Available online at www.sciencedirect.

com

Industrial Marketing Management 37 (2008) 686 697

The B2B-RELPERF scale and scorecard: Bringing relationship marketing theory into business-to-business practice
Luis Filipe Lages a,b,, Andrew Lancastre a,c, , Carmen Lages d
a

Faculdade de Economia, Universidade Nova de Lisboa, Campus de Campolide, 1099-032 Lisboa, Portugal b MIT's Deshpande Center for Technological Innovation, USA c IADEInstituto de Artes Visuais, Design e Marketing, Av. D.Carlos I, no. 4, 1200, Lisboa, Portugal d ISCTE Business School-Lisbon, Av. das Foras Armadas 1649-026 Lisboa, Portugal Received 23 August 2005; received in revised form 2 July 2006; accepted 7 May 2007 Available online 8 August 2007

Abstract It is becoming increasingly important from both theoretical and managerial perspectives to measure Customer Relationship Management (CRM) as a key intangible asset. This paper seeks to bring relationship marketing theory into practice by developing a new measure of relationship performance between two firms, the business-to-business relationship performance (B2B-RELPERF) scale. Survey findings from a sample of approximately 400 purchasing managers operating in a B2B e-marketplace reveal that relationship performance is a high-order concept, composed of several distinct, yet related, dimensions: (1) relationship policies and practices, (2) relationship commitment; (3) trust in the relationship, (4) mutual cooperation; and (5) relationship satisfaction. Findings reveal that the B2B-RELPERF scale relates positively and significantly with customer loyalty. The paper also presents the B2B-RELPERF balanced scorecard, which combines tangible and intangible metrics. While existing IT solutions usually focus exclusively on the use of tangible CRM indicators, this new tool includes the voice of the customer. At the managerial level, both the scale and scorecard could act as useful instruments for short- and long-term management, controlling, planning, and improvement of B2B relationships. Implications for relationship marketing theory are also presented. 2007 Elsevier Inc. All rights reserved.
Keywords: Relationship marketing theory; Balanced scorecard; Electronic markets; e-Commerce; e-Marketplace; Performance measurement; CRM; B2B; Purchasing

1. Introduction If you do not measure it, then you cannot manage it! (Jack Welch, former CEO of General Electric) At the beginning of the 21st century it is widely accepted that existing Customer Relationship Management (CRM) solutions

Corresponding author. Faculdade de Economia, Universidade Nova de Lisboa, Campus de Campolide, 1099-032 Lisboa, Portugal. Tel.: +351 213801601; fax: +351 213886073. E-mail addresses: lflages@fe.unl.pt (L.F. Lages), alancastre@iade.pt (A. Lancastre), carmen.lages@iscte.pt (C. Lages). URL: http://prof.fe.unl.pt/~lflages (L.F. Lages). Editor's note: Three days after this paper was accepted for publication, Andrew Lancastre passed away due to leukemia. This issue of IMM is dedicated to his memory. 0019-8501/$ - see front matter 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.indmarman.2007.05.008

have much room for improvement. The main reason may be found in a statement by Einstein: everything that can be counted does not necessarily count; everything that counts cannot necessarily be counted. In order to measure what cannot necessarily be counted, existing CRM Information Technologies (IT) solutions employ tangible metrics to assess intangible dimensions (e.g. trust and cooperation). Many firm's intangible assets constitute unique opportunities for economic (added) value, such as customer relationships, and these can and should be scientifically assessed (Hunt, 1997). As managers and researchers observe that good versus poor relationships significantly affect business performance, there is an increasing concern with achieving a better understanding of relationship development with business partners (Lages, Lages, & Lages, 2005a; Lages, Lages, & Lages, 2005b; Lemon, White, & Winer, 2002). Although intangible metrics are of interest to academics and managers, these two worlds discuss the issue quite

L.F. Lages et al. / Industrial Marketing Management 37 (2008) 686697

687

differently (Melnyk, Stewart, & Swink, 2004; Likierman, 2004). While most managers do not have the scientific knowledge to develop reliable measures, most academic researchers are not concerned with the development of scientific metrics for application at the managerial level (Weick, 2001). Within this context, the development of business performance metrics requires the exploration of synergies between researchers and managers in order to develop scientific and reliable measures that might be of interest to practitioners, bridging the gap. Moreover, although marketing academics and practitioners have been examining relationship marketing since the mid-1980s, a significant criticism of most relationship marketing studies is the fact that many studies are based on a single dimension or a single financial indicator, intended to capture the nature of complex relationships between buyers and suppliers (Yau et al., 2000). A major priority for the upcoming years is the development of B2B metrics, namely in an e-commerce environment (Parasuraman, Zeithaml, & Malhotra, 2005; Parasuraman & Zinkhan, 2002). Despite both managers and academics' interest in understanding relationships in e-commerce, concerted efforts have not materialized (Grewal, Comer, & Mehta, 2001; Klein & Quelch, 1997). This article attempts to help bridge the gap by scientifically developing a new scale that enables, from a customer perspective, the assessment of relationship performance in business-to-business (B2B) relationships named the B2B-RELPERF scale. Furthermore, the authors use the B2B-RELPERF scale to suggest the development and testing of the respective relationship performance scorecard for inclusion in periodic business reports and/or existing CRM IT solutions. It is believed that the scale (and future scorecard) might help firms to administer resources more efficiently, by allocating them to different customers, and identifying deviations from objectives. Given the development of different customer relationship processes, this can also help a firm to establish its annual priorities in terms of marketing efforts. Moreover, a firm can use relationship performance metrics as a motivation and reward tool for managers and their teams (e.g., bonus, promotion) by relying on comprehensive data. Finally, these metrics can support the development, monitoring, improvement and benchmarking of customer relationship processes (see Lages et al., 2005a,b). This paper starts by presenting the five dimensions of the B2B-RELPERF scale. We then refine the preliminary scale using qualitative research and testing it through a field survey of approximately 400 purchasing managers of small and medium-sized enterprises (SMEs) in an e-marketplace. We also analyze the impact of the B2B-RELPERF scale on customers' loyalty intentions. Finally, we present implications for theory and practice, suggest a B2B-RELPERF scorecard, point out research limitations and define directions for further research. 2. The B2B-RELPERF scale A recent meta-analysis of relationship marketing literature (Palmatier, Dant, Grewal, & Evans, 2005) indicates that re-

search in the field should follow a multidimensional perspective because there is no single or best dimension able to capture the full essence of this phenomenon. Indication of the main constituents of a relationship process can be found in the literature. While building on past RM literature, the B2B-RELPERF scale reflects the performance of a buyersupplier relationship marketing process at a specific point in time. This scale is a higher-order construct composed of five distinct, yet related, dimensions: (1) relationship policies and practices; (2) trust in the relationship; (3) relationship commitment; (4) mutual cooperation; and (5) relationship satisfaction. Following the findings of Dwyer, Schurr and Oh (1987), supported some years later by Jap and Ganesan (2000), these five dimensions are key ingredients of a relationship process. Relationship policies and standards of conduct that mark a relational contract take form during the exploration phase, and go on building during the relationship process. The rudiments of trust and joint satisfaction established during this phase develop in the next stages of the process, leading to increased risk taking within the dyad. The expansion process that follows the exploration phase is a consequence of each party's satisfaction with the other's role performance. The resulting perceptions of goal congruence and cooperation lead to interactions beyond those that are strictly required at the outset, resulting in the parties purposefully committing resources to maintain and develop the relationship. These findings, supported by the relationship marketing literature (Anderson & Narus, 1990; Morgan & Hunt, 1994; Sirdeshmukh, Singh, & Sabol, 2002), suggest the significance of the identified five dimensions to relationship process performance. 2.1. Relationship policies and practices Relationship policies and practices represent one of the most important dimensions during a relationship process (Jap & Ganesan, 2000). By establishing clear relationship policies and practices, the supplier becomes motivated to behave in a way that is beneficial to the relationship as a whole, and as a result, emerging exchange partners start setting the ground rules for future exchange (Dwyer et al., 1987: 17). Despite the well-recognized significance of such policies, few studies have examined company behaviors and practices specifically or the mechanisms by which they strengthen a relationship (Sirdeshmukh et al., 2002). Nevertheless, it is recognized that relationship policies and practices should include ethical values, such as the supplier showing respect for the customer, because they contribute to the development of the relationship between firms and customers (Morgan & Hunt, 1994). Indeed, the extent to which partners have common beliefs about what behaviors and policies are important, appropriate, and right, is important to the process of relationship development. Strategic considerations motivate firms to serve their customers better, and technology is viewed as a means to build competitive advantage. In e-markets, strategic considerations, such as providing better customer service, are particularly significant because the virtual world increases the possibility of

688

L.F. Lages et al. / Industrial Marketing Management 37 (2008) 686697

suppliers acting opportunistically and customers perceiving augmented risk and uncertainty (Grewal et al., 2001). Thus, providing better service through quicker and easier problem solving represents an added relationship value in the electronic context and a contribution to the buyersupplier relationship. 2.2. Trust In interpersonal relationships trust is defined as the willingness to rely on another party and to take action in circumstances where such actions make one vulnerable to the other party (Doney, Cannon, & Mullen, 1998, 604). Trust has been extended to relations between organizations since interorganizational exchanges are made by individuals from each organization (Aulakh, Kotabe, Sahay, 1996). Thus, trust exists when one party has confidence in an exchange partner's reliability and integrity (Morgan & Hunt, 1994). Trust is an essential ingredient in the creation, development, and maintenance of long-term relationships between buyers and suppliers (Anderson & Narus, 1990; Ganesan, 1994). According to the findings of Palmatier et al. (2005), trust is a key dimension of (objective) performance. As trust is the cornerstone of the strategic partnership and relationship development process (Moorman, Deshpand, & Zaltman, 1993; Morgan & Hunt, 1994), it is widely accepted that trust plays a central role in the development of relationship performance (Dwyer et al., 1987; Morgan & Hunt, 1994). Trust becomes even more important for relationship and loyalty development purposes when perceived risk is more pronounced. Perceived risk in an e-commerce environment is high when compared to traditional commerce, due to spatial and temporal separation between buyers and sellers. The internet transaction typically takes place from different locations at different times, and goods or services are delivered after a seller has verified payment. When selecting a supplier, a buyer might consider the possibility that the supplier may be an expert in attracting customers and cashing credit cards but not in actually delivering goods or services (Smith, Bailey, & Brynjolfsson, 1999). Thus, given the higher perceived risks and uncertainty in the marketspace, trust may be even more critical in the buyersupplier relationship development process than in the traditional marketplace. 2.3. Relationship commitment

tomer-oriented assortment (Anderson & Weitz, 1992), and similarly buyers in an e-marketplace receive more relevant upto-date market and product information, a better assortment choice, and order/payment automation (Smith et al., 1999; Weiber & Kollman, 1998). Because both parties receive new benefits from each other, each has a stronger motivation to build, maintain and develop the relationship through renewed committed efforts (Kumar et al., 1995). Thus, strong relationships are built on the foundation of mutual commitment (Berry & Parasuraman, 1991: 139). In an organizational environment, commitment may be affective (attachment to an organization), continued (perceived cost of leaving an organization), or normative (perceived obligation to stay with an organization) (Meyer, Allen, & Smith, 1993). Of these three, only affective commitment influences the degree to which the customer wants to maintain a relationship with the firm (Roberts et al., 2003). Thus, in the electronic setting, we assume that relationship commitment is the buyer's attachment to the supplier and that it leads to the development of stable, long-term relationships. 2.4. Mutual cooperation Relationships require cooperative behavior (Morgan & Hunt, 1994). Cooperation between partners in a relationship exchange process grows as each partner perceives greater benefits from working together than from working independently. Thus, mutual cooperation is defined as similar or complementary coordinated actions taken by firms in interdependent relationships to achieve mutual outcomes or singular outcomes with expected reciprocation over time (Anderson & Narus, 1990: 45). The e-marketplace is a networked information system that serves as an enabling infrastructure for buyers and sellers to exchange information, transact, and perform other activities (Varadarajan & Yadav, 2002: 297). Mutual cooperation in an electronic environment is viewed in terms of the regularity of the interactions and the openness in communication activities between the buyer and the supplier (Hewett & Bearden, 2001; O'Keefe, O'Connor, & Jung, 1998), and represents an important element in relationship performance planning, management, and control. 2.5. Relationship satisfaction

Relationship commitment is defined as the desire to develop a stable relationship, a willingness to make short-term sacrifices to maintain the relationship, and a confidence in the stability of the relationship (Anderson & Weitz, 1992: 19). Commitment is essential for the development of long-term relationships (Anderson & Narus, 1990; Kumar, Scheer, & Steenkamp, 1995), and is an important indicator of both objective and relationship performance (Roberts, Varki, & Brodie, 2003). Moreover, relationship commitment is a means for differentiating successful relationships from unsuccessful ones (Morgan & Hunt, 1994). Suppliers in a committed relationship gain greater access to market information, which enables them to select a better cus-

Relationship satisfaction refers to the cognitive and affective evaluation based on personal experience across all episodes within the relationship (Roberts et al., 2003: 175). Thus, relationship satisfaction can be considered to be a positive emotional and rational state resulting from the assessment of the buyer's working relationship with the supplier (Geyskens, Steenkamp, & Kumar, 1999). Relationship satisfaction summarizes the customer's previous interactions with the supplier, which in turn influence expectations of future relationship development (Roberts et al., 2003). Customer evaluation of relationship satisfaction with the supplier is critical for the development of future business exchanges (Cannon &

L.F. Lages et al. / Industrial Marketing Management 37 (2008) 686697

689

Perreault, 1999). Positive interactions lead to long-term continuation of the relationship, thus emphasizing that relationship satisfaction is essential to the improvement of relationship performance. 3. Method 3.1. Unit of analysis and research setting The unit of analysis for this research is a specific buyer's relationship with a specific supplier. The analysis is conducted by examining the buyer's perspective. It was decided to collect customers' data because the customer ultimately decides whether to purchase from the supplier, and the customer's perception of the relationship is likely to predominantly determine its development and performance (Cannon & Perreault, 1999). The research setting has two main characteristics we believe to be worth examining when considering relationships: business-to-business and electronic. Business-to-business because recent research findings (Palmatier et al., 2005) reveal that customer relationships lead to higher levels of objective performance in business markets versus consumer markets. Electronic because the e-market environment might be perceived as risky and uncertain, and thus relationship development can be particularly relevant and difficult. Electronic business has added a new dimension to business relationship discussions (Morgan & Hunt, 2003). In an emarket environment, business is conducted at a distance, and risks and uncertainties are enhanced because the Internet, which is a relatively new and complex technology from the consumer's perspective, presents security problems. For example, online suppliers can easily register and track customer data, which increases the possibility for them to act opportunistically. Furthermore, many online buyers may not yet have accumulated the necessary shopping experience and relevant knowledge about potential market suppliers or other partners in this new shopping channel (Einwiller, Ingenhoff, & Schmid, 2003), which might undermine their confidence in using it. 3.2. The online business center The online business center pmelink.pt, selected for this research, sells office goods and services (e.g., office paper, consumables, computers, office furniture, recruitment services, customer credit reports) to SMEs in order to support their core businesses. We use pmelink.pt to analyze the extent to which it might be possible to monitor marketing relationship performance in a non-critical products' context. The purchasing literature suggests that non-critical items do not have a significant impact on the finances of most firms, representing a low supply risk to most B2B buyers (Kraljic, 1983). It has been argued that for routine products, value added strategies play a secondary role because the buyer normally relies on exploitation of full purchasing power, order volume optimization, and targeted pricing strategies. As long as key purchasing variables (e.g. lowest price and availability/delivery) are addressed by the seller, the buyers will become committed, and all other added-

value activities will become secondary.1 The pmelink.pt supplier delivers non-critical products to SMEs by focusing on added-value services, such as offering management relevant information, customer problem solving and advice, product portfolio access and payment convenience, customization and personalization, following retention and loyalty strategies. Interestingly, findings from a previous study (see Lancastre & Lages, 2006) revealed that although pmelink.pt sells routine, or at best non-critical products, the key determinants of buyers' cooperation (a key dimension in a relationship process) with this seller were essentially marketing relationship dimensions. Surprisingly, product prices ( .05, P b.05) resulted in a minor determinant of cooperation, and product delivery showed a nonsignificant association with that relationship dimension. Hence, by using a sample of pmelink.pt's business customers, it is our goal to better understand how it is possible for a firm to monitor and strengthen relationship performance with its customers even when dealing with non-critical products. Essentially, pmelink.pt operates among diverse businesses and suppliers: a customer places an order with pmelink.pt, which is forwarded to one of its 30 suppliers, with whom they have previously established contractual agreements; an express cargo carrier takes care of delivery logistics, and pmelink.pt bills the customer. The pmelink.pt online business center was formed when three major Portuguese groups (Portugal Telecom and two of the main national Banks) recognized an opportunity to market a variety of goods and services to their common client base. Portugal Telecom's market penetration is almost 100% and the other two partners together retain around 65% of all SMEs operating in the country as clients. The two banks realized that during their day-to-day relationships with their SME's customers, the SMEs were generally strong concerning their core businesses but relatively poor in providing support areas clerical, office supplies and purchasing, IT, marketing, logistics and so on vital for relationship development. The company pmelink.pt offered an opportunity through technology to offer those supporting tools to their SME's customers as well as to increase their knowledge and business with them. The online business center pmelink.pt uses an Internet and CRM electronic platform to formulate an integrated e-business infrastructure that allows the provision of adequate and reliable information, a variety of added-value services, and personalization. In addition to products, pmelink.pt offers a range of business expertise, advice and information through the same online connection. As an example, pmelink.pt realized that SMEs often had difficulty dealing with various legal requirements. In response, it developed a package of core services that includes a search engine for all types of legal documents, a simulator for various fees and taxes, a fiscal calendar with reminders of major dates, and a library of printable official forms. Not only does the supplier promise fast and efficient delivery of goods (pmelink.pt promises a 99% delivery success rate within a 24-h timeframe), it also leverages direct cost reductions to its clients through bulk ordering and strategic sourcing of materials from key suppliers.
1

We acknowledge an anonymous reviewer for this insight.

690

L.F. Lages et al. / Industrial Marketing Management 37 (2008) 686697

Monitoring user activities on the site allows individual visitors to be segmented, so that campaigns and newsletters can be targeted more effectively, and loyalty programs can be developed. Based on its unique relationship with its investors and customers, pmelink.pt aimed to contact and start business with around 10% of the 200,000 SMEs operating in Portugal by the end of its third year of existence. 3.3. Survey instrument development and test Scale development follows Churchill's (1979) traditional approach. On the basis of the literature review and preliminary findings, we specified the domain of the construct to include five Business-to-Business Relationship Performance (B2BRELPERF) dimensions relationship policies and practices, trust, relationship commitment, mutual cooperation, and relationship satisfaction. Fifty-two items were developed to measure these five dimensions based on CRM and electronic marketing literature. These items were then refined through indepth interviews with the five managers of pmelink.pt involved in the e-business operations. On the basis of their feedback, some items were removed so that the measures became better adjusted to the reality of e-markets. Three SME customers then assessed the final set of items of the B2B-RELPERF scale for content and face validity. To pre-test the B2B-RELPERF scale, data were collected through an initial online survey targeting 40 SMEs previously selected by the supplier pmelink.pt, considering loyal, regular and infrequent buyers. At the end of this process the scale included the 5 dimensions considered previously. As a consequence of the above-mentioned stages of the refinement process, the final solution included 38 of the 52 initial items. 3.4. Data collection The exploratory stage was followed by a survey based on an online questionnaire that we administered to a sample of the SME customers who were responsible for purchasing operations. In line with previous studies conducted by pmelink.pt, the online survey attached to the firm's periodic online newsletter, resulted in 395 valid questionnaires, which is greater than the minimum number (i.e., 381) required for a 95% confidence level. Stratified sampling was accomplished and controlled through the data collecting process, using the supplier's customers' loyalty degree strata grouping loyal, regular, infrequent and new buyers. 3.5. Sample profile and nonresponse bias Respondents were representative of the main industry and economic activities in the primary sector (5%), industrial sector (21%), and services sector (74%). They also regularly purchased main product categories (as classified by the supplier): paper (74%), consumable goods (73%), other office products (57%), systems equipment (29%), office furniture (5%), and services (6%). On the basis of the supplier database, we directed the survey to individuals who were primarily responsible for

the SMEs' buying centers. Although the job titles of the respondents ranged from general manager to financial manager, purchasing manager, and administrative manager, all of the respondents were the main person responsible for their SME's purchasing operations and most interacted with the supplier on a daily basis. In terms of profile, 20% of the respondent firms had fewer than 6 months of business experience with the supplier, 30% varied between 6 and 12 months, and the remaining 50% had more than 12 months of experience. In this last group, 70% had more than 2 years of experience. This indicates that though the title of the respondents' positions may be wide ranging, all appear to have significant knowledge about the specific purchasing activities of the firm. Although the respondents' profile suggests a high degree of confidence in the sample, we considered it important to test for possible nonresponse bias. By assessing the differences between the early and late respondents (Armstrong & Overton, 1977), we found no significant differences.2 Exploratory Factor Analysis (EFA) was conducted to examine the factor structure of each variable presented in the conceptual model. EFA is a data simplification technique particularly useful for reducing the number of indicators to a manageable set. EFA with principal components analysis (PCA) and varimax rotation (when Kaiser's rule is used for factor selection) was performed on the survey data. Varimax rotation is the most widely used analytical format to analyze orthogonal factors. This method was selected because although the subdimensions may be related, the relationships cannot be anticipated a priori. The use of an orthogonal rotation has the advantage of allowing a more intuitive interpretation of the results (Hair, Anderson, Tatham, & Black, 1998). Coefficient alpha was also computed for the emerged factors in order to assess the validity of the initial scales and further purify the number of items for each factor. Five items were removed during this initial process, two from the factor relationship policies and practices (Sirdeshmukh et al., 2002) and three from the factor relationship commitment (Anderson & Weitz, 1992). The other 19 items were removed as a consequence of CFA, Fornell and Larcker's (1981) test of variance extracted, and Bagozzi's (1980) test of composite reliability. 4. Data analysis 4.1. CFA CFA was performed to assess the measurement properties of the scales, using full-information maximum likelihood estimation procedures in LISREL 8.3 (Jreskog & Srbom, 1993). In this model, each item is restricted to load on its prespecified factor, with the five first-order factors correlating freely. The chi-square for this model is significant (2 = 143.58, d.f. = 67, P b.05). Because the chi-square is sensitive to sample size, we

2 Early respondents were defined as the first 75% of the returned questionnaires, and the last 25% were considered to be late respondents. These proportions approximate the actual way in which questionnaires were filled in.

L.F. Lages et al. / Industrial Marketing Management 37 (2008) 686697

691

also assessed additional fit indices: (1) non-normed fit index (NNFI), (2) comparative fit index (CFI), incremental fit index (IFI), and standardized root mean square residual (SRMR). The NNFI, CFI, and IFI, of this model are .99, and the SRMR is .41. Because fit indices can be improved by allowing more terms to be freely estimated, we also assessed the root mean square error of approximation (RMSEA), which, for our measurement model, is .054. A full listing of the 14 final items and their scale reliabilities appear in Table 1. The average internal reliability (Cronbach ) was .86. The significant standardized loadings of each item on its intended construct (average loading size was .83) evidenced unidimensionality. Table 1 shows that all constructs present the desirable levels of composite reliability (Bagozzi, 1980).

Table 1 The B2B-RELPERF scale: scale dimensions and items, reliabilities, and variance extracted Please rate your agreement with each of the following statements, regarding your relationship with Firm X: RPP: V1 V2 V3 RCO: V4 V5 V6 Relationship policies and practices (Adapted from Sirdeshmukh et al., 2002) Firm X has policies that show respect for the customer. Firm X has practices that make solving problems easy. Firm X solves my firm's problems quickly. / vc(n) /

.81/.61/.82

Relationship commitment .86/.69/.87 (Adapted from Kumar et al., 1995) Our relationship with Firm X is a long-term partnership. We would not drop Firm X because we like being associated with it. We want to remain as a customer of Firm X because we have pride in being associated with a firm that carries a technological image. .91/.78/.91

TRUST: Trust in the relationship (Adapted from Morgan & Hunt, 1994) In our relationship, Firm X V7 is someone to whom I give my confidence. V8 has high integrity. V9 gives us reliable information and advice. MCO: V10 V11 SAT: V12 V13 V14 Mutual cooperation (Adapted from Hewett & Bearden, 2001) My firm and Firm X regularly interact. There is an open communication between our firms. Relationship satisfaction (Adapted from Cannon & Perreault, 1999) Overall, we are satisfied with Firm X. We are pleased with what Firm X does for us. If we had to do it again, we would still choose to use Firm X.

.82/.71/.83

.88/.71/.88

Table 1 also shows that Fornell and Larcker's (1981) index of variance extracted was greater than the recommended level of .50 for all five constructs. The construct intercorrelations, which are significantly different from 1, reveal evidence of discriminant validity, and the shared variance between any two constructs (i.e., the square of their intercorrelation) is less than the average variance explained in the items by the construct (Fornell & Larcker, 1981; MacKenzie, Podsakoff, & Rich, 1999). The largest squared multiple correlation between any two constructs was .53 (.73 was the highest correlation [between satisfaction and commitment and between satisfaction and trust]), whereas the variance extracted ranged from .61 to .78. To assess nomological validity, the measures were tested with respect to a dimension to which the constructs are supposed to be theoretically related (Churchill, 1995). Expecting a positive relationship between relationship performance and loyalty is well grounded in theoretical reasoning. Buyer loyalty is the intention to perform a set of behaviors that indicate a motivation to maintain a relationship with a supplier, including allocating a higher share of wallet, engaging in positive word of mouth, and repeat purchasing (Zeithaml, Berry, & Parasuraman, 1996). When suppliers are oriented to the relationship and act in a way that strengthens relationship policies and practices, trust, relationship commitment, cooperative activities, and satisfaction, the perceived risk with the specific service supplier is likely to be reduced, enabling the customers to make confident predictions about the supplier's future behaviors (Morgan & Hunt, 1994). This helps shape the customers' perceptions about their relationship with suppliers, and enhances their loyalty (Sirdeshmukh et al., 2002). According to recent findings (Palmatier et al., 2005), relationship marketing may build customer loyalty more effectively in this type of research setting (i.e. service offering and business markets). We used three items for measuring buyer loyalty ( = .93): (1) the buyer's intention to make most future purchases from the supplier, (2) the buyer's intention to recommend the supplier to other firms, and (3) the buyer's intention to use the supplier the next time he/she needs to purchase products or services. We measure all items on 7-point Likert scales, anchored by strongly disagree and strongly agree (Sirdeshmukh et al., 2002). We found that loyalty is positively correlated with relationship orientation (r = .50), relationship commitment (r = .64), trust (r = .53), mutual cooperation (r = .58), and relationship satisfaction (r = .80). Given that all of the coefficients are positive and significant (P b.01 or better), we conclude that the performance of the buyerseller relationship has a positive impact on loyalty. Thus, we have support for the nomological validity of the five proposed measures (Cross & Chaffin, 1982). 4.2. Higher-order factor We also estimate a second-order factor model of B2BRELPERF. In addition to obtaining respondents' evaluations on the five dimensions, the second-order factor model captures the common variance among them, which reflects an assessment of the B2B relationship performance process. This model includes

Notes: Firm X = The supplier (i.e., pmelink.pt). For purposes of replication of this study in other types of B2B relationships, Firm X may be considered any type of firm (e.g., a manufacturer, buyer, distributor, franchisor, franchisee, exporter, importer, multinational). Cronbach's = internal reliability (Cronbach, 1951); vc(n) = variance extracted (Fornell & Larcker, 1981); and = composite reliability (Bagozzi, 1980). All items are 7-point Likert scales, anchored by strongly disagree and strongly agree.

692

L.F. Lages et al. / Industrial Marketing Management 37 (2008) 686697

Fig. 1. The B2B-RELPERF scale: CFA standardized coefficients for higher-order model. Legend: RP:Relationship performance; RPP:Relationship policies and practices; RCO:Relationship commitment; TRUST:Trust in the relationship; MCO:Mutual cooperation; SAT:Relationship satisfaction.

the five first-order factors along with their standardized coefficients, observable indicators, and measurement errors (see Fig. 1). Each of the first-order factors has significant (P b.01) loadings of .54, .80, .79, .74, and .94, respectively, on the second-order factor. Although the chi-square for the secondorder model is significant (2 = 152.66, d.f. = 72, P b.05), the NNFI, CFI, and IFI are .99; the SRMR is .41; the RMSEA is .053; and the chi-square difference test between the first-order and second-order models is not significant (2 = 9.08, d.f. = 5, P N.10). Overall, this suggests that the higher-order model accounted for the data well. Further evidence is demonstrated in the correlations between the five constructs. All correlations are significant (P b.01), and the coefficients are large and positive, indicating that the five scales converge on a common underlying construct (Cadogan, Diamantopoulos, & de Mortanges, 1999; Lages & Fernandes, 2005). 5. Main findings and discussion In this study we propose the B2B-RELPERF scale as a broad conceptualization and specific operationalization of relationship performance. Although we cannot claim to have captured the dimensions of this concept fully, we argue that we have come close to doing so because the second-order factor extracts the underlying commonality among the five dimensions, as CFA results suggest. The hierarchical structure of the B2BRELPERF scale reveals that relationship performance is composed of five elements: relationship orientation, relationship commitment, trust, mutual cooperation and relationship

satisfaction. The B2B-RELPERF scale reflects an entire set of relationship marketing metrics that might be influenced and, to a certain extent, are within the control of management. For example, our findings reveal that all five dimensions are positively correlated with loyalty, which measures the buyer's intention to repeat and make most future purchases from the supplier, beyond recommending it to other firms. Moreover, CFA results suggest that the second-order factor model captures the common variance among the five dimensions, which reflects an assessment of the B2B relationship performance process, following retention and loyalty strategic orientations. Earlier studies on B2B relationship determinants (e.g. Lancastre & Lages, 2006) suggest that relationship marketing activities offer added-value to the customer even when selling routine products and are worth investing in as they promote customer loyalty. Moreover, the non-critical products' context represents an excellent opportunity to develop cross-selling and up-grading business activities with the buyers, outcomes that normally follow the implementation of retention and loyalty strategies. Development of tools to assess the performance of a long-term relationship process between two firms becomes crucial for managers to better understand and efficiently handle their relationships. Moreover, a gap between academics and practitioners has been identified (see 2001 special issue of the British Journal of Management). By developing the B2BRELPERF scale, we are striving to help bridge that gap. We believe that such a research tool helps to enhance relationship performance at the individual customer level in order to better guide the respective relationship effort.

L.F. Lages et al. / Industrial Marketing Management 37 (2008) 686697

693

One of the implications of these concerns is the suggestion that from a managerial perspective, the possibility of grouping the 5 dimensions into a single score allows managers to have a starting point to more easily recognize and monitor the overall relationship performance with each customer. Indeed, the function of the customer relationship development process is to build relationships with preferred customers (Knox, 1998). Not all customers are willing to join a relationship process with the supplier, nor are they all-important for the supplier to add value to his business (Christy, Oliver, & Penn, 1996; Grnroos, 1997). In fact, the supplier may expend significant sums of money to retain an unprofitable customer, or to establish and maintain a relationship when the customer does not wish for one. Segmentation strategies, based on customer requirements, profitability, and customer loyalty are essential (Knox, 1998; Payne & Frow, 1999). This will lead to more productive marketing activities by developing and increasing individual customer loyalty (Palmatier et al., 2005). This research develops a buyerseller relationship measure in an e-business and routine products context. The e-marketplace is a particularly interesting environment in which to develop these metrics because e-business relationships are conducted at a distance, and risks and uncertainties are magnified. Despite purchasing literature suggestions that the buyer normally relies on order volume optimization and pricing strategies, the routine and non-critical products context represents an opportunity for the supplier to follow added-value relationship strategies, in order to develop up-selling, up-grading and crossselling of its products and services. Moreover, because the relationship process develops through various stages in which different dimensions contribute in different ways during the process (Dwyer et al., 1987; Jap & Ganesan, 2000) we suggest that the use of a balanced scorecard that weights dimensions differently may capture this process. 6. The B2B-RELPERF scorecard At a time when B2B relationships are instrumental in the determination of enterprises' value and performance, the B2BRELPERF scale has the potential to help practitioners plan, manage, monitor, and improve their ongoing B2B relationships if properly integrated with objective metrics through a balanced scorecard often available through existing Enterprise Resource Planning (ERP) and traditional CRM solutions (see Appendix).3 This approach of matching objective and subjective metrics has the potential to provide a better understanding of relationship performance constituents, and to increase return on relationship marketing investments (see: Bolton, 2004; Lehmann, 2004). Managers are often judged not only by the performance of different relationship processes, but also by the priorities they assign to the relationships. The recommendation is to compare

each current year (Y) weighted score with that of the previous year (Y 1) because problems are easier to detect by the size of the gap between a current year's metrics and the base score from the previous year. This annual feedback enables marketers to make corrections and helps them define goals and priorities for the relationships in the following year. To define each relationship process goal and degree of priority, firms might also consider existing objective metrics per customer and industry benchmarks, such as sales or margins by product category, frequency of sales and respective deviation standard, promotion sales and costs (in the case of pmelink.pt). Another possibility is to set sub-goal intervals for the B2B-RELPERF scorecard (e.g. on a quarterly or half-year basis). The major advantage of this is that quarterly or half-year feedback would enable marketers to review relationship performance trends and to make corrections more frequently, in response to the changing environment (Abernathy, 1997). In this way, periodic scheduled reviews with the B2B-RELPERF scorecard could be extremely useful in monitoring and improving relationship and loyalty strategies. Regarding the B2B-RELPERF scorecard development and implementation, data from longitudinal studies are required to test it. We present some recommendations below. The first step is to enter under customer description the name of the customers to be analyzed. Second, under the objective metrics column(s), identify existing financial or other objective metrics (e.g. sales, profits, sales frequency, operational costs, lifetime value) and establish a ranking for each customer. Then, ask customers to assess the 14 items in the B2B-RELPERF scale on a 7-point Likert scale (see Table 1). The next step is to assess the average score for each of the five B2B-RELPERF dimensions and multiply them by their weights. These results should be presented under the relationship performance dimensions column. The relationship process develops through mutual learning stages in which relationship policies and practices, trust, relationship commitment, mutual cooperation, and relationship satisfaction are critical issues. Depending on the relationship stage, different factors might present different weightings. Weightings should differ across the various customers' relationship processes, depending on the different phases of relationship development. Hence, it is important that the management team agrees on the weightings of each dimension before its implementation and future assessment. The sum of the five weighted dimensions represents the final RPScore per customer. The RPWeightedScore is the result of multiplying the final RPScore by the weight of each customer to overall relationship performance. Each RP weight to overall relationship performance must take into consideration different factors, such as the firm's mission and objectives, the firm's strategy, and different insights from managers. A final RELPERFScore results from the sum of all individual RPWeightedScores. 7. Limitations and further relationship marketing research Although this study provides a number of new insights, it is important to note its limitations. A first limitation regards the specificities of this study's research context and respondents Portuguese SMEs' relationships with an internet-based supplier.

3 An IT solution is also being developed while using as a basis the B2BRELPERF Scale and the B2B-RELPERF Scorecard. For further information please contact the first author.

694

L.F. Lages et al. / Industrial Marketing Management 37 (2008) 686697

While enhancing the focus of this research, it may limit the generalization of the results to some degree, at the same time creating the need for further research, in this environment and other contexts. Another limitation results from the fact that research relies only on the responses of the buyers. Ideally, researchers should collect data from both sides of the dyad. However, as research in relationship marketing is still at an early stage of development, we begin with the buyer's view to better understand the components of B2B relationship performance, as in many situations the buyer is the one that decides if he wishes or not to continue the relationship, while evaluating the addedvalue that results from his relationship with the supplier. A third limitation is that our research instrument (i.e., the questionnaire) may have created common method variance. This could be particularly problematic if the respondents were aware of the conceptual framework of interest. However, we did not explain the specific purpose of the study to them, and we mixed and separated the construct items (Jap, 2001; Lages & Montgomery, 2005). Furthermore, we guaranteed confidentiality to all survey participants for self-presentation reasons, which also helps to reduce the possibility of bias in issues such as relationship policies and practices, trust, commitment, cooperation, and satisfaction (Singh, 2000). In addition, if common method bias exists, a CFA containing all constructs should produce a singlemethod factor (Podsakoff and Organ, 1986). The goodness-offit indices (NNFI = .85, CFI = .87, IFI = .87, and RMSEA = .191) indicate a poor fit, which suggests that biasing from common method variance is unlikely (Lages and Lages, 2004). Hence, due to the potential pitfalls of this tool, our efforts to develop the B2B-RELPERF scale should be seen as a starting point, considered as part of a larger effort to help managers and researchers capture relationship performance. Nevertheless, future operationalization of the B2B-RELPERF scorecard is encouraged, through a combination of objective and intangible metrics, as earlier discussed, to help overcome the issue of common method bias.4 This study provides a foundation for significant research endeavors to advance the field. Additional research is required that analyzes the antecedents and consequences of the B2BRELPERF scale (see Lages et al., in press). In addition to its relationship with loyalty, further research must investigate how this scale is related to other established constructs in the relationship marketing field, such as transaction-specific investments. We recommend further research that tests the B2B-RELPERF scale on the suppliers' side and on both sides of the dyad. Consider testing the scale in different settings and in a wide range of activities, such as exporterimporter and franchiserfranchisee relationships, and whether findings are different depending on the buying situation of the customers or on the customer profile. Relationships in an international context transcend national boundaries, and are particularly complex because they are affected by socio-cultural, and other environmental differences. When applying the scale to different

contexts, we encourage researchers to add new items and factors in order to continue refining the B2B-RELPERF scale. 8. Conclusions In this economic environment when corporate budgets are being squeezed, Chief Marketing Officers are kept up at night by worry, trying to justify their expenditures and their existence. They believe that what they are doing has value, and they have to figure out how to demonstrate that value to skeptical CEOs and CFOs (Reibstein, 2004). As a direct response to a recent observation in the literature (Morgan & Hunt, 2003), we hope that this paper will help cultivate knowledge on relationship management theory while shedding light on B2B relationships that are supported by new information and communication technologies. Moreover, at a time when researchers are challenged to present studies with managerial implications (Marketing Science Institute, 2004), the B2B-RELPERF scale and the B2B-RELPERF scorecard might be used to address the managerial needs of B2B relationship performance planning, implementation, and control. By using the B2B-RELPERF scale and the B2B-RELPERF scorecard to assess the performance of a buyersupplier relationship process, managers can better understand the constituent elements of the relationship process, which in turn aids managers in selecting, using, and controlling the most adequate marketing tools for each of these elements. These tools may also help managers understand differences that exist in the relationship process development phases among different customers or groups and can enable managers to handle such differences more efficiently and effectively. Moreover, by defining actions that address potential problems during the relationship marketing process development, managers can ultimately influence their firm's relationship orientation, retention, and loyalty strategies. Finally, by balancing and complementing financial and other operational measures with intangible customer metrics, the B2B-RELPERF balanced scorecard allows for better monitoring of relationship performance at both the customer and corporate levels, while helping to focus the entire organization on long-term strategy and improving shareholder value (Kaplan & Norton, 1992, 1996, 2000). As a whole, in addition to addressing the relationship marketing area, this paper may also enrich knowledge in the e-commerce and purchasing literature. Acknowledgements This research was funded by the following research grants to the first author: NOVA FORUM, NOVA EGIDE, and 6th European Framework Program while affiliated with Universidade Nova de Lisboa; FCT-EU while a Visiting Scholar at London Business School and MIT's Deshpande Center for Technological Innovation. Andrew Lancastre thanks UNIDCOM/IADE, and Carmen Lages thanks UNIDE/ISCTE. The authors acknowledge IMM editor, three IMM anonymous reviewers, and the participants and reviewers of 2005 CRM (Russia), 2005 ICRM (Canada) and 2007 EMAC (Iceland) conferences for comments on earlier versions of this manuscript.

We acknowledge an anonymous reviewer for this insight.

Appendix A
L.F. Lages et al. / Industrial Marketing Management 37 (2008) 686697 The B2B-RELPERF Scorecard Customer description Objective metrics Subjective metrics Relationship performance dimensions Objective metric
a

Relationship performance MCOScore: 2-item average weight (%) ___ ___ ___ ___ ___ ___ SATScore: 3-item average weight (%) ___ ___ ___ ___ ___ ___ _______ _______ _______ RPScore Weight to overall relationship perform.(%) = RPWeightedScore

Year Y 1 Base(s)
b

Year Y + 1 Goal(s)b Priorityb

Ranking

RPPScore: 3-item average weight (%)

RCOScore: 3-item average weight (%) ___ ___ ___ ___ ___ ___

TRUSTScore: 3-item average weight (%) ___ ___ ___ ___ ___ ___

Customer1 Customer2 Customer3 Customern

_____ _____ _____

____ ____ ____

____ ____ ____ .

___ ___ ___ ___ ___ ___

_______ _______ _______ Sum =

=_______ =_______ =_______ RELPERFScore

______ ______ ______

______ ______ _____

______ ______ _____

Notes: RPScore = RP1WeightedScore + RP2WeightedScore + RP3WeightedScore + RP4WeightedScore + RPnWeightedScore RPScore = RPPScore + TRUSTScore + RCOScore + MCOScore + SATScore RELPERFScore = overall relationships performance, RPP = relationship policies and practices, RCO = relationship commitment, TRUST = trust in the relationship, MCO = mutual cooperation, and SAT = relationship satisfaction. a This column should be repeated according to the number of existing objective performance metrics (e.g., sales volume, profit, costs). If more than one objective metric is available, a final ranking of objective relationship performance should be created while taking into consideration different weights for each objective metric. b These fields should take into consideration the RPWeightedScore and, if possible, existing objective metric(s).

695

696

L.F. Lages et al. / Industrial Marketing Management 37 (2008) 686697 Jreskog, K. G., & Srbom, D. (1993). LISREL 8: Structural equation modeling with the SIMPLIS command language. Hillsdale, NJ: Lawrence Erlbaum Associates. Kaplan, R. S., & Norton, D. (1992). The balanced scorecard-measures that drive performance. Harvard Business Review, 70(1), 7179. Kaplan, R. S., & Norton, D. (1996). The balanced scorecard. Cambridge, MA: Harvard Business School Press. Kaplan, R. S., & Norton, D. (2000, September-October). Having trouble with your strategy? Then map it. Harvard Business Review, 167176. Klein, L., & Quelch, J. (1997). Business-to-business market making on the internet. International Marketing Review, 14(5), 345361. Knox, S. (1998, December). Loyalty-based segmentation and the customer development process. European Management Journal, 16(8), 729737. Kraljic, P. (1983, September-October). Purchasing must become supply management. Harvard Business Review, 109117. Kumar, N., Scheer, L. K., & Steenkamp, J. E. (1995, February). The effects of supplier fairness on vulnerable resellers. Journal of Marketing Research, 32, 565. Lages, L. F., & Lages, C. R. (2004). The STEP scale: A measure of short-term export performance improvement. Journal of International Marketing, 12 (1), 3656. Lages, L. F., & Fernandes, J. (2005). The SERPVAL scale: A multi-item instrument for measuring service personal values. Journal of Business Research, 58(11), 15621572. Lages, L. F., & Montgomery, D. B. (2005). The relationship between export assistance and performance improvement in Portuguese export ventures: An empirical testing of the mediating role of pricing strategy adaptation. European Journal of Marketing, 39(7/8), 755784. Lages, C., Lages, C. R., & Lages, L. F. (2005). The RELQUAL scale: A measure of relationship quality in export market ventures. Journal of Business Research, 58(8), 10401048. Lages, L. F., Lages, C., & Lages, C. R. (2005). Bringing export performance metrics into annual reports: The APEV scale and the PERFEX scorecard. Journal of International Marketing, 13(3), 79104. Lages, L.F., Jap, S.D., Griffith, D.A., (in press). The role of past performance in export ventures: A short-term reactive approach. Journal of International Business Studies. Lancastre, A., & Lages, L. F. (2006). The relationship between buyer and a B2B e-marketplace: Cooperation determinants in an electronic market context. Industrial Marketing Management, 35(6), 774789. Lehmann, D. R. (2004, Oct). Metrics for making marketing matter, MSI Special Section. Journal of Marketing, 68, 7375. Lemon, K. N., White, T. B., & Winer, R. S. (2002). Dynamic customer relationship management: Incorporating future considerations into the service retention decision. Journal of Marketing, 66(1), 114. Likierman, A. (2004). Ideas and practice: Why academic should not mean irrelevant.. Keynote speech at the 2004 Conference of the Performance Measurement Association, Centre for Business Performance Edinburgh, UK: Performance Measurement Association. MacKenzie, S. B., Podsakoff, P. M., & Rich, G. A. (1999). Transformational and transactional leadership and salesperson performance. Journal of the Academy of Marketing Science, 29(2), 115134. Marketing Science Institute, 20022004 research priorities. 2004 (http://www. msi.org/msi/rp0204.cfm#Overview). Melnyk, S. A., Stewart, D. M., & Swink, M. (2004). Metrics and performance measurement in operations management: Dealing with the metrics maze. Journal of Operations Management, 22(3), 209217. Meyer, J., Allen, N., & Smith, C. (1993). Commitment to organizations and occupations: Extension and test of a three-component conceptualization. Journal of Applied Psychology, 78(4), 538551. Moorman, C., Deshpand, R., & Zaltman, G. (1993, January). Factors affecting trust in market research relationships. Journal of Marketing, 57, 81101. Morgan, R, & Hunt, S. (2003). An interview with Rob Morgan and Shelby Hunt. (http://www.incites.com/papersMMorgan_n_Hunt.html) accessed May 1, 2004. Morgan, R. M., & Hunt, S. D. (1994, July). The commitmenttrust theory of relationship marketing. Journal of Marketing, 58, 2038. O'Keefe, R., O'Connor, G., & Jung, H. (1998). Early adopters of the Web as a retail medium: Small company winners and losers. European Journal of Marketing, 32(7/8), 629643.

References
Abernathy, W. (1997). Balanced scorecards make teamwork a reality. Journal for Quality and Participation, 20(5), 5859. Anderson, J. C., & Narus, J. A. (1990, January). A model of distributor firm and manufacturer firm working partnerships. Journal of Marketing, 54, 4258. Anderson, E., & Weitz, B. (1992, February). The use of pledges to build and sustain commitment in distribution channels. Journal of Marketing Research, 29, 1834. Armstrong, J., & Overton, T. (1977, August). Estimating nonresponse bias in mail surveys. Journal of Marketing Research, 14, 396402. Aulakh, P., Kotabe, M., & Sahay, A. (1996). Trust and performance in crossborder marketing partnerships: A behavioural approach. Journal of International Business Studies, 27(5), 10051032. Bagozzi, R. P. (1980). Causal models in marketing. New York: John Wiley. Berry, L., & Parasuraman, A. (1991). Marketing services: Competing through quality. New York: The Free Press. Bolton, R. N. (2004, Oct). Linking marketing to financial performance and firm value, MSI Special Section. Journal of Marketing, 68, 7375. Cadogan, J. W., Diamantopoulos, A., & de Mortanges, C. P. (1999). A measure of export market orientation: Scale development and cross-cultural validation. Journal of International Business Studies, 30(4), 689707. Cannon, J., & Perreault, W. (1999, November). Buyerseller relationships in business markets. Journal of Marketing Research, 36, 439460. Churchill, G. A., Jr. (1979, February). A paradigm for developing better measures of marketing constructs. Journal of Marketing Research, 16, 6473. Churchill, G. A., Jr. (1995). Marketing research: Methodological foundations. Chicago: Dryden Press. Christy, R., Oliver, G., & Penn, J. (1996). Relationship marketing in consumer markets. Journal of Marketing Management, 12, 175187. Cronbach, L. J. (1951, September). Coefficient alpha and the internal structure of tests. Psychometrika, 16, 297334. Cross, E. M., & Chaffin, W. W. (1982). Use of the binomial theorem in interpreting results of multiple tests of significance. Educational and Psychological Measurement, 42, 2534. Doney, P. M., Cannon, J., & Mullen, M. (1998). Understanding the influence of national culture on the development of trust. Academy of Management Review, 23(3), 602621. Dwyer, F., Schurr, P., & Oh, S. (1987, April). Developing buyerseller relationships. Journal of Marketing, 51, 1127. Einwiller, S., Ingenhoff, D., & Schmid, B. (2003). A model of trust and reputation in electronic commerce. Proceedings from the EMAC 32nd Conference Glasgow: University of Strathclyde. Fornell, C. D., & Larcker, F. (1981, February). Evaluating structural equation models with unobservable variables and measurement errors. Journal of Marketing Research, 18, 3950. Ganesan, S. (1994, April). Determinants of long-term orientation in buyerseller relationships. Journal of Marketing, 58, 119. Geyskens, I., Steenkamp, J., & Kumar, N. (1999, May). A meta-analysis of satisfaction in marketing channel relationships. Journal of Marketing Research, 36, 223238. Grewal, R., Comer, J., & Mehta, R. (2001, July). An investigation into the antecedents of organizational participation in business-to-business electronic markets. Journal of Marketing, 65, 1733. Grnroos, C. (1997). Value-driven relational marketing: From products to resources and competencies. Journal of Marketing Management, 13, 407419. Hair, J., Jr., Anderson, R., Tatham, R., & Black, W. (1998). Multivariate data analysis, 5th. New York: Prentice Hall International. Hewett, K., & Bearden, W. (2001, October). Dependence, trust, and relational behavior on the part of foreign subsidiary marketing operations: Implications for managing global marketing operations. Journal of Marketing, 65, 5166. Hunt, S. (1997). Competing through relationships: Grounding relationship marketing in resource-advantage theory. Journal of Marketing Management, 13, 431445. Jap, S. D. (2001, February). Pie sharing in complex collaboration contexts. Journal of Marketing Research, 38, 8699. Jap, S. D., & Ganesan, S. (2000). Control mechanisms and the relationship life cycle: Implications for safeguarding specific investments and developing commitment. Journal of Marketing Research, 37(2), 227245.

L.F. Lages et al. / Industrial Marketing Management 37 (2008) 686697 Palmatier, R. W., Dant, R. P., Grewal, D., & Evans, K. (2005). A meta-analysis of the nomological network surrounding relationship marketing. Cambridge, Mass.: Marketing Science Institute. Parasuraman, A., Zeithaml, V., & Malhotra, A. (2005). E-S-QUAL: A multipleitem scale for assessing electronic service quality. Journal of Service Research, 7(3), 213233. Parasuraman, A., & Zinkhan, G. (2002). Marketing to and serving customers through the Internet: An overview and research agenda. Journal of the Academy of Marketing Science, 30(4), 286295. Payne, A., & Frow, P. (1999). Developing a segmented service strategy: Improving measurement in relationship marketing. Journal of Marketing Management, 15, 797818. Podsakoff, P. M., & Organ, D. W. (1986). Self-reports in organizational research: Problems and perspectives. Journal of Management, 12(4), 531544. Reibstein, D. (2004). Connecting marketing metrics to financial consequences. Speech given at the Wharton Marketing Conference, 18 October 2004 http:// knowledge.wharton.openn.edu/ Roberts, K., Varki, S., & Brodie, R. (2003). Measuring the quality of relationships in consumer services: An empirical study. European Journal of Marketing, 37(1/2), 169196. Singh, J. (2000, April). Performance productivity and quality of frontline employees in service organizations. Journal of Marketing, 64, 1534. Sirdeshmukh, J., Singh, J., & Sabol, B. (2002, January). Consumer trust, value, and loyalty in relational exchanges. Journal of Marketing, 66, 1537. Smith, M., Bailey, J., & Brynjolfsson, E. (1999, March). Understanding digital markets: Review and assessment. eBusiness Center, MIT Sloan School of Management, vol. 140 http://ebusiness.mit.edu Varadarajan, P. R., & Yadav, M. (2002). Marketing strategy and the Internet: An organizing framework. Journal of the Academy of Marketing Science, 30 (4), 296312. Weiber, R., & Kollman, T. (1998). Competitive advantages in virtual markets: Perspectives of information-based marketing in cyberspace. European Journal of Marketing, 32(7/8), 603615.

697

Weick, K. E. (2001). Gapping the relevance bridge: Fashions meet fundamentals in management research.British Journal of Management, 12, 7175 Special Issue. Yau, O., McFetridge, P., Chow, R., Lee, J., Sin, L., & Tse, A. (2000). Is relationship marketing for everyone? European Journal of Marketing, 34(9/10), 11111127. Zeithaml, V. A., Berry, L. L., & Parasuraman, A. (1996). The behavioral consequences of service quality. Journal of Marketing, 60(2), 3146. Luis Filipe Lages (PhD, Warwick) is Associate Professor of Marketing and International Business at Faculdade de Economia, Universidade Nova Lisboa, Portugal and Visiting Scholar at MIT's Deshpande Center for Technological Innovation, USA. His research interests include measurement and monitoring of intangibles, international marketing, innovation/disinnovation strategy, and technology-market transfer. His publications appeared in Industrial Marketing Management, Journal of International Business Studies, Journal of International Marketing, Journal of Business Research, European Journal of Marketing, MSI Reports, among others. Andrew Lancastre (PhD, ISCTE) is Associate Professor of Marketing at IADEInstituto de Artes Visuais, Design e Marketing in Lisbon, and Visiting Professor of Marketing Strategy at Faculdade Economia, Universidade Nova Lisboa, Portugal. Previously, he was a marketing manager at British Petroleum and Renault. Main areas of interest cover relationship marketing, B2B activities, SMEs, and international marketing. His work is published in Industrial Marketing Management and in the proceedings of leading conferences in the marketing field. Carmen Lages (PhD, Warwick) is Assistant Professor of Marketing at ISCTE Business School-Lisbon, Portugal. Her research interests include relationship marketing, social marketing, and marketing communications. She has published in the Journal of Business Research, European Journal of Marketing, Journal of International Marketing, among others.