled and knowl- PMOs, with the objective of providing strategic
edgeable professionals who apply business business guidance and direction to those sub-
acumen and advanced business and project ordinate PMOs. The latter is typical in a large
management concepts to solutions imple- global organization where the center of excel-
mented in the project management environ- lence provides some aspect of oversight, con-
ment. These individuals help implement trol, and support to PMOs serving regional
such functionality as mentoring services, business interests.
project audits, and project recovery services. Consequently, the center of excellence as-
They monitor and manage project results in sumes a strategic alignment role in the relevant
t e r m s o f b u s i n e s s p e r fo r m a n c e . T h e organization and guides the project manage-
advanced PMO staff also can include busi- ment environment in its continuous-improve-
ness analysts and specialists from diverse ment efforts.These include:
professional disciplines such as legal, con-
tract, and procurement management, cus- Providing direction and influence for enter-
tomer service, and so forth, as needed full time prise project management operations. It also
or part time to achieve PMO functionality. may oversee subordinate PMO functionality
where the relevant organization has con-
As part of the advanced PMO, implementa- structed other PMO operations relative to its
tion of the 20 PMO functions will be reassessed international, national, or other expanded
to introduce expanded capacity and programs geographical business focus.
to manage the project management environ- Building both project management environ-
ment. Because it has a focus on integrating ment and project stakeholder awareness and
business interests, the advanced PMO also en- representation across business units, cus-
sures that all PMO functions are integrated for tomer relationships, as well as vendor and
efficient and effective operations. partner relationships.
Sponsoring and conducting studies and eval-
STAGE 5: THE CENTER OF EXCELLENCE uations of project management functionality
and business effectiveness, which may
The center of excellence is a separate business
include affiliated PMOs.
unit within the relevant organization and has
Representing the business interests of the
responsibility for enterprisewide project man-
relevant organization in the project manage-
agement operations. The PMO functionality pre-
ment environment, and vice versa.
scribed for the center of excellence has a focus
on strategic business interests across the relevant The Stage 5 center of excellence PMO reas-
organization. Although lower-stage PMOs may sesses the 20 PMO functions for strategic busi-
also be assigned such tasks, it is most distinct at ness implications, together with how they can
this highest PMO level. Lower-stage PMOs may be adapted, adjusted, or redesigned for optimal
50 W W W . I S M - J O U R N A L . C O M
F A L L 2 0 0 4

95
ISMFall2004Book Page 51 Tuesday, August 17, 2004 9:37 AM

IS GOVERNANCE

use, including application by other subordi- implement all 20 functions. Instead, adapta-
nate PMOs within the relevant organization. tions and adjustments will be the rule, rather
than the exception.
CONCLUSION
The underlying premise of the PMO model pre- Note
sented here is that the PMO is a business inte- 1. For a complete description of the 20 PMO
gration activity. Not all organizations may need functions and their associated activities, see the
to evolve to Stage 5 of the continuum to authors The Complete Project Management
achieve their organizational objectives. It is Office Handbook, published by Auerbach
also unlikely that any individual PMO will Publications, 2004.

I N F O R M A T I O N S Y S T E M S
F A L L 2 0 0 4
M A N A G E M E N T
51
96
PAPERS A Fresh Look at the Contribution of
Project Management to Organizational
Performance
Monique Aubry, Universit du Qubec Montral, Montreal, Canada
Brian Hobbs, Universit du Qubec Montral, Montreal, Canada

ABSTRACT INTRODUCTION

A better understanding of organizational per- he aim of this article is to enrich the current discussion on the value
formance and the contribution that project
management can make is the aim. The article
adopts the Competing Values Framework, a
rich framework that is well established both
theoretically and empirically but is not well
T of project management by presenting empirical results from a
research on the performance of project management offices (PMOs).
It proposes a novel approach to performance inspired by the
Competing Values Framework (Quinn & Rohrbaugh, 1983).
Performance is often identified as the ultimate dependent variable in the
known in the field of project management. The literature on organizations. It is currently the focus of much attention in
framework is summarized and applied in an the project management literature (Thomas & Mullaly, 2008). The current
empirical investigation of the contribution of focus on the topic seems to be driven by the belief that organizations will
project management in general and project adopt project management only if it can be shown to generate value. After
management offices (PMOs) in particular to more than a half-century of history in the management of projects, its con-
organizational performance. The examination of tribution to performance is still not acknowledged outside the group of
11 case studies revealed multiple concurrent professionals who believe in project management. The community of pro-
and sometimes paradoxical perspectives. The fessionals and academics within project management associations are
criteria proposed by the framework have been mostly preaching to the converted. However, outside of this community, the
further developed through the identification of a value of project management is not generally recognized, particularly at sen-
preliminary set of empirically grounded per- ior levels (Thomas, Delisle, Jugdev, & Buckle, 2002).
formance indicators. The empirical results con- A major piece of research on the value of project management led by
tribute to a better understanding of the role of Thomas and Mullaly has recently been completed (Thomas & Mullaly, 2008).
project management generally and PMOs They propose a framework where project management implementation and the
specifically. They also demonstrate the useful- value of project management are aligned within the organizational context
ness of this framework for the study of project through the notion of fit. The notion of value has been used to focus on
managements contribution to organizational what project management is worth to different stakeholders. The level of
performance. analysis is the organization in both Thomas and Mullaly (2008) and the pres-
ent article. A major part of the research presented in this article was realized
KEYWORDS: organizational performance; prior to the publication of papers and the monograph by Thomas and
competing values framework; PMO; value of Mullaly (2008). However, efforts have been made to acknowledge their
project management results.
The empirical work reported in the present article centers around PMOs.
Centering the investigation on the PMO facilitates the empirical study of dif-
ferent means of contributing to organizational performance and different
perceptions of the value of these contributions. In brief, it increases the like-
lihood of producing good results for several reasons. First, organizations that
have PMOs have chosen to centralize several aspects of project management in
and around these organizational entities, making project management more
visible in the organization and easier to study. Studying the role of PMOs is,
Project Management Journal, Vol. 42, No. 1, 316 therefore, a practical means for studying project management as it is prac-
2010 by the Project Management Institute ticed in these organizations. Second, PMOs are small units that are often
Published online in Wiley Online Library located outside the major organizational units. They are thus in a position to be
(wileyonlinelibrary.com). DOI: 10.1002/pmj.20213 appraised by stakeholders in many other units. This improves the likelihood of

February 2011 Project Management Journal DOI: 10.1002/pmj 3


97
A Fresh Look at the Contribution of Project Management
PAPERS

capturing multiple conceptions of their performance by the PMO seems to take significant (Thomas & Mullaly, 2008).
contribution to the performance of the different forms. And it should be distin- The clear demonstration of the direct
organization. Third, research by Hobbs guished from the contribution of proj- influence of project management on
and Aubry (2007) has shown that the ects. The PMOs contribution is, at least return on investment (ROI) is not easily
legitimacy of PMOs is being challenged potentially, behind the performance of accomplished, as explained by Thomas
in approximately 50% of organizations. each individual project. and Mullaly (2008). In addition, the
The discourse that surrounds PMOs is The context of diversity supports reduction of project management value
thus often charged with tensions that the definition proposed here for organi- exclusively to financial indicators
make differing points of view more vis- zational performance based upon the underestimates major contributions
ible and more easily captured in empir- competing values framework. There are that project management brings to
ical studies. Fourth, Hobbs and Aubry two problems: the first one is to establish organizational successfor example,
(2007) have shown that PMOs fill many a clear definition as to what constitutes innovation (Turner & Keegan, 2004),
different organizational roles. In doing organizational performance, and the process (Winch, 2004), and people
so, they potentially contribute to the second is to propose a realistic and reli- (Thamhain, 2004). Furthermore, the
organization in many different ways, able approach to its measurement. This multifaceted concept of project per-
making the diverse contributions more leads to the research questions: What is formance is acknowledged by several
visible and easier to study. In addition organizational performance in the con- authors (Dietrich & Lehtonen, 2004;
to facilitating the study of the contribu- text of project management and how Shenhar, Dvir, Levy, & Maltz, 2001). The
tion of project management to organi- can it be assessed? balanced scorecard is based on the eco-
zational performance generally, the The next section of the article nomic conception. The balanced score-
PMO is a legitimate object of study in explores the literature on performance. card approach has been proposed to
its own right. This is followed by a presentation of the assess project management perfor-
Organizational performance is a sub- competing values framework, an inte- mance (Norrie & Walker, 2004; Stewart,
jective construct. This construct is subjec- grative model that has the ability to 2001). It has the advantage over the
tive because it exists in the minds of capture the diversity of conceptualiza- traditional economic vision of project
those who are evaluating. The organi- tions of organizational performance performance in encompassing four
zational performance of PMOs will vary found within organizations. Empirical complementary perspectives. However,
depending on who the evaluator is. results will then be presented, which the foundation of this approach rests
Most of these stakeholders belong to illustrate the usefulness of this frame- on ROI. It structures the creation of
different units that have different cul- work. The empirical portion of the arti- value hierarchically with financial value
tures and different values. cle concludes with the presentation of a at the top (Kaplan & Norton, 1996;
A construct is not directly observ- set of practical indicators that provides Savoie & Morin, 2002).
able. In order to evaluate it, the vari- a more concrete representation of orga- The second conception of perfor-
ables that form it must be identified and nizational performance and facilitates mance in the literature on project per-
examined. Justification of the PMO the construction of metrics. Finally, a formance is pragmatic. Several authors
remains a recurring problem in organi- conclusion closes the article. have encompassed the problem of per-
zations, with almost 50% reporting that formance in an approach that seeks to
the existence of their PMO has been Organizational Performance in identify success factors ( Jugdev &
recently questioned (Hobbs & Aubry, the Project Management Mller, 2005). A clarification should be
2007). A PMO would be legitimate if it Literature made here to distinguish between suc-
could convincingly demonstrate its Two conceptions of performance dom- cess factors and success criteria.
contribution to organizational perform- inate the project management litera- Success factors refer to a priori condi-
ance. However, the evaluation of its ture: economic and pragmatic. In the tions that contribute to positive results,
contribution to organizational perform- former, researchers try to demonstrate while success criteria are used to assess
ance is a complex question that may the direct economic contribution of a concrete and measurable result a pos-
have as many variations as the PMO project management to the bottom line teriori (Cooke-Davies, 2002). Cooke-
itself. This highlights the subjective side (Dai & Wells, 2004; Ibbs, Reginato, & Davies (2000, 2004) has examined
of organizational performance. Kwak, 2004). Interestingly, none of the empirical evidence supporting the
PMOs are performing many differ- these researchers have been able to many best practices and success factors
ent functions (Hobbs & Aubry, 2007). Are convincingly demonstrate the econom- found in the literature. He concludes
these different functions regarded with ic value of investment in project man- that most of the contributions have
the same value by different stakehold- agement. The results of the research been based on the opinion of members
ers? The contribution to organizational by Ibbs et al. (2004) are not statistically of the project management community

