GEORGE R.

FRANKE and JEONG-EUN PARK*

The authors combine findings from 155 samples of more than 31,000
salespeople to test alternative models of antecedents and consequences
of adaptive selling behavior (ASB) and customer orientation (CO). A
random-effects meta-analysis yields average values for 28 different
correlations ranging from –.16 to .35, 19 of which are significant.
Controlling for salesperson gender and selling experience, structural
equation modeling indicates that ASB increases self-rated, manager-
rated, and objective measures of performance, whereas CO increases
only self-rated performance. Both ASB and CO increase job satisfaction.
Tests of reciprocal relationships indicate that ASB increases CO and job
satisfaction increases performance rather than vice versa. Selling
experience increases performance but not job satisfaction, and
saleswomen rate their performance and satisfaction slightly higher than
salesmen do. The magnitudes of the relationships indicate that ASB and
selling experience have greater effects than CO and gender on
salesperson performance.

Salesperson Adaptive Selling Behavior and

Customer Orientation: A Meta-Analysis

Understanding the characteristics of effective salespeople
has been a long-standing goal of managers and researchers.
Quantitative syntheses of hundreds of empirical studies
indicate that personal characteristics, role perceptions, and
job attributes typically account for 10% or less of the vari-
ance in salesperson performance and job satisfaction (e.g.,
Brown and Peterson 1993; Churchill et al. 1985). Therefore,
identifying additional useful predictors could prove helpful
in selecting, training, and managing salespeople.
Two salesperson characteristics that have been the focus
of prominent research streams in sales force research have
not yet been examined in a meta-analysis. One stream
examines adaptive selling, that is, “the altering of sales
behaviors during a customer interaction or across customer
interactions based on perceived information about the
nature of the selling situation” (Weitz, Sujan, and Sujan
1986, p. 175). This approach enables salespeople to tailor
messages to fit individual customers’ needs and prefer-
ences. The other stream involves customer-oriented selling,
which “can be viewed as the practice of the marketing con-
*George R. Franke is Professor and Reese Phifer Fellow of Marketing,
Department of Management and Marketing, University of Alabama
(e-mail: gfranke@cba.ua.edu). Jeong-Eun Park is an assistant professor,
Marketing Department, University of New Hampshire (e-mail: John.
Park@unh.edu). Tom Brown, Eric Harris, and Charles Pettijohn provided
valuable unpublished correlations for use in the meta-analysis. The authors
also gratefully acknowledge the helpful comments of David Ford, the
anonymous JMR reviewers, and the late Dick Wittink.
cept at the level of the individual salesperson and customer”
(Saxe and Weitz 1982, p. 343). This approach emphasizes
long-term customer satisfaction rather than short-term sales
objectives.
The purpose of this study is to provide an integrative
meta-analysis of research on both adaptive selling and
customer-oriented selling. Using correlations obtained from
journal articles, dissertations, and other sources, we test a
structural model in which performance and satisfaction are
outcomes of adaptive selling behaviors (ASB) and customer
orientation (CO). We extend previous research on the rela-
tionships between objective and subjective performance
measures (Rich et al. 1999) by examining objective out-
comes, self-ratings, and managerial ratings of performance.
We also assess the effects of salesperson gender and selling
experience. Finally, we compare alternative model specifi-
cations and test various potential moderators of the relation-
ships in the model. The results should suggest useful
avenues for sales force management and research.
BACKGROUND AND HYPOTHESES
Figure 1 presents an overview of the model. Hypothesis
numbers indicate proposed antecedents and consequences
of ASB and CO. Because the model allows these relation-
ships to be completely or partially mediated by intervening
variables, we consider total effects rather than direct effects
in testing the hypotheses. We include additional paths in the
model because of their substantive relevance and to avoid
confounding of other variables’ effects with those of ASB