4 February 2011 Project Management Journal DOI: 10.1002/pmj


98
and that only a small number have How to Define Organizational life cycle of the organization or of the
been empirically validated. Based on Performance? unit. There exists simultaneously in the
the empirically validated data, Cooke- The concept of organizational perfor- same organization a variety of contra-
Davies (2004) proposes a set of 12 fac- mance is not new. At the end of the 1950s dictory preferences that this type of
tors related to three distinct ways of and in the early 1960s, sustained efforts definition cannot capture.
looking at performance: project man- were made notably to understand the The second approach to a defini-
agement success (time, cost, quality, success of organizations. This literature tion is based on the identification of
etc.), project success (benefits), and developed in the 1960s and 1970s, and limits/borders, which is a semantic def-
corporate success (processes and deci- after 1980 narrowed down to concepts inition. A semantic definition describes
sions that translate strategy into pro- like quality (Boyne, 2003). Several words the meaning of a term by its similarities
grams and projects). It is noteworthy are used almost as synonyms to organi- (positive semantic definition) or by its
that the success factors are different at zational performancefor example, differences (negative semantic defini-
each level of analysis. Cooke-Davies efficiency, output, productivity, effec- tion) with other terms (Van de Ven,
(2004) argues these three groups are tiveness, health, success, accomplish- 2007). Addressing the question What is
intimately linked; corporate project ment, and organizational excellence organizational performance? also
and program practices create the con- (Savoie & Morin, 2002). The concept of comes back to trying to define the
text for individual project and program organizational performance has been scope of the total construct by delimit-
practices. While the research on success adopted in this research because it is ing the components located inside and
factors has identified some conditions more appropriate in the context of outside its borders. Cameron (1981)
in organizational project management organizational project management. discusses this question from two view-
that are associated with performance at Trying to give a clear definition of points: the theoretical borders and the
different levels of analysis, the under- organizational performance is not an empirical borders. Practically speaking,
standing of performance and the a easy task. Attempts made to clarify it by the theoretical borders of organization-
priori conditions that contribute to definition have not led to an acceptable al performance do not exist. No theory
performance remains limited. result. Two alternative approaches are is completely satisfying, and the
There is no consensus on the way to explored: (1) a definition of the concept research undertaken so far is made up
assess either performance or the value by the identification of its characteris- of a collection of individual essays that
of project management. The financial tics and (2) a definition of the concept lack integration (Cameron, 1981). It is
approach alone cannot give a correct by the identification of its limits/ difficult to grasp the construct when
measure of the value of project man- borders. The definition by its character- approaching it theoretically, and still
agement for the organization. Project istics is called a definition of com- today, there is no clear definition of the-
success is a vague approximation and, ponents, where a term (in this case, oretical borders (Savoie & Morin, 2002).
as such, a rather imperfect system for organizational performance) is given in A few authors have tried instead to
measuring results. New approaches are reference to its constituent parts or its define an empirical border. Moreover,
needed in order to extricate ourselves characteristics (Van de Ven, 2007). this inductive approach is appropriate
from what looks like a dead end. Organizational performance has been when there is a high level of complexity,
Organizations are multifaceted, leading approached in the literature using dif- which is the case here (Patton, 2002).
to a variety of perspectives and evalua- ferent sets of characteristics or vari- This being the case, each study has
tion criteria. The international research ables. A first difficulty with this type of been done as in a silo, each author
on the value of project management definition is the uniformity of the levels observing in an isolated fashion a par-
draws similar conclusions (Thomas & both conceptual and operational ticular type of organization (Cameron,
Mullaly, 2008). among the characteristics (Cameron & 1981). Therefore, approaching a defini-
Whetten, 1983; Quinn & Rohrbaugh, tion of organizational performance by
What Is Organizational 1983; Van de Ven, 2007). There are other the identification of a border does not
Performance? difficulties with this type of definition. It allow us to determine what is inside the
Performance has its origin in the old is inherently subjective (Cameron, 1981). border, because theoretical research is
French parfournir and is defined today It can be difficulteven impossibleto insufficient and empirical studies are
as something accomplished (Merriam- reconcile the multiplicity of points of varied and lack integration.
Websters Collegiate Dictionary, 2007). view from different stakeholders. It can To overcome the problem of defini-
The etymology brings us straight to the be difficult even for individuals to iden- tion, Cameron (1981) suggests that orga-
point: what indeed is accomplished by tify their own preferences for an orga- nizational performance be defined as a
project management, and how should it nization. Preferences change over time, subjective construct anchored in values
be evaluated? in keeping with social values and the and preferences of the stakeholders.

February 2011 Project Management Journal DOI: 10.1002/pmj 5


99
A Fresh Look at the Contribution of Project Management
PAPERS

This definition offers significant poten- (Thompson, McGrath, & Whorton, on the values of those who are evaluat-
tial for adaptation to organizational sit- 1981). Rather than imagine a new ing (Cameron, 1986).
uations and offers the possibility of model, Quinn and Rohrbaugh The third dimension (orientation
acknowledging that a variety of per- approached the problem in a highly and purpose) was not often used in
formance evaluation models may exist original way by undertaking research empirical research based on the com-
simultaneously. This construct is also based on criteria already identified by peting values approach, including
coherent with the constructivist per- Campbell (1976, cited in Quinn & research by Cameron and Quinn
spective, which recognizes the exis- Rohrbaugh, 1983). They treated these (1999). In a fashion consistent with this
tence of several competing logics. criteria using a combination of the stream of research, only the structure
In this perspective, organizational Delphi approach and statistical model- dimension (paradox between flexibility
performance is anchored in the values ing with the participation of a group of and control) and the focus dimension
and preferences of the stakeholders. In very reputable researchers on two pan- (paradox between internal and external)
the context of project management in els. The research led to a set of 17 unique have been employed in the present
general and PMOs in particular, stake- criteria grouped into three significant research (see Figure 1).
holders are individuals and groups who dimensions: the structure dimension The research of Quinn and
have a substantial interest in the man- (paradox between flexibility and con- Rohrbaugh (1983) thus led to the for-
agement of the projects of the organiza- trol), the focus dimension (paradox mulation of a framework that presents
tion. The stakeholders could include between internal and external), and the 17 criteria and their dimensions in four
the project governance board, the busi- dimension of purpose and orientation. quadrants, each associated with a spe-
ness unit managers, the customers of These dimensions formed three sets cific preexisting model of organizational
the projects, the users, the PMO man- of values that explicitly expressed the performance: the open system model,
ager, the project portfolio managers, dilemmas or paradoxes present in the human relations model, the internal
the functional managers, project man- organizations. These values are in con- process model, and the rational goals
agers, project controllers, and so on. stant competition in organizations, and model. Sixteen of the seventeen criteria
to succeed, organizations must reach are associated with one of the four
The Competing Values good overall results, without necessari- models, each representing a different
Framework ly seeking a balance. In this context, conception of organizational perfor-
Origin and Development of the organizational performance depends mance. The 17th criterion, output quality,
Competing Values Framework
Organizational performance was the
object of a worldwide study for a nucle- Flexibility

us of researchers (Cameron & Whetten,


HUMAN RELATIONS MODEL OPEN SYSTEM MODEL
1983; Quinn & Rohrbaugh, 1983)
toward the end of the 1970s and the
beginning of the 1980s. Quinn and 1. Value of human resources working in project 12. Growth
Rohrbaugh (1983) were, however, the 2. Training and development emphasis 13. Flexibility/adaptation/innovation in project management
3. Moral of project personal 14. Evaluation by external entities (audit, benchmarking, etc.)
first to have proposed the competing 4. Conflict resolution and search for cohesion 15. Links with external environment (PMI, IPMA, etc.)
16. Readiness
values approach. This approach came
out of a research program over a period
of several years at the Institute for 17. OUTPUT QUALITY
Internal External
Government and Policy Studies, intend-
ed to evaluate performance in the pub-
lic sector. This sector is enormously
5. Information mangement and communications 8. Profit
complex, and at a time when the econ- 6. Processes stability 9. Productivity
omy was affected by high inflation, it 7. Control 10. Planning goals
11. Efficiency
was important to ensure the best possi-
ble use of public funds in all public
INTERNAL PROCESS MODEL RATIONAL GOAL MODEL
institutions (Rohrbaugh, 1981).
The theoretical basis of the compet-
ing values approach rests on the fol- Control
lowing assumption: tensions exist in all Note. The 17 elements listed in the figure are the criteria associated with each conception.

organizations where needs, tasks, val-


Figure 1: Models of organizational performance and their associated criteria.
ues, and perceptions must compete

6 February 2011 Project Management Journal DOI: 10.1002/pmj


100
was not associated specifically with any stability. One of the most important one hand, the PMO has an active role to
of the models. roles for the PMO is to monitor and play relative to the internal focus
A precision must be made to differ- control the performance of projects. through the development and dissemi-
entiate between open systems and Another important role is the standard- nation of project management metho-
rational goal models. The open systems ization of methods and processes. At dology, the fostering of internal com-
model values effectiveness. On the the same time, the PMO is part of mul- munication including the presentation
other hand, the rational goal model val- tiple project management networks of project results to upper manage-
ues efficiency, profitability, and ROI. where projects and ad hoc committees ment, and the development of compe-
The competing values approach are created, dissolved, and re-created tencies. A PMO is often responsible for
has been applied in a variety of areas. according to project management creating the common language relative
Originally, it emerged in the public sec- needs. Projects are temporary organiza- to project management. At the same
tor (Rohrbaugh, 1981), but several tions often associated with innovation time, the PMO is connected to the exter-
sectors have been studied since: higher and change, disruptive or incremental, nal world by means of consultant firms
education (Pounder, 2002), manufactur- as each project brings a new and and project management associations.
ing (McDermott & Stock, 1999), research unique solution to a particular prob- When a PMO is asked to benchmark the
and development (Jordan, Streit, & lem. In this context, the PMO supports internal project management process-
Binkley, 2003), and banking (Dwyer, creativity and innovation, or at the very es, the internal common language must
Richard, & Chadwick, 2003), as well as a least should not impede it. The PMO be translated to a universal common
cross-sector study (Stinglhamber, participates in the line of control, giving language. The PMO is an entity in which
Bentein, & Vandenberghe, 2004). More- the necessary stability while at the there exists a permanent arbitrage
over, Cameron and Quinn (1999) same time encouraging innovation and between internal and external focuses.
account for more than a thousand inter- change with flexibility. In this sense, a The examination of the two dimen-
ventions in organizations from several PMO can be said to be an ambidextrous sions in the specific context of the PMO
industrial sectors as diversified as agri- entity in developing ability in both con- confirms the existence of paradoxes
culture, insurance, and construction. trol and flexibility (Tushman & OReilly, identified by the competing values
This wide empirical base confirms the 1996). These examples illustrate the framework within a project manage-
applicability of this approach in various paradox between control and flexibility ment context. The evaluation of the
organizational contexts. as it applies in the context of PMOs. organizational performance of the
This approach has two important PMO can shed light on the different
strengths: the values underlying the eval- The Focus Dimension: Paradox Between perspectives from which organizational
uation become obvious and the changes Internal and External performance can be examined based
in the way these values are exerted are The PMO adopts an outright external on the values of those evaluating. The
also identified (Morin, Savoie, & focus when, to measure project and competing values framework repre-
Beaudin, 1994; Quinn & Rohrbaugh, project management results, it looks at sents a means to make these values
1983). In conclusion, the competing val- quantitative financial indicators and explicit, which will then lead to an
ues approach has the potential to grasp compares itself to other organizations understanding of what constitutes a
the dynamic of organizations by creat- or industries. Kendall and Rollins (2003) contribution to the performance of the
ing a dialogue between people having suggest that the main indicators for PMO and of the entire organization.
different, sometimes opposite, values measuring the value added of a PMO
that underlie their evaluation of organi- are related to three major elements: Methodology
zational performance. reduction of the life cycle of projects; The methodological framework for this
completion of more projects during research is based upon a constructivist
The Competing Values Framework in
the fiscal year with the same epistemology. In this epistemology,
the Context of PMOs
resources; and the phenomenon is in the reality and the
Because the empirical portion of this
tangible contribution for reaching researcher in part of the interaction
research is centered on the PMO, the
organizational goals in terms of cost that takes place between the researcher
two dimensions and the paradoxes that
reduction, revenue increase, and a and the object of study. Knowledge cre-
these dimensions give rise to are exam-
better return on investment. ation is the ultimate objective (Allard-
ined in the context of the PMO.
Poesi & Marchal, 1999). It modifies the
The Structure Dimension: Paradox The professionalization of project more traditional researcher role by lis-
Between Flexibility and Control management also contributes to the tening to the reality (Midler, 1994). In
The PMO usually belongs to the hierarchy fact that organizations want to compare the case of PMOs, this position is
and, as such, participates in maintaining and share their best practices. On the appropriate, as theories are almost

February 2011 Project Management Journal DOI: 10.1002/pmj 7


101
A Fresh Look at the Contribution of Project Management
PAPERS

Industrial Sector Telecommunication Financial Multimedia Financial


Years since implementation of first PMO 13 9 5 2
Number of PMOs (n 11) 4 3 3 1
Number of interviewees (n 49) 13 16 15 5
Interviewees by role
Project manager 3 3 1 1
PMO director 0 5 2 1
Manager in PMO 4 2 0 0
Executives 2 1 1 1
HR 1 0 2 0
Financial 1 1 1 0
Other manager 2 1 1 1
PMO staff 0 3 7 1
Table 1: Profile of respondents.

nonexistent and the complexity found is proposed that can capture these per- into the four conceptions (Quinn &
in the reality cannot be explained using spectives and tensions. Rohrbaugh, 1983) adapted for use with
existing simple models and a positivist This research is part of a mixed- PMOs (see Figure 1). Respondents were
approach (Hobbs & Aubry, 2007). Just method program of research built for chosen to represent different roles,
as organizations are complex social robustness (Brown & Eisenhardt, 1997). potentially leading to different concep-
entities, so too are the specific organi- In this specific research project, a case- tions of the PMOs contribution to orga-
zational project management struc- study approach has been used to nizational performance (see Table 1).
tures that encompass PMOs. The explore and better understand the con- In addition to interviews, a ques-
methodological strategy is designed to tribution of PMOs to organizational tionnaire was built with the objective
understand such complexity. performance (Yin, 1989). Four organi- of capturing the different conceptions of
Drawing on Van de Vens (2007) zations participated in this research. A the PMOs contribution and their under-
engaged scholarship brings together retrospective historical approach cov- lying values. The questionnaire contains
different points of view of key people ering the period since before the imple- the same 17 criteria. Respondents were
involved with PMOs, using a combina- mentation of the first PMO was adopt- asked to assess the importance of each
tion of qualitative and quantitative ed. The periods covered ranged from 2 of the criteria in their current context
instruments. PMOs represent a com- to 13 years, with an average of 7.24 using a 5-point Likert scale, where 1 was
plex phenomenon not only by the vari- years. As is common among PMOs not important at all and 5 was very
ety of their expressions, but also by the globally, the PMOs in these organiza- important. Criteria with a score of 4 or 5
number of entities they relate to in a tions were restructured every few years were considered important.
single organization. In matrix organiza- (Hobbs & Aubry, 2007). A total of 11 dif-
tions, projects naturally form networks, ferent PMOs were analyzed, each con-
Empirical Results
which converge in one or more PMOs. stituting a case study. A Typology of PMOs Based on
Yet, within a single organization there Two types of data were collected: Organizational Performance Criteria
are multiple managers and profession- interviews and a questionnaire. The As mentioned previously, the compet-
als in relationships with the PMO. How most important data came from inter- ing values framework takes into
do they value the PMOs contribution to views where open-ended questions were account the values within organiza-
organizational performance? It asked specifically on the performance of tions, and it provides an instrument
depends on the perspective of each of the PMO. Interviews were codified and that helps highlight paradoxes between
these stakeholders. In the quest for a analyzed in a grounded theory approach values. The diagram shown in Figure 1
better understanding of the PMOs con- (Strauss & Corbin, 1998). Transcripts forms a typology based upon the four
tribution to organizational perfor- were coded using the 17 criteria from the different conceptions of organizational
mance, a global methodological strategy competing values framework grouped performance. Each PMO from the case

8 February 2011 Project Management Journal DOI: 10.1002/pmj


102
Equilibrium can be observed that project managers
Internal Focus Internal/External External Focus do recognize the importance of the
PMOs contribution in the human rela-
Flexibility 1 1 3
tions and rational goals criteria. In this
Equilibrium 1 0 2 particular case, the PMO is active in
flexibility/control human resource functions participat-
Control 2 1 0 ing in the career path of people working
in projects. However, project managers
Table 2: Number of PMOs classified by the importance of their organizational performance criteria.
do not recognize that internal process-
es are as important. This may be
because project managers perceive
studies has been assigned a position single organization. This should trans- these processes to be a constraint on
within this framework so that they can late in this research into different pat- their freedom to act. The PMO director
be compared more easily with each terns for different stakeholders in their considers all criteria as important. This
other (see Table 2). evaluation of the importance of organi- situation is not specific to this case
Each of these case studies has its zational performance criteria. Figure 2 the same result was observed in almost
own dynamics. Space restrictions pre- illustrates this phenomenon within one all cases. This confirms the positive bias
vent these from being explored here. In organization from the case studies. It of PMO managers when asked to assess
all, the results are coherent with the pertains to the evaluation of the existing the PMOs contribution to organiza-
proposals from the competing values PMO at the time of interviews. As can be tional performance. Other managers
framework. There is no such thing as a observed, differences exist between within the PMO are more critical
perfect balance, but rather different val- actors in their assessment of the impor- specifically of human resource criteria.
ues underlie what organizational per- tance of the performance criteria. Otherwise, these managers recognize
formance represents in organizations Globally, results confirm the posi- the importance to other groups of crite-
(Cameron, 1986). A plurality of perspec- tive contribution of PMOs to organiza- ria. Executives recognize some impor-
tives on the contribution to organiza- tional performance. The Likert scale tance for all groups of criteria, but none
tional performance is observed. The offers the choices of low values of reach the level of significant impor-
results show that certain perspectives importance, but no criteria falls under tance. This is consistent with the poor
prevail at certain times and evolve with the middle position, which indicates perception of project managements
the context. One problem when try- that all criteria are of at least some ability to contribute to organizational
ing to understand the contribution of importance. The results of one case performance reported by Thomas et al.
PMOs to organizational performance is with 15 respondents are illustrative but (2002). Curiously, the human resource
related to the fleetingness of the PMO cannot be generalized. manager does not attribute that much
itself (Hobbs & Aubry, 2007). Assessing An examination of the variations in importance to the PMOs contribution
something that is fast-moving contains responses between stakeholders in dif- to human resource performance. This
in itself a major limitation. In this ferent roles is informative. In Figure 2, it may be a reflection of issues related to
research, the evolution of the PMO and
the evolution of the perception of its
contribution to organizational per-
5
formance were tracked. Results from
Importance of criteria

the pre-PMO period have been intro- 4


duced in this analysis. However, no
(means)

pattern resembling a predetermined 3

life cycle in the evolution of their con- 2


tribution to organizational perfor-
mance was found. 1
Human relations Internal processes Rational goals Open systems
The Actors View of the PMOs Groups of criteria
Contribution to Organizational Project Manager PMO Director Manager within PMO
Performance
Executive HR Manager Financial Manager
As discussed earlier, the competing val-
Manager Elsewhere PMO Employee
ues framework is based upon the
assumption that many conceptions of
Figure 2: Importance of organizational performance criteria by role.
organizational performance coexist in a

February 2011 Project Management Journal DOI: 10.1002/pmj 9


103
A Fresh Look at the Contribution of Project Management
PAPERS

jurisdiction over human resource each of the four models and the 17 cri- according to employee wishes. It is
issues. The human resource managers teria. The sets of indicators create value unusual that an organization structured
do recognize the PMO within the inter- in two ways. First, they enrich the by project (which is the case here)
nal processes and rational goals crite- understanding of the PMOs contribu- stresses the contribution that personnel
ria. PMO employees are of particular tion to organizational performance make to projects and intensifies the
interest; they attribute significant by providing detail that is meaningful role of the PMO in human resources
importance to human resource and in this context. Second, they provide management. But management of
open system criteria but not that much the basis for instruments to measure the human resources, in this organizational
to internal process and rational goal presence of the models in real organiza- context, is a particularly critical func-
criteria. tional settings. tion. The personnel are exceedingly
These results also show some of the The goal here is to be more specific young: the average age is less than 30
paradoxes in the expectations relative and to identify relevant concrete indica- years old. This fact accounts for the
to the PMOs contribution to perfor- tors in the context of PMOs. Transcripts strength of the company at the same
mance. For example, the financial man- of the interviews have first been coded time as it produces its own nightmares.
ager considers the internal processes using the 17 criteria. Then, excerpts These teenagers require a consider-
criteria to be the PMOs most important have been scrutinized for their meaning able amount of supervision in order to
contribution to organizational per- in order to group multiple variations respect the project constraints and the
formance. However, project managers under a common indicator. From this never-ending challenges. There is an
consider these to be the least impor- second step, a list of 79 unique indica- important shortage in qualified person-
tant. When it comes time for the PMO tors was produced. nel in this high-technology sector. This
manager to discuss the contribution of Indicators inform us about the vari- company has invested extensively in
his/her unit with the financial manager, ety of possible ways the models mani- training in conjunction with local gov-
arguments concerning internal fest themselves and the different ways ernments. It has also implemented an
processes will probably be important, that measurements can be made in the internal school to provide skilled work-
but the same arguments are not as like- different PMOs. An advantage of this ers for its own development needs.
ly to convince the project managers. It exercise is to render explicit and opera- Furthermore, personnel turnover is sig-
is easy to see how these differences in tional notions about the contribution of nificant. In summary, the management
perceptions and appreciations can lead the PMO to organizational perfor- of human resources is an important
to tensions and even to conflicts. mance that until now may have function in which the PMO plays an
Results from the competing values remained abstract. See Appendix A for active role.
framework may also offer the opportu- the complete list of indicators. The significant number of indica-
nity to open up discussion between dif- tors identified within the human rela-
ferent and sometimes opposite ways of Indicators Within the Human tion conception shows that underlying
understanding the PMOs contribution Resources Conception values exist in organizations to assess the
to organizational performance. The indicators related to human re- contribution of the PMO to organiza-
sources foster a clearer understanding tional performance regarding the human
The Development of Indicators of the role that the PMO can play in this resources. A PMO manager confirms the
Specific to the Evaluation of the area. Indicators vary greatly from one impact of his entity on the degree of sat-
PMOs Contribution to Organi- organization to the next. It appears that isfaction of project managers:
zational Performance each of the four organizations has a dif-
Well, we took it all [multiple PMOs]
The four conceptions and the 17 gener- ferent flavor in the way the human
and centralized it; it [the degree of
ic criteria initially proposed within the resource contribution of the PMO is
satisfaction of the employee] went
competing values framework (Quinn & valued. For example, the organization from 0 to 24 in 12 months. The ener-
Rohrbaugh, 1983) can be applied in dif- in the multimedia industry stands out gy, the empowermentthere was a
ferent contexts. For this reason they are with the largest number of indicators in huge improvement.
at a more abstract level. Cameron and human resources. The scope of these
Quinn (1999) recognize that the four indicators often covers all of the It is also notable that the PMO plays
models and the 17 criteria are quite resources working on projects rather a social role and that it has an influence
abstract and recommend that sets of than only PMO employees or project on the work-family balance. Analysis
criteria be developed that are specific managers. In this case, the PMO plays a also revealed that the capacity for nego-
to a particular use. In a manner consis- direct role in the development of com- tiation is a competence essential to the
tent with this recommendation, specific petencies for personnel, according to resolution of conflicts surrounding
sets of indicators were developed for the needs of upcoming projects, and the state of advancement of projects.

10 February 2011 Project Management Journal DOI: 10.1002/pmj


104
line, and there was a need in anoth-
The role of PMOs within human One of the financial services case
er product line, we could move that
resource management is often neglected organizations had two interrelated
person over if the competence and
in the literature on PMOs with the excep- PMOs: a central one and one in a busi-
the profile both matched the
tion of a few authors that dedicated their ness unit. These two PMOs dont value requirements.
efforts to emphasizing this role the same elements as far as the quality
(Crawford & Cabanis-Brewin, 2006; of deliverables and communications Productivity in project management
Huemann, Keegan, & Turner, 2007). management is concerned. This is is a constant challenge. The challenge is
From the qualitative analysis, it can be understandable in complementary even more evident in international
seen that the PMO can make a signifi- but paradoxical terms: the business- organizations where there is competi-
cant contribution to organizational unit PMO values product quality and tion between different units in different
performance regarding human business results, while the central PMO locations. Productivity in project man-
resources and that concrete indicators values process maturity and project agement becomes an important factor
can be used to assess this. performance in terms of cost, schedule, for decisions as to where projects will be
and the project requirements. As can be executed. The role of PMOs in project
Indicators Within the Internal
seen from this example, two PMOs in productivity is often recognized in the
Processes Conception
the same organization may have com- literature (Kendall & Rollins, 2003).
The internal processes conception of
plementary but conflicting priorities. These authors link the PMOs productiv-
organizational performance shows the
ity directly to its legitimacy.
largest number of individual indicators Indicators Within the Rational Goals
The criterion of planning in the
of the four conceptions. This empha- Conception
PMO context mostly refers to their
sizes the position of project manage- The indicators for the rational goals or
strategic and multiproject functions.
ment and the PMO in their traditional efficiency conception are less numer-
Indicators proposed by respondents
roles of process management. ous but are the most frequently cited.
give some idea of the concrete out-
Many indicators bear on the criteria Indicators included the profit criterion,
comes that relate to the strategic action
of information and communication which is not surprising; they reflect the
of PMOs in selecting the right projects.
management. The PMO seems to col- interest in selecting the right projects
Indicators also emphasize the role of
laborate in many networks and play a the ones that contribute to the busi-
the PMO in the portfolio equilibrium
central role in the circulation of infor- nesss bottom line. The contribution
regarding their risks and their short-
mation. A respondent emphasizes this of the PMO to organizational per-
and long-term benefits. Capacity plan-
role when saying: formance is recognized through its
ning indicators recognize the PMOs
involvement in portfolio and program
I think that the PMO has an impor-
role in the allocation of resources on
management. Regarding the productiv-
tant role in the sense that they have the long run, the capacity to deliver,
ity criterion, the contribution of PMOs
a vision of what is going on else- and the capacity for internal resources
can be significant, particularly in the
where in the organization. . . . to absorb changes from projects. The
allocation and efficient use of
Normally, the PMO has antennae in alignment of employees objectives
resources. This point highlights an
each portfolio. . . . I think it could with organizational objectives was also
have a unifying role.
important issue for organizations hav-
found under the planning criteria. This
ing multiple highly specialized expert
indicator recognized that PMOs are
profiles working on multiple projects.
Indicators also reflect both the qual- involved in the appraisal process of
From the case studies, this issue was of
ity of information and the ease of its flow individuals working on projects.
prime importance in two organizations
throughout the organization. The crite- There are two indicators related to
having projects where 200 to 300
ria dedicated to the stability of processes efficiency criterion. The first one refers
employees work in parallel. In those
pinpoints more specifically the tradi- to the relationship that a PMO has with
two organizations, PMOs centralize the
tional role of PMOs in standardization of other parts of the organization. From
allocation of human resources. The idea
project management. The criteria of interviewees, this refers to numerous
here is to not leave anyone on the
control included of course meeting inefficient meetings with PMO employ-
bench. The director of a PMO pin-
costs, deadlines, and project scope. ees or managers. PMOs often perform a
pointed his role in the allocation of
However, PMO control is becoming monitoring and controlling function on
project managers:
increasingly diversified and is often project performance. In order to
exercised on the processes themselves. We wanted to use project manage- accomplish this mission, additional
Of particular interest is the situation ment resources in a better way so information to that available on reports
with multiple PMOs where the values that, for example, if a project man- or Web sites is needed. Different com-
given to indicators are quite different. ager was freed up in one product mittees or meetings are called to share

February 2011 Project Management Journal DOI: 10.1002/pmj 11


105
A Fresh Look at the Contribution of Project Management
PAPERS

information. People working on proj- justify the number of staff working in it. formance. Research on organizational
ects repeat the same information in The criterion of responsiveness includ- performance in project management
several different meetings, resulting in ed two indicators that were mentioned does not produce entirely satisfactory
inefficiency and frustration. At the quite often by respondents: (1) the results. Each piece of research brings
same time, the role of the PMO in nego- PMO should be able to respond quickly important contributionsbut consid-
tiation when it comes time to decide on in order to make projects succeed and ered all together, a global vision of proj-
the status report color is recognized. (2) the PMO should be able to adapt to ect management performance at the
The second indicator mentioned is different situations. organizational level is still lacking.
project success and, more specifically, Indicators in this open system con- The competing values framework
the role of the PMO in fostering project ception contrast with the ones included has the advantage of integrating the
success. in the internal processes conception. financial perspective of performance
This confirms that paradoxes exist. There with the other conceptions in order to
Indicators Within the Open System are individuals that value the respect of form a multidimensional perspective.
Conception project management processes, while at Indeed, the four conceptions of the
Indicators within the open systems or the same time in the same organization, framework give us a multifaceted repre-
effectiveness conception are the fewest others value exactly the opposite and sentation of the performance of organi-
in number. These mostly deal with flex- encourage delinquency. The competing zational project management. The
ibility, adaptation, and innovation in values framework offers an opportunity rational goals and efficiency conception
project management. The first criteri- to acknowledge these paradoxes and, integrates the economic values of prof-
on, growth of the organization, refers from there, to open up a dialogue to itability, project management efficiency,
directly to the business side of the develop a common basis and under- and return on investment. The open sys-
organization, taking into account sales, standing of organizational performance. tems and effectiveness conception
qualitative results, and effectiveness. This work supports the recognition of the includes variables that measure growth
These elements relate to the benefits diversity of the contributions a PMO can and take into consideration innovation
from projects. It emphasizes that the make to an organization. And also it and project effectiveness. The human
PMO could be involved in a wider proj- should help to develop the awareness of relations conception emphasizes the
ect life cycle, covering the benefits from PMO managers and their employees of development of human resources, cohe-
projects. This stretches project man- the paradoxes that are at work in their sion, and personnel morale. All of these
agement toward the product life cycle organizations regarding their perform- elements are often absent from the eval-
(Jugdev & Mller, 2005). ance. This approach can be a valuable uation of organizational performance.
The criterion of flexibility, adapta- instrument to initiate a dialogue and The internal processes conception
tion, and innovation show numerous come to a common understanding of captures measurements related to cor-
indicators, few of which are shared what is valued. PMO actions could then porate processes tied to project manage-
from one case to another, except for be aligned on this common understand- ment such as project delivery method-
delinquency, with respect to project ing of organizational performance. ologies, communication processes, and
methodology. One respondent stated: knowledge management processes.
Indicators of Output Quality
A lot of flexibility, what matters to me is Overall, the competing values model
The quality criteria include three indi-
the result; I couldnt care less if we used bears directly on performance (objective
cators. First, the quality of the product
a saw or a screwdriver to get there. This variable) instead of bearing on success
has been included here, as many inter-
highlights the fact that the contribution factors (explanatory variables).
viewees mentioned this element in
of the PMO to organizational perfor- Organizational performance must
relation with the PMOs contribution to
mance is not limited to the establishment be examined from different viewpoints
the overall quality performance. The
of a methodology in project manage- and be scrutinized at several loci of
second and third are indicators of satis-
ment (from the internal processes con- analysis. PMOs are positioned at the
faction of the PMO sponsor and clients
ception), but also the flexibility with interface of several entities, some of
of the PMO. These indicators are quite
which the PMO encourages its use. which belong to project networks and
common when assessing quality.
While no indicators were mentioned in others to operational organizations
the evaluation by the external entities Conclusion (Lampel & Jha, 2004). They are in touch
criterion, respondents mentioned some The aim of this study is to understand with the projects, programs, project port-
for the criteria of having links with the the contribution of the PMO to organi- folios, corporate strategy, and functional
external environment. Benchmarking zational performance with a view to and business units. The PMO is there-
was mentioned often in a context of jus- understanding project managements fore at the center of numerous perspec-
tification of the PMO, particularly to contribution to organizational per- tives on organizational performance.

12 February 2011 Project Management Journal DOI: 10.1002/pmj


106
References with a strategic project office: Select, management system. Harvard Business
Allard-Poesi, F., & Marchal, C. (1999). train, measure, and reward people for Review, 74(1), 7585.
Construction de lobjet de recherche organization success. Boca Raton, FL: Kendall, G. I., & Rollins. S. C. (2003).
[Construction of the research object]. Auerbach. Advanced project portfolio manage-
In R.-A. Thitart (Ed.), Mthodes de Dai, C. X. Y., & Wells, W. G. (2004). An ment and the PMO: Multiplying ROI at
recherche en management [Research exploration of project management warp speed. Boca Raton, FL: J. Ross
methods in management] (pp. 3456). office features and their relationship to Publishing.
Paris: Dunod. project performance. International Lampel, J., & Jha, P. J. (2004). Models of
Boyne, G. A. (2003). What is public Journal of Project Management, 22, project orientation in multiproject
service improvement? Public 523532. organizations. In P. W. G. Morris & J. K.
Administration, 81, 211227. Dietrich, P., & Lehtonen, P. (2004). Pinto (Eds.), The Wiley guide to manag-
Brown, S. L., & Eisenhardt, K. M. Successful strategic management in ing projects (pp. 223236). Hoboken,
(1997). The art of continuous change: multi-project environment: Reflections NJ: Wiley.
Linking complexity theory and time- from empirical study. Paper presented McDermott, C. M., & Stock, G. N.
paced evolution in relentlessly shifting at the IRNOP VI Conference, Turku, (1999). Organizational culture and
organizations. Administrative Science Finland. advanced manufacturing technology
Quarterly, 42, 134. Dwyer, S., Richard, O. C., & Chadwick, K. implementation. Journal of Operations
Cameron, K. S. (1981). Construct space (2003). Gender diversity in manage- Management, 17, 521533.
and subjectivity problems in organiza- ment and firm performance: The influ-
Merriam-Websters collegiate diction-
tional effectiveness. Public Productivity ence of growth orientation and organi-
ary (11th ed.). (2007). Springfield, MA:
Review, 5(2), 105121. zational culture. Journal of Business
Merriam-Webster.
Cameron, K. S. (1986). Effectiveness as Research, 56, 10091019.
Midler, C. (1994). Lauto qui nexistait
paradox: Consensus and conflict in Hobbs, B., & Aubry, M. (2007). A multi-
pas [The car that didnt exist]. Paris:
conceptions of organizational effective- phase research program investigating
Interditions.
ness. Management Science, 32, 539553. project management offices (PMOs):
The results of phase 1. Project Morin, E. M., Savoie, A., & Beaudin, G.
Cameron, K. S., & Quinn, R. E. (1999).
Management Journal, 38(1), (1994). Lefficacit de lorganisation:
Diagnosing and changing organiza-
7486. Thories, reprsentations et mesures
tional culture: Based on the competing
[Effectiveness of the organization:
values framework. Reading, MA: Huemann, M., Keegan, A., & Turner, R. J.
Theories, representation and measures].
Addison-Wesley. (2007). Human resource management
Montral, Qubec: Gatan Morin diteur.
Cameron, K. S., & Whetten, D. A. in the project-oriented company: A
review. International Journal of Norrie, J., & Walker, D. H. T. (2004).
(1983). Organizational effectiveness: A
Project Management, 25, 315329. A balanced scorecard approach to proj-
comparison of multiple models. New
ect management leadership. Project
York: Academic Press. Ibbs, C. W., Reginato, J., & Kwak, Y. H.
Management Journal, 35(4), 4756.
Cooke-Davies, T. J. (2000). Towards (2004). Developing project manage-
improved project management prac- ment capability: Benchmarking, matu- Patton, M. Q. (2002). Qualitative
tice: Uncovering the evidence for effec- rity, modeling, gap analysis, and ROI research & evaluation methods.
tive practices through empirical studies. In P. W. G. Morris & J. K. Pinto Thousand Oaks, CA: Sage.
research. Unpublished doctoral thesis, (Eds.), The Wiley guide to managing Pounder, J. S. (2002). Public accounta-
Metropolitan University, Leeds, United projects (pp. 12141233). Hoboken, NJ: bility in Hong Kong higher education:
Kingdom. Wiley. Human resource management implica-
Cooke-Davies, T. J. (2002). The real Jordan, G. B., Streit, D., & Binkley, S. J. tions of assessing organizational effec-
success factors on projects. (2003). Assessing and improving the tiveness. International Journal of Public
International Journal of Project effectiveness of national research labora- Sector Management, 15, 458474.
Management, 20(3), 185190. tories. IEEE Transactions on Engineering Quinn, R. E., & Rohrbaugh, J. (1983).
Cooke-Davies, T. J. (2004). Project Management, 50(2), 228235. A spatial model of effectiveness crite-
management maturity models. In Jugdev, K., & Mller, R. (2005). A ret- ria: Towards a competing values
P. W. G. Morris & J. K. Pinto (Eds.), rospective look at our evolving under- approach to organizational analysis.
The Wiley guide to managing projects standing of project success. Project Management Science, 29, 363377.
(pp. 12341264). Hoboken, NJ: Wiley. Management Journal, 36(4), 1931. Rohrbaugh, J. (1981). Operationalizing
Crawford, K. J., & Cabanis-Brewin, J. Kaplan, R. S., & Norton, D. P. (1996). the competing values approach: Mea-
(2006). Optimizing human capital Using balanced scorecard as a strategic suring performance in the employment

February 2011 Project Management Journal DOI: 10.1002/pmj 13


107
A Fresh Look at the Contribution of Project Management
PAPERS

service. Public Productivity Review, Lessons for team leadership. organisations as information pro-
5(2), 141159. International Journal of Project cessing systems? Paper presented
Savoie, A., & Morin, E. M. (2002). Les Management, 22, 533544. at the PMI Research Conference,
reprsentations de lefficacit organisa- Thomas, J. L., Delisle, C., Jugdev, K., & London.
tionnelle: Dveloppements rcents Buckle, P. (2002). Selling project man- Yin, R. K. (1989). Case study research:
[Representations of the organizational agement to senior executives: Framing Design and methods. Newbury Park,
effectiveness: Recent developments]. In the moves that matter. Newtown Square, CA: Sage.
R. Jacob, A. Rondeau, & D. Luc (Eds.), PA: Project Management Institute.
Transformer lorganisation: La gestion Thomas, J. L., & Mullaly, M. E. (Eds.).
stratgique du changement [Transfor- (2008). Researching the value of
ming the organization: The strategic project management. Newtown Monique Aubry, PhD, is a professor in the grad-
management of change] (pp. 206231). Square, PA: Project Management uate programs in project management at the
Montral, Qubec: Revue Gestion. Institute. School of Business and Management at the
Shenhar, A. J., Dvir, D., Levy, O., & Thompson, M. P., McGrath, M. R., & Universit du Qubec Montral. She is an
Maltz, A. C. (2001). Project success: Whorton, J. (1981). The competing active researcher within the Project
A multidimensional strategic concept. values approach: Its application and Management Research Chair under the aegis of
Long Range Planning, 34, 699725. utility. Public Productivity Review, 5(2), project governance. Before her academic
Stewart, W. E. (2001). Balanced score- 188200. career, she worked for more than 20 years in
card for projects. Project Management the management of major projects in the finan-
Turner, R. J., & Keegan, A. E. (2004).
Journal, 32(1), 3853. cial sector. She is a member of the Project
Managing technology: Innovation,
Management Institutes Standards Member
Stinglhamber, F., Bentein, K., & learning, and maturity. In P. W. G.
Advisory Group.
Vandenberghe, C. (2004). Congruence Morris & J. K. Pinto (Eds.), The Wiley
de valeurs et engagement envers lor- guide to managing projects (pp.
ganisation et le groupe de travail 567590). Hoboken, NJ: Wiley.
[Values congruence and engagement Tushman, M. L., & OReilly, C. A., III. Brian Hobbs, PhD, PMP, Project Management
towards the organization and working (1996). Ambidextrous organizations: Research Chair (www.pmchair.uqam.ca), has
group]. Psychologie du Travail Managing evolutionary and revolu- been a professor at the Universit du Qubec
et des Organisations, 10(2), tionary change. California Manage- Montral in the Masters Program in Project
165187. ment Review, 38(4), 830. Management for 25 years, a program accredited
Strauss, A., & Corbin, J. (1998). Basics of Van de Ven, A. H. (2007). Engaged by the Project Management Institutes Global
qualitative research: Techniques and scholarship: Creating knowledge for Accreditation Center. He has been a member of
procedures for developing grounded the- science and practice. Oxford, UK: PMIs Standards and Research Member Advisory
ory (2nd ed.). Thousand Oaks, CA: Sage. Oxford University Press. Groups. He has presented many papers at both
Thamhain, H. J. (2004). Linkages of Winch, G. M. (2004, July). Rethinking research and professional conferences world-
project environment to performance: project management: Project wide.

14 February 2011 Project Management Journal DOI: 10.1002/pmj


108
Appendix: List of Indicators contribution of PMOs to organizational Table A1 presents indicators within con-
This Appendix presents the list of all 79 performance. Indicators have been clas- ceptions of human resources, output
unique indicators that were identified sified using the conceptions on per- quality, and internal processes. Table A2
from the four case studies. Indicators formance from the competing values presents indicators within rational goals
provide an explicit element to assess the framework (Quinn & Rohrbaugh, 1983). and open systems conceptions.

CRITERIA INDICATORS CRITERIA INDICATORS


Indicators Within Human Resources Conception Indicators Within Internal Processes Conception
1. Value of human 1. Empowerment 6. Information and 1. Accuracy of information in progress
resources working 2. Stimulating projects (participate to communication report
in project something big) management 2. Transparency of information in
3. Visibility for good work in projects progress report
4. Individual assessment 3. Circulation of the information on
5. Internal recruitment privileged projects (transverse role)
6. Team work valued 4. Keeping the memory of projects for
7. Trust in PMO forecasting (historical statistics)
5. Existence of project documentation
2. Training and 8. Training in project management 6. Capacity to absorb a lot of infor-
emphasis on 9. Level of experience of the personnel mation (project managers and
development working in PMO coordinators)
10. Encouragement for PMP 7. Creation of open places for people
11. Individual development plan for project to discuss
management competencies 8. Politicsvisibility of the CEO
12. Diversity in competencies 9. Learning from errors
13. Coaching
14. Organization of eventsknowledge 7. Stability in 10. Standardization in the way things
transfer processes are done
15. Change management in project 11. Importance of the resource
management appointment process
12. Existence and stability of project
3. Moral on project 16. Pleasure in working management processes
personal 17. Career job security
18. Employee satisfaction in project 8. Control 13. Rigor in the project management
19. Work-family equilibrium process
20. Number of overtime hours 14. Control of the appointment process
to avoid thieving
4. Conflict 21. Conflict prevention 15. Capacity to act (difference between
resolution and 22. Resolution of conflict in HR management monitoring and controlling)
search for 23. Negotiation on progress report 16. Control of project delivery date
cohesion (e.g., color code) 17. Control of costs
24. Negotiation on actions to be taken from 18. Control of scope
progress report 19. Control of earned value
25. Negotiation on project selection in portfolio 20. Ratio number of changes/respect
of cost
Indicators Within Output Quality
21. Equilibrium between time and
5. Output quality 1. Quality of the product budget
2. Satisfaction of the sponsor 22. Control of risks
3. Satisfaction of clients 23. Percent of precision in control data
Table A1: List of indicators within conceptions: Human resources, output quality, and internal processes.

February 2011 Project Management Journal DOI: 10.1002/pmj 15


109
A Fresh Look at the Contribution of Project Management
PAPERS

CRITERIA INDICATORS CRITERIA INDICATORS


Indicators Within Rational Goals Indicators Within Open System
9. Profit 1. Profit from projects 13. Growth 1. Sales results
2. Benefits planning within 2. Qualitative element from busi-
project business case ness case (business positioning)
3. Effectiveness
10. Productivity 3. Order in productivity
4. Best utilization of resources in 14. Flexibility/ 4. Innovator, creator, and good at
project management (leave less adaptation/ conflict or problem resolution
people on the bench) innovation in 5. Hiring of project management
5. Index of productivity project personnel having creative skills
6. Bureaucracy management 6. Existence of initiatives in project
7. Internal competition (e.g., between management methodology
units in different countries) (sometimes being delinquent)
8. Existence of an organizational 7. PMO product a variety of reports
structure to deliver projects 8. Hiring of external consultants to
know the best practices in project
11. Planning in 9. Importance of the strategic dimension in management
goals to reach the selection of the good projects 9. Evolution in project management
10. Equilibrium in projects of a portfolio process and tools
(risk, benefits on the short-, medium-, 10. Participation of stakeholders in
and long-term value) the development and evolution of
11. Prediction of the delivery capabilities project management processes
(resource allocation)
12. Alignment of enterprise objectives with 15. Assessment by none
the employees objectives external entities

12. Efficiency 13. Efficiency in the relations between 16. Links with 11. Link with the local PMI (some-
PMO and functional or business units external times too much!)
negotiation on projects environment 12. Benchmarking
14. Project success (PMO impacts on
projects) 17. Readiness 13. Being agile
14. Responsiveness in appointment
when urgent need
Table A2: List of indicators within conceptions: Rational goals and open system.

16 February 2011 Project Management Journal DOI: 10.1002/pmj


110
111
112
113
114
115
116
117
118
119
120
www.hbrreprints.org

Strategy at many companies is


almost completely
disconnected from execution.
The Office of Strategy
Establishing a dedicated unit
to orchestrate both will help to
Management
bridge the divide.
by Robert S. Kaplan and David P. Norton

Included with this full-text Harvard Business Review article:

1 Article Summary
The Idea in Briefthe core idea
The Idea in Practiceputting the idea to work

2 The Office of Strategy Management

11 Further Reading
A list of related materials, with annotations to guide further
exploration of the articles ideas and applications

Reprint R0510D

121
The Office of Strategy Management

The Idea in Brief The Idea in Practice


Most large organizations fail to achieve Design your office of strategy management to perform these functions:
profitable growthdespite ambitious
plans. Why the gap between intended and CREATE AND OVERSEE YOUR STRATEGY REFINE STRATEGY
actual performance? Theres an alarming MANAGEMENT SYSTEM Evaluate new strategic ideas coming from
disconnect between the parts of the orga- Help the executive team select performance within the organization and convey promising
nization that formulate corporate strategy targets and identify required strategic initia- ones to senior management.
and the functions, processes, and people tives. Initiate and administer your companys
required to execute it. strategic performance reporting system. To MANAGE STRATEGIC INITIATIVES
67% of HR and IT departments strategies maintain integrity of performance data, create Manage strategy-related initiatives that cross
dont reflect corporate strategy. 60% of or- and enforce uniform organization-wide metrics. unit and functional lines, to ensure they re-
ganizations dont link their financial bud- Incorporate changes in corporate strategy ceive sufficient resources and attention. Moni-
gets to strategic priorities. Compensation into all documents and tools that the com- tor progress of all strategic initiatives and re-
packages of 90% of frontline employees pany uses to track strategic performance port on them to top management.
show no connection to the success or fail- such as strategy maps and the Balanced
ure of strategy execution. 95% of the typical Scorecard. CONSULT WITH KEY STRATEGY SUPPORT
companys workers are unaware of, or dont FUNCTIONS
understand, its strategy. ALIGN THE ORGANIZATION Planning and budgeting. Work with the fi-
How to close the breach between strategy Actively manage organizational alignment nance department to ensure that corporate
formulation and execution? Create an office with corporate strategy. Institutionalize the and unit budgets reflect those established
of strategy management (OSM). Your OSM use of a common strategic performance re- during the strategic planning process and
couples the units responsible for strategic porting system by all units. Ensure that busi- that each units budget includes resources
planning with those performing the activi- ness unit and support unit strategies are needed for the units contribution to cross-
ties required to implement strategysuch linked to one another and to the companys functional strategic initiatives.
as establishing budgets, communicating strategy. Human resource alignment. See that the
COPYRIGHT 2005 HARVARD BUSINESS SCHOOL PUBLISHING CORPORATION. ALL RIGHTS RESERVED.

strategy to the workforce, and designing HR function manages employee incen-


compensation systems that reward strate- COMMUNICATE STRATEGY tives, competency development programs,
gic performance. Through newsletters, CEO speeches, and and annual performance reviews in a man-
The payoff for designing an effective OSM? other channels, communicate corporate strat- ner consistent with corporate and business
A corporate strategy that delivers on its egy, targets, and initiatives to the workforce. unit strategic objectives.
promises. Thanks in part to its OSM, the Coordinate with HR to ensure that education
Knowledge management. Coordinate
Chrysler Group generated $1.2 billion in about the strategy management process is in-
with the chief learning officer to ensure that
earnings and launched a series of exciting cluded in training programs.
the best practices and ideas most critical to
new cars in 2004while the rest of the U.S. the corporate strategy are shared through-
domestic auto market languished. REVIEW STRATEGY
out the organization.
Organize and lead monthly strategy-review
meetings, briefing the CEO about strategic
concerns in advance. Document needed ad-
justments to strategy and execution identified
during meetings and follow up to ensure that
changes are implemented. Help the chief fi-
nancial officer prepare strategy updates for
board meetings.

page 1

122
Strategy at many companies is almost completely disconnected from
execution. Establishing a dedicated unit to orchestrate both will help to
bridge the divide.

The Office of Strategy


Management
by Robert S. Kaplan and David P. Norton

Most companies have ambitious plans for It doesnt have to be like this. For the past 15
growth. Few ever realize them. In their book years, we have studied companies that have
Prot from the Core, Chris Zook and James achieved performance breakthroughs by
COPYRIGHT 2005 HARVARD BUSINESS SCHOOL PUBLISHING CORPORATION. ALL RIGHTS RESERVED.

Allen report that between 1988 and 1998, adopting the Balanced Scorecard and its associ-
seven out of eight companies in a global sam- ated tools to help them better communicate
ple of 1,854 large corporations failed to strategy to their employees and to guide and
achieve protable growth. That is, these com- monitor the execution of that strategy. (For
panies were unable to deliver 5.5% annual real background on the Balanced Scorecard, see
growth in revenues and earnings while earn- our book The Strategy-Focused Organization,
ing their cost of capital (a rather modest hur- Harvard Business School Press, 2000.)
dle). Yet 90% of the companies in the study Some companies, of course, have achieved
had developed detailed strategic plans with better and longer-lasting improvements than
much higher targets. others. The organizations that have managed
Why is there such a persistent gap between to sustain their strategy focus have typically es-
ambition and performance? The gap arises, we tablished a new unit at the corporate level to
believe, from a disconnect in most companies oversee all strategy related activities, an ofce
between strategy formulation and strategy exe- of strategy management (OSM), as we call it.
cution. Our research reveals that, on average, This might appear to be nothing more than
95% of a companys employees are unaware of, a new name for the familiar strategic planning
or do not understand, its strategy. If the em- unit. But the two are quite different. The typi-
ployees who are closest to customers and who cal planning function facilitates the annual
operate processes that create value are un- strategic planning process but takes little or no
aware of the strategy, they surely cannot help leadership role in seeing that the strategy gets
the organization implement it effectively. executed. The companies we studied, however,

harvard business review october 2005 page 2

123
The Office of Strategy Management

recognize that effective strategy execution re- veys that we conducted of HR and IT
quires communicating corporate strategy; en- managers reveal that the strategies of fully 67%
suring that enterprise-level plans are translated of those organizations are not aligned with
into the plans of the various units and depart- business unit and corporate strategies; nor do
ments; executing strategic initiatives to deliver HR and IT departmental plans support corpo-
on the grand plan; and aligning employees rate or business-unit strategic initiatives. Bud-
competency development plans, and their per- geting is similarly disconnected: Some 60% of
sonal goals and incentives, with strategic objec- organizations do not link their nancial bud-
tives. Whats more, they recognize that the gets to strategic priorities. Incentives arent
companys strategy must be tested and aligned, either: The compensation packages of
adapted to stay abreast of the changing compe- 70% of middle managers and more than 90%
tition. The OSM becomes the central point for of frontline employees have no link to the suc-
coordinating all these tasks. It does not do all cess or failure of strategy implementation. Pe-
the work, but it facilitates the processes so that riodic management meetings, corporate com-
strategy execution gets accomplished in an in- munication, and knowledge management are
tegrated fashion across the enterprise. similarly not focused on strategy execution.
In the following pages, we will describe how What can companies do to change this state
the concept of the ofce of strategy manage- of affairs? The experience of the Chrysler Group
ment came into being and how it has helped rst suggested to us that the answer lies in
companies align key management processes to bringing all strategy-related activities into a sin-
strategy. Although the companies we have gle functional unit. After a string of innovative
studied use the Balanced Scorecard as the successes in the early 1990s, Chrysler had hit a
framework for their strategy management sys- dry spell. Performance problems were exacer-
tems, we believe that the lessons we draw are bated by an economic downturn, rising costs,
also applicable to companies that do not use and encroaching imports, and by 2000, the
the Balanced Scorecard. company was staring at a projected decit of
more than $5 billion for the coming year. At this
Strategy Management: The New point, the parent company, DaimlerChrysler,
Support Function appointed a new CEO, Dieter Zetsche, who in-
The exhibit The Old Strategy Calendar de- troduced the Balanced Scorecard as part of a
picts the strategy management schedule at a major change in strategy. The project was spear-
typical large company. The process starts headed by Bill Russo, vice president of business
about midway through the scal year, when strategy, whose unit worked with Chryslers ex-
the CEO and the executive team get together ecutive team to translate the companys new
to clarify their strategic vision and update the strategy into a Balanced Scorecard. Russos unit
strategy. Sometime afterward, similar pro- also served as trainer and consultant to help
cesses take place at the business and func- Chryslers business and support units create
tional units, led by unit heads and other senior local scorecards that were aligned with corpo-
executives. Toward the end of the third quar- rate objectives and customized to local opera-
ter, the nance function takes the baton, nal- tions. Once the design phase had been com-
izing corporate and unit budgets. At the end of pleted and scorecards had been cascaded
the year, the HR function conducts employ- throughout the company, the strategy group
ees annual performance reviews and orches- maintained responsibility for the data collection
trates the setting of professional goals and de- and reporting processes for the scorecards.
velopment programs. Throughout the year, Up to this point, Chryslers Balanced Score-
meanwhile, different teams and units have en- card project had followed a traditional course.
Robert S. Kaplan (rkaplan@hbs.edu) gaged in performance reviews, corporate com- Where Chrysler broke new ground was in the
is the Baker Foundation Professor at munication, and knowledge sharing. roles assumed by the strategy group. The
Harvard Business School in Boston. The problem with this approach is that the group took the lead in preparing scorecard-
David P. Norton (dnorton@bscol.com) activities are carried out largely in isolation related materials to communicate the strat-
is the founder and president of the and without guidance from the enterprise egy to the more than 90,000 employees.
Balanced Scorecard Collaborative strategy. This partition of responsibilities cre- Russo began to brief Zetsche before each
(www.bscol.com), based in Lincoln, ates the gulf between an organizations strat- management meeting about issues that had
Massachusetts. egy and its processes, systems, and people. Sur- been revealed through the scorecard report-

harvard business review october 2005 page 3

124
The Office of Strategy Management

ing and that required management attention much the same way. A central project team at
and action. In his capacity as a member of the the Pentagon headquarters, under the leader-
executive team, Russo followed up after each ship of the Army chief of staff, developed the
meeting to make sure that the required items initial scorecard, which the Army called the
were communicated and acted upon. As a re- Strategic Readiness System (SRS). The project
sult of this proactive involvement in agenda team also selected the software to be used for
setting and follow-up, the responsibilities of scorecard reporting and established systems and
the business strategy function expanded to in- processes so that the scorecard would be regu-
corporate many new cross-enterprise strategy larly populated with valid, timely data. In the
execution processes. Thus was born Chryslers next phase, the team helped to cascade score-
Ofce of Strategy Managementa unit cur- cards to 13 major subcommands and subse-
rently employing some 13 full-time people quently to more than 300 subsidiary commands
who not only manage the companys strategy throughout the world. The centralized project
but also assist the business units in developing team provided training, consulting, software,
new products. Chryslers new approach to and online support for the dispersed project
strategy execution appears to have paid off teams. The central team also reviewed the score-
handsomely. In 2004, despite a weak domes- cards produced by local project teams to ensure
tic automobile market, Chrysler successfully that their goals were aligned with those articu-
launched a series of exciting new cars and lated on the chief of staffs scorecard.
generated $1.2 billion in earnings. The Armys project team, like its counter-
The U.S. Armys Balanced Scorecard project part at Chrysler, soon took on more than the
produced an ofce of strategy management in traditional roles of scorecard custodian and

The Old Strategy Calendar


Strategy management at most companies consists of processes carried out in
isolation by different groups with different reporting lines. Thats why strategy
becomes disconnected from the units responsible for executing it.

Copyright 2005 Harvard Business School Publishing Corporation. All rights reserved.
Q1 Q2 Q3 Q4

Senior executives
Strategy update: CEO and executive have no consistent way
team clarify vision. to describe strategy.

Line-of-business and support-unit


Two-thirds of HR and IT leaders conduct strategic planning. 70% of
organizations are
middle managers do
not aligned with strategy.
not have strategy-linked
60% of CFO oversees budgeting. incentive pay.
The vast majority of companies do not link
executive teams spend budgets to strategy.
less than one hour per HR coordinates personal goal setting,
month discussing strategy. incentives, and personal development.

Top executives conduct monthly management reviews.


95% of the
Corporate communications unit disseminates information. workforce does not
understand the strategy.
Chief knowledge officer oversees knowledge sharing.

= deciencies in old management process

harvard business review october 2005 page 4

125
The Office of Strategy Management

How to Wield Influence and Stay Informed


by Graham Sher
As the chief executive of the nonprot that with easy access to timely, unltered informa- mary responsibility for most of these pro-
manages the supply of blood products for tion about our strategy implementation. cesses, but not all. For example, in 2004, the
all of Canada except the province of Que- Because of my urgent need to accomplish OSM led the project team that developed
bec, I instituted an ofce of strategy man- change, I followed the unconventional route the strategy maps and scorecards for the en-
agement to help me cope with three big of establishing an ofce of strategy manage- terprise, our three operating divisions, and
challenges in implementing a strategic ment at the outset of our Balanced Score- two support unitshuman resources and
agenda. First, I spend a great deal of time card project. I also wanted the OSM to re- information technology. For some pro-
dealing with external demands and constit- port directly to methat was a way to cesses, however, the OSMs role is more inte-
uents. In addition to reporting to the board highlight the importance of this ofce to my grative and facilitative than direct. For ex-
of directors of my organization, Canadian strategic agenda. But the OSM needed other ample, the chief nancial ofcer has
Blood Services (CBS), I must also focus on clearly dened linkages or relationships, too; primary responsibility for budgeting, with
the 12 Canadian provincial and territorial I want change at CBS to come from within, the OSM playing a coordinating role.
governments that provide its funding. So I not to be imposed from above. To that end, I We launched the OSM with three full-
have limited time and information with created a dotted-line reporting relationship time individuals. The OSM leader is a vice
which to manage internal issues. between the OSM and two other key execu- president and a member of the executive
Also, while many people believe that chief tives at CBS, the CFO and the COO, who ulti- management team; her position in the orga-
executives wield direct and easy inuence, mately are going to help execute the change nization is consistent with the importance
the reality is that any CEO has a difcult agenda. I did not create the new corporate- we give this function. She leads and facili-
time inuencing his or her organization. A level OSM unit lightly. Its positioning in the tates the integration of strategy into all our
CEOs attempts to command and control un- organization enables me to fulll my inter- core processes. In addition, we have two in-
dermine the authority of senior executives. I nal duties as a change leader but doesnt af- dividuals reporting to the OSM leader to
want to exert my inuence indirectly and in fect my ability to meet the many external provide day-to-day management of the of-
a way that empowers my executives and cre- obligations I have as the CEO of a rapidly ce; to manage the multiple work streams
ates an environment in which they can lead evolving public-sector organization emerg- and cross-functional teams; to lead and facil-
and manage their parts of the organization. ing from crisisCanadas blood-supply sys- itate meetings; to educate people on the Bal-
I set the tone, and I dene the strategic tem was completely revamped after thou- anced Scorecard and other strategy-focused
agenda, communicate it, and ensure that it sands of people received contaminated practices and tools; and to perform analyses
gets undertaken, but I dont command any blood in the 1980s and 1990s. of problems, performance, and metrics. This
parts of the organization. As for the OSMs responsibilities, I see should be the right complement of individu-
My third challenge is staying informed. strategy management as being made up of als to help support the leader of the OSM,
Information, particularly bad news, is l- three high-level processes: strategy formula- and ultimately the rest of the executive
tered before it gets to me. I typically do not tion, leading to strategy execution, leading in team, in undertaking our ambitious change
see the most timely, valid information about turn to strategy learning, which then cycles agenda for this year.
CBSs current performance. Before our OSM back to strategy formulation. The exhibit
was implemented, we were spending way The Processes of Strategy shows the activi- Graham Sher is the CEO of Canadian Blood
too much time debating the quality of our ties within the categories. The OSM has pri- Services, based in Ottawa.
informationobviously an unwieldy way of
executing strategy and a very time-intensive The Processes of Strategy
way of conducting management meetings.
I see the Balanced Scorecard, managed by The Canadian Blood Services Ofce of Strategy Management has direct or indirect
an ofce of strategy management, as a way of (shaded items) responsibility for strategic processes, which fall into three categories.
overcoming these three barriers to success.
STRATEGY EXECUTION
Corporation. All rights reserved.

The Balanced Scorecard empowers execu- STRATEGY FORMULATION STRATEGY LEARNING


tives, as opposed to invading their territory
Business School Publishing

Environmental assessments Balanced Scorecard Benchmarking


Copyright 2005 Harvard

and undermining their authority. It gives me performance reporting


performance management information that Strategic planning Initiative management Best-practice sharing
is aligned at all executive levels and appropri-
Budgeting Communicating strategy Internal coaching and
ately validated before it comes to my atten- change management
tion. Much of management is a search for the
Personal scorecards
truth. The Balanced Scorecard provides me

harvard business review october 2005 page 5

126
The Office of Strategy Management

consultant. It established and took ownership Can$900 million, more than 4,000 employees,
of a strategy communication program. The and 17,000 volunteers, is an excellent example
Army team created a Web site that was acces- of an organization that created an OSM at the
sible from around the world in both classied beginning of its journey to becoming more
and unclassied versions, developed an online strategy focused. (See the sidebar How to
portal and library containing information Wield Inuence and Stay Informed, by CEO
about the SRS, wrote articles about the initia- Graham Sher.)
tive, published a bimonthly newsletter, con- What should people designing an OSM bear
ducted an annual conference, led periodic in mind as they embark on the project?
conference calls with SRS leaders at each com- Through research into Balanced Scorecard
mand level, and conducted scorecard training, best practices, weve identied the activities
both in person and on the Web. This extensive that should be directly managed by or coordi-
communication process was critical for edu- nated with an OSM. Some of these activities
cating soldiers and civilian employees and specically those involved in creating and
gaining their support for the new strategy. managing the scorecard, aligning the organiza-
And the Army project team, much as tion, and setting the agenda for monthly strat-
Chryslers did, began to facilitate the monthly egy reviewsare the natural turf of an OSM.
discussions at headquarters about the readi- They did not exist prior to the introduction of
ness status of units around the world. Once the Balanced Scorecard, so they can be given
again, an ad hoc project team had turned into to a new unit without infringing on the cur-
a sustainable part of the organizations struc- rent responsibilities of any other department.
ture (the team and the SRS survived the ap- But many other activitiesstrategic planning,
pointment of a new chief of staff in June budget supervision, or HR training, for in-
A unit with responsibility 2004). stanceare already the territory of other
The creation of a central ofce for strategy units. In these cases, the company needs to be
for the implementation execution may appear to risk reinforcing top- explicit about the allocation of responsibilities
down decision making and inhibiting local ini- between the OSM and other functional units.
of strategy becomes a tiative, but it does just the opposite. A unit We have identied the following basic OSM
convenient focal point with responsibility for the implementation of tasks:
strategy becomes a convenient focal point for Create and manage the scorecard. As the
for ideas that percolate ideas that percolate up through the organiza- owner of the scorecard process, the OSM must
up through the tion. These emerging ideas can then be put on ensure that any changes made at the annual
the agendas of quarterly and annual strategy strategy-planning meeting get translated into
organization. reviews, with the best concepts being adopted the companys strategy map and Balanced
and embedded in enterprise and business unit Scorecard. Once the executive team has ap-
strategies. The OSM is a facilitating organiza- proved the objectives and measures for the
tion, not a dictating one. subsequent year, the OSM coaches the team in
selecting performance targets on the score-
What Good OSMs Do card measures and identifying the strategic in-
Most of the organizations we have studied fol- itiatives required to achieve them. As guardian
low the path Chrysler and the Army took: The of the scorecard, the OSM also standardizes
Balanced Scorecard project team incremen- the terminology and measurement denitions
tally and organically assumes more and more across the organization, selects and manages
responsibilities on its own initiative. But thats the scorecard reporting system, and ensures
not the only way to institute an OSM. From the integrity of the scorecard data. The OSM
these cases, we have learned what functions need not be the primary data collector for the
an effective OSM must perform and how an scorecard, but it should oversee the processes
OSM must relate to other functions within the by which data are collected, reported, and val-
organization. As a consequence, a few organi- idated. Finally, the OSM serves as the central
zations we advise have recently opted to make scorecard resource, consulting with units on
the creation of an OSM an early and integral their scorecard development projects and con-
part of their scorecard initiatives. Canadian ducting training and education.
Blood Services, the main provider of blood Align the organization. A company can ex-
services in Canada with an annual budget of ecute its strategy well only if it aligns the strat-

harvard business review october 2005 page 6

127
The Office of Strategy Management

egies of its business units, support functions, the executive team on strategic options. But
and external partners with its broad enterprise developing strategy should not be a onetime
strategy. Alignment creates focus and coordi- annual event. After all, performance mea-
nation across even the most complex organiza- sures, such as those supplied by the Balanced
tions, making it easier to identify and realize Scorecard, provide continual evidence about
synergies. At present, few companies actively the validity of the assumptions underlying a
manage the process of alignment; in many companys strategy. Those assumptions can be
cases, unit strategies have only rhetorical links discussed periodically by the executive team,
with corporate strategy. The OSMs weve stud- which can update the strategy if appropriate.
ied help the entire enterprise to have a consis- And strategy development should not be done
tent view of strategy and to systematically only by senior managers. The OSM or strategic
manage organizational alignment. The OSM planning unit can act as a lter for new ideas
oversees the process of developing scorecards that come from within the organization.
and cascading them through the levels of the Weve found that most planning units adapt
organization. It denes the synergies to be cre- fairly quickly to the continual strategy devel-
ated through cross-business behavior at lower opment process we observe at scorecard-
organization levels and ensures that individ- driven companies. The additional processes
ual business unit and support unit strategies represent a natural extension of, and comple-
and scorecards are linked to each other and to ment to, their traditional work. Problems arise
the corporate strategy. when a scorecard project is managed by a
Review strategy. For all their professed group from outside planning (such as HR,
commitment to strategy, senior managers quality, or an ad hoc team). As the scorecard
spend remarkably little time reviewing it. Our acquires strategic importance, conicts over
research suggests that 85% of executive leader- strategy development can arise between the
ship teams spend less than one hour per planning unit and the scorecard team. If this
month discussing their units strategy, with occurs, top management should quickly
50% spending no time at all. Companies that merge the two groups.
manage strategy well behave differently. Top Communicate strategy. Effective commu-
managers usually meet once a month for four nication to employees about strategy, targets,
to eight hours. This meeting provides the op- and initiatives is vital if employees are to con-
portunity to review performance and to make tribute to the strategy. Canon U.S.A., a score-
adjustments to the strategy and its execution. card user, describes its internal communica-
The underlying hypotheses of the companys tion process as democratizing strategy, and it
strategy can be tested and new actions initi- actively promotes understanding of the com-
ated. Managing this meeting is a core function panys strategy and the scorecard in all busi-
of the OSM. It briefs the CEO in advance ness units and support functions. Strategy
about the strategic issues identied in the communication, therefore, is a natural turf for
most recent scorecard so that the agenda can an OSM. But as with strategy planning, inter-
focus on strategy review and learning, rather nal communication is sometimes another
than just a short-term nancial performance units existing responsibility. In these situa-
review and crisis management. The OSM then tions, the OSM has tended to take an editorial
monitors the meeting to determine action role, reviewing the messages to see that they
plans and follows up to ensure that the plans communicate the strategy correctly. In cases
are carried out. Since the board of directors where the corporate communications group
also plays an important role in reviewing and has little knowledge of or focus on strategy,
guiding strategy, the OSM helps the chief - such as at Chrysler and the U.S. Army, the
nancial ofcer prepare the board packet and OSM takes on primary responsibility for com-
agenda for board meetings. municating both the scorecard and strategy to
Develop strategy. Typically, strategy formu- employees. In either situation, the OSM
lation is the responsibility of the existing stra- should always take the lead in crafting strat-
tegic planning unit. The unit performs exter- egy messages delivered by the CEO, because
nal and internal competitive analysis, one of the most effective communication
conducts scenario planning, organizes and channels is having each employee hear about
runs an annual strategy meeting, and coaches strategy directly from the CEO. Finally, as part

harvard business review october 2005 page 7

128
The Office of Strategy Management

of its communication responsibilities, the by customer service) or by an ad hoc team


OSM must cooperate with HR to ensure that drawn from the functions or units affected.
education about the scorecard and its role is Responsibility for managing initiatives that al-
included in employee training programs. ready have a natural home should remain
Manage strategic initiatives. Strategic ini- with the associated unit or function. The OSM
tiativessuch as a TQM program or the im- intervenes only when an initiative falls behind
plementation of CRM softwareare discre- schedule, is over budget, or is not delivering
tionary programs that help companies expected results. But the OSM should manage
accomplish strategic objectives. The executive initiatives that cross unit and functional
team typically identies these initiatives as linesit can thus make sure that they get the
part of its annual planning process, although resources and attention they need. In all cases,
new initiatives may arise throughout the year. the OSM retains responsibility for monitoring
Ideally, the entire portfolio of such initiatives the progress of strategic initiatives and report-
should be assessed and reprioritized several ing on them to top management.
times annually. The screening, selection, and Integrate strategic priorities with other
management of strategic initiatives are what support functions. Existing functional depart-
drive change in the company and produce re- ments retain prime responsibility for three
sults. Our experience suggests that such initia- other key processes necessary for successful
tives should be managed separately from rou- strategy implementation: planning and bud-
tine operations. Typically, they are managed geting, human resource alignment, and
by the units most closely associated with them knowledge management. These processes are
(a CRM project, for instance, is best managed critical for effective strategy execution, and

The New Strategy Calendar


At scorecard-driven companies, the strategic processes are carried out or super-
vised by the ofce of strategy management in coordination with the appropriate
management teams or executives. This ensures that the strategy is fully reected
in all strategy-related activities at all levels of the company.

Q1 Q2 Q3 Q4

Copyright 2005 Harvard Business School Publishing Corporation. All rights reserved.
Strategy update: CEO and executive team clarify
vision. Balanced Scorecard team facilitates
development of corporate scorecard and strategy map.

Line-of-business and support-unit leaders align their units


with strategy. Board of directors becomes aligned. Balanced
Scorecard team facilitates development of unit scorecards.

Ofce of strategy management


oversees alignment CFO, HR head, CIO, and COO
of all management processes conduct planning and budgeting.
with strategy.
HR oversees alignment of personal goal setting,
incentives, and personal development with strategy.

Top executives conduct monthly management reviews.

Corporate communications unit disseminates information.

Chief knowledge officer oversees knowledge sharing.

harvard business review october 2005 page 8

129
The Office of Strategy Management

the OSM should play a consultative and inte- HR performs these activities in a manner con-
grative role with the respective functional de- sistent with corporate and business unit strate-
partments. gic objectives. The goal is to make strategy ev-
Planning and Budgeting. At most corpora- eryones job.
tions, the various functional departments are Knowledge Management. Finally, the OSM
responsible for planning how the corporation needs to ensure that knowledge management
will allocate resources over the year. The - focuses on sharing the best practices most criti-
nance department oversees budgeting and the cal for the strategy. If managers use the wrong
allocation of cash to the units and cross-func- benchmarks, the companys strategy will fall
tional initiatives; IT makes recommendations short of its potential. At some companies,
about investments in databases, infrastruc- learning and knowledge sharing are already
ture, and application programs; and HR makes the responsibility of a chief knowledge or learn-
plans for hiring, training, and leadership de- ing ofcer; in those cases, the OSM needs to co-
velopment. For a strategy to be effective, all ordinate with that persons ofce. But if such a
the functional plans must be aligned with the function does not already exist, the OSM must
strategy. The budgets prepared by the nance take the lead in transferring ideas and best prac-
department, for example, should reect those tices throughout the organization.
established in the strategic planning process The exhibit The New Strategy Calendar
and should incorporate funding and person- illustrates the activities that a properly consti-
nel resources for cross-functional strategic ini- tuted OSM will be engaged in during the year.
tiatives. To ensure this alignment, the OSM The strategy cycle launches at the beginning of
must work closely with all these functional the second quarter, when the OSM starts to
units. plan strategy and update the enterprise score-
Human Resource Alignment. No strategy can card. After the enterprise strategy meeting, the
be effective unless the people who have to OSM starts the process of aligning the organi-
carry it out are motivated and trained to do so. zation with the enterprise goals. Before the
Motivation and training is, of course, the natu- end of the third quarter, it will be coordinating
ral domain of HR, which typically carries out with nance to bring unit-level plans and bud-
annual performance reviews and personal gets in line with strategy, and by the beginning
goal setting and manages employee incentive of the fourth quarter, it will be working with
and competency development programs. It is HR on aligning the competency development
the responsibility of the OSM to ensure that and incentives of employees with scorecard ob-
jectives. While these calendar-driven processes
are going on, the unit continually engages in
control and learning: reviewing and communi-
To fulll its responsibilities success-
cating strategy, managing initiatives, and shar-
fully, an ofce of strategy management
ing best practices.
at a large company typically needs
only six to eight full-time people. Positioning and Stafng the OSM
Executing strategy usually involves making
STRATEGY MANAGEMENT TYPICAL changes that only a CEO can empower, and the
PROCESS # OF FTE OSM will be most effective when it has direct
Scorecard management 1.0 access to the CEO. Barbara Possin, the director
Organization alignment 1.0 1.5 of strategic alignment at St. Marys Duluth
Copyright 2005 Harvard Business School
Publishing Corporation. All rights reserved.

Clinic, told us she was able to overcome resis-


Strategy reviews 0.5 1.0
tance to her initiatives because managers knew
Strategic planning 0.5
she had a direct reporting line to the companys
Strategy communication 0.5 1.0 chief operating and chief executive ofcers. An
Initiative management 1.0 1.5 OSM buried deep in the nance or planning de-
Planning and budgeting 0.5
partment may nd it difcult to command sim-
ilar respect and attention from senior execu-
Workforce alignment 0.5
tives for strategy management priorities.
Best-practice sharing 0.5 1.0 The simplest solution, therefore, is to place
TOTAL FTE POSITIONS 6.0 8.5 the OSM on a par with major functions, such

harvard business review october 2005 page 9

130
The Office of Strategy Management

Its simplest to place the as nance and marketing, that report directly does not usually involve hiring expensive new
office of strategy to the CEO. The OSM serves, in effect, as the talent. The OSM is typically staffed with peo-
CEOs chief of staff. But if the OSM has origi- ple who led the Balanced Scorecard project
management on a par nated within a powerful function, such a posi- they often come from the planning and -
with functions that tioning may not be feasible. In that case, the nance functions, but some come from other
OSM will usually report to the chief of the staff groups such as quality, HR, and IT. Several
report directly to the function in which it is nestedsuch as the organizations we studied have reported that
CFO or vice president of strategic planning the people assigned to their OSMs do not con-
CEO. The office serves, in
but with occasional direct access to the CEO. stitute a net increase in the organizations head
effect, as the CEOs chief At the Mexican insurance company Grupo Na- count. In many cases, the evolution of a well-
cional Provincial (GNP), for example, the OSM functioning OSM actually helps reduce overall
of staff. reports both to the chief executive and to the head count, thanks to the OSMs role in
chief nancial ofcer. The OSM sets the streamlining and focusing management pro-
agenda for a weekly meeting with the CEO and cesses and helping managers eliminate layers
CFO and for a broader weekly meeting with of staff engaged in data gathering and report-
the six top company executives. The ofce of ing. The OSM, however, should be assessed by
strategy management at GNP also has a ma- the value it creates through successful strategy
trixed relationship with 20 Balanced Score- execution, not by whether it can reduce head
card managers in the two major business units count.
and nine support units and with the owners of
the major strategic initiatives. The relationship Many organizations have achieved dramatic
enables the OSM to coordinate the strategic performance improvements by sustaining a
planning done in the business and support focus on implementation of strategy. We have
units. captured and codied a body of knowledge
The OSM may be an important functional from these successful organizations that pro-
unit, but it doesnt have to be large; it is cer- vides the foundation for an emerging profes-
tainly not our goal to encourage companies to sional function focusing on the management
build a new bureaucracy. Although Chrysler of strategy. An ofce of strategy management
employs 13 full-time people in its OSM, re- that is positioned at the level of other senior
ecting the units involvement in product de- corporate staff ofces and has responsibility
velopment, our experience suggests that rms for managing and coordinating all the key
with sales of $500 million to $5 billion and strategy management processes can help com-
1,000 to 10,000 employees can get by with panies realize the benets from this body of
fewer than ten people. In principle, as the ex- knowledge.
hibit on the previous page shows, a fully func-
Reprint R0510D

tioning OSM should not need more than six To order, see the next page
to eight full-time-equivalent positions to cope or call 800-988-0886 or 617-783-7500
with its activities. or go to www.hbrreprints.org
We have observed that establishing an OSM

harvard business review october 2005 page 10

131
The Office of Strategy Management

Further Reading
ARTICLES
Turning Great Strategy into Great Transforming Corner-Office Strategy into
Performance Frontline Action
by Michael C. Mankins and Richard Steele by Orit Gadiesh and James L. Gilbert
Harvard Business Review Harvard Business Review
July 2005 May 2001
Product no. R0507E Product no. R0105D

The authors provide additional guidelines for Gadiesh and Gilbert argue that managers can
coupling strategy formulation with execu- improve strategy execution by clarifying and
tion: 1) Articulate your companys strategy in communicating strategy. To that end, they
clear, simple terms describing what your com- recommend crafting a strategic principlea
pany will and wont do. 2) Challenge the as- pithy, memorable, action-oriented phrase that
sumptions underlying your strategic plans to distills your companys strategy and commu-
ensure that they reflect real market econom- nicates it throughout the workforce. Examples
ics and your organizations capacities. 3) Use include AOLs Consumer connectivityany-
common, agreed-on approachessuch as time, anywhere, The Vanguard Groups Creat-
benchmarkingto assess strategic perfor- ing unmatchable value for investors/owners,
mance. 4) Discuss resource deployments and Southwests Meet customers short-haul
among business units early to create feasible travel needs at fares competitive with the cost
forecasts. 5) Identify key actions that must be of automobile travel.
taken to deliver planned performance, and
An effective strategic principle helps people
determine when they must be taken. 6) Track
make trade-offs between competing re-
real-time results against your strategic plan,
sources, links leaders strategic insights with
revising assumptions and reallocating re-
line operators pragmatic sense, and sets clear
sources where needed. 7) Strengthen execu-
boundaries within which employees operate
tion ability, by making selection and develop-
and experiment. Communicate your strategic
ment of managers a top priority.
principle consistently, simply, and repeatedly.

To Order

For Harvard Business Review reprints and


subscriptions, call 800-988-0886 or
617-783-7500. Go to www.hbrreprints.org

For customized and quantity orders of


Harvard Business Review article reprints,
call 617-783-7626, or e-mai
customizations@hbsp.harvard.edu

page 11

132

Potrebbero piacerti anche