© 2006, American Marketing Association Journal of Marketing Research

ISSN: 0022-2437 (print), 1547-7193 (electronic) 693 Vol. XLIII (November 2006), 693–702
694
Figure 1
INITIAL AND ALTERNATIVE MODELS OF ANTECEDENTS AND CONSEQUENCES OF ASB AND CO
and CO. The dashed arrows indicate alternative structural
models, though we also test the implied reciprocal relation-
ships with further model modifications.
ASB
Salespeople can use information they gather about
prospects before or during an interaction to customize the
content and format of their messages for more effective
communications. Weitz (1981) discusses ASB in his contin-
gency model of salesperson effectiveness, and Spiro and
Weitz (1990) develop the concept further in their presenta-
tion of the adaptive selling (ADAPTS) scale. They argue
that the benefits of ASB are likely to outweigh the costs of
gathering and responding to information under particular
circumstances, such as when the salesperson has the neces-
sary resources, when buying tasks are complex and may
result in large orders, and when customer relationships have
little conflict and are expected to continue in the future.
However, even simple adaptations in sales interactions,
such as reactions to questions, comments, body language,
and so on, may improve rapport and reduce objections.
Accordingly, research largely concludes that ASB improves
salesperson performance regardless of the circumstances
(e.g., Boorom, Goolsby, and Ramsey 1998; Spiro and Weitz
1990; Weitz, Sujan, and Sujan 1986). Therefore, we pro-
pose the following:
H1: Adaptive selling increases salesperson performance.
The practice of adaptive selling may be intellectually
stimulating, increasing the salesperson’s enjoyment of the
selling experience (Sujan, Weitz, and Sujan 1988). In addi-
tion, the conscious effort to notice and respond to differ-
ences between prospects may give salespeople a feeling of
satisfaction with their performance (Keillor, Parker, and
Pettijohn 1999). Work effort tends to increase job satisfac-
tion in both sales and nonsales contexts (Brown and Peter-
son 1994). Therefore, we propose the following:
H2: Adaptive selling increases salesperson job satisfaction.
CO
With its emphasis on long-term customer satisfaction, a
high CO reflects a high concern for others, low-pressure
selling, and problem–solution selling approaches (Saxe and
JOURNAL OF MARKETING RESEARCH, NOVEMBER 2006
Weitz 1982). In general, empirical findings indicate a strong
correspondence between customer perceptions of sales-
person CO and satisfaction, trust, and salesperson relation-
ship development (e.g., Stock and Hoyer 2005; Williams
and Attaway 1996). However, as with ASB, the benefits of
customer-oriented selling must be weighed against its costs,
such as the salesperson’s time spent identifying customer
problems and solutions, possibly reduced margins or
increased service costs entailed in satisfying customers, and
lost sales that might have been made with more aggressive
sales approaches. In their presentation of the selling
orientation–customer orientation (SOCO) scale, Saxe and
Weitz (1982) report a significant, positive correlation
between CO scores and performance for the salespeople
who are above average on their ability to help the customer
and the strength of their customer relations. Otherwise, the
correlation is negative or nonsignificant, and it is nonsignif-
icant within each of four sales forces and for the sample as
a whole (Saxe 1979).
Despite the limited initial results, subsequent authors
have argued for a positive effect of CO on salesperson per-
formance (e.g., Keillor, Parker, and Pettijohn 2000).
Schwepker (2003) concludes from his narrative review of
the literature on CO that this relationship holds for both
business-to-business and retail sales. Therefore, we propose
the following:
H3: Customer-oriented selling increases salesperson
performance.
The literature suggests that there is a positive relationship
between salesperson CO and job satisfaction. Customer-
oriented salespeople should find the process of satisfying
customers’ needs intrinsically pleasing (e.g., Brown et al.
2002). In addition, salespeople who are high in CO may
better understand their service role and experience less role
stress than salespeople who are low in CO, so that high CO
should reduce role conflict and ambiguity (e.g., Saxe and
Weitz 1982). In turn, this would increase job satisfaction.
Therefore, we propose the following:
H4: Customer-oriented selling increases salesperson job
satisfaction.
Adapting to customers entails focusing on their individ-
ual needs and preferences, which may lead to a customer-
Salesperson Adaptive Selling Behavior and Customer Orientation
centered, problem-solving orientation. An empathetic rela-
tionship between the salesperson and the customer may also
result from the process of adaptive selling (Siguaw 1991).
Therefore, a possible antecedent of customer-oriented sell-
ing is ASB:
H5: Adaptive selling behavior increases customer-oriented
selling.
A different perspective is that CO entails concern for
solving customers’ specific problems and therefore entails
gathering information and adapting to individual customers’
needs (Bodkin 1989). The effect of CO could be similar to
that of empathy, which Spiro and Weitz (1990) hypothesize
and find to be positively correlated with ASB. Accordingly,
H5 competes with an alternative hypothesis that reverses the
direction of influence:
H5r: Customer-oriented selling increases ASB.
Gender and Experience
Women tend to be less aggressive and more agreeable,
empathetic, and relationship-oriented than men (e.g., Fein-
gold 1994). Empathy and other characteristics that increase
awareness of situational factors and the willingness to adapt
behaviors accordingly may increase ASB (Spiro and Weitz
1990). Female salespeople may also tend to be more con-
cerned with customer satisfaction and opposed to “hard-
sell” techniques, making them higher in CO (e.g., O’Hara,
Boles, and Johnston 1991). Therefore, gender may influ-
ence both ASB and CO:
H6: Female salespeople practice adaptive selling more than
male salespeople.
H7: Female salespeople are more customer oriented than male
salespeople.
Adaptive selling draws on skills and knowledge of selling
situations, customer types, and selling strategies that
increase with experience (e.g., Spiro and Weitz 1990; Weitz,
Sujan, and Sujan 1986). Experienced salespeople may also
have a greater ability to identify ways to help satisfy cus-
tomer needs, a longer-term orientation, and more repeat
customers. These factors should increase CO (O’Hara,
Boles, and Johnston 1991; Pettijohn, Pettijohn, and Taylor
2002; Saxe and Weitz 1982). Therefore, we propose the
following:
H8: Increased levels of sales experience increase adaptive
selling.
H9: Increased levels of sales experience increase customer-
oriented selling.
There is no obvious reason to predict an effect of gender
on salesperson performance or job satisfaction except
through the mediating effects of ASB and CO (cf. Siguaw
and Honeycutt 1995). However, sales experience would
logically appear to be an antecedent of both job perform-
ance and satisfaction. Ineffective or unhappy salespeople
may eventually look for a more suitable occupation and
consequently not be represented among the more experi-
enced respondents to a sales force survey (cf. Churchill,
Ford, and Walker 1976). Therefore, the model allows for
direct effects of experience on performance and satisfaction
in testing the antecedents and consequences of ASB and
CO.
695
Job Performance and Satisfaction
An enduring question in sales force research is the rela-
tionship between job performance and satisfaction. The
majority view is that performance leads to desired out-
comes, such as pay, recognition, and promotion, which in
turn lead to satisfaction (Yilmaz 2002). Effective sales-
person job performance should be intrinsically satisfying,
and if the organization’s systems for performance measure-
ment and compensation are not defective, it should also be
extrinsically rewarding. The opposing view is that satisfac-
tion affects job-related attitudes and emotions, which in
turn influence performance. To test both perspectives, we
examine alternative models in which performance affects
satisfaction, and vice versa.
Potential Moderator Variables
In general, prior research has shown little evidence of
moderators of ASB or CO. One exception is the work of
Stock and Hoyer (2005), who show that empathy, expertise,
and other factors moderate the effect of the salesperson’s
customer-oriented attitudes on customer perceptions of the
salesperson’s customer-oriented behavior. The current
meta-analysis covers a wide range of sales force samples
and research procedures, which may facilitate the detection
of moderator effects on antecedents or consequences of CO.
We examine the following potential moderators: customer
type (organizational or consumer), product type (goods or
services), salesperson gender and selling experience, meas-
ures used (ADAPTS or others, and SOCO or others), publi-
cation year, and publication outlet (journal or others). As is
the case with the other moderators, we assess gender and
experience relative to studies rather than to individuals. We
examine gender as the proportion of women in the sample,
and experience is the average years of total selling experi-
ence in the sample. Unfortunately, we cannot test
individual-level interactions, such as ASB × CO, with the
data available for the meta-analysis.
METHOD
Collection and Coding of Studies
To limit potential cultural influences on model relation-
ships, the meta-analysis uses only studies of native English-
speaking sales forces. In addition, to reduce redundancy
with Churchill and colleagues’ (1985) meta-analysis and to
correspond with the date of Saxe’s (1979) development of
the SOCO scale, we use only studies dated 1979 or later.
Subject to these constraints, we identified potentially rele-
vant studies in multiple ways. We posted a request for arti-
cles and working papers on the electronic marketing list-
server ELMAR. We searched ABI/Inform, ProQuest Digital
Dissertations, and other computerized databases using key-
words related to adaptive selling, CO, and personal selling
in general. We examined the Social Sciences Citation Index
for articles that cited Saxe and Weitz’s (1982) and Spiro and
Weitz’s (1990) seminal studies. We manually searched jour-
nals, such as Industrial Marketing Management and Jour-
nal of Personal Selling & Sales Management. When we
identified relevant studies, we obtained them when possible
and examined their references for additional sources.
We treated multiple publications based on the same sam-
ple as a single study, and we treated results from distinct
samples within a single study as independent. In all, articles
in 33 different journals, 48 dissertations, and six conference
696 JOURNAL OF MARKETING RESEARCH, NOVEMBER 2006
proceedings and working papers yielded correlations from
155 different samples that included 31,428 salespeople. The
samples represent multiple sales approaches, such as tele-
marketing, door-to-door, organizational, and retail sales,
and products, such as pharmaceuticals, real estate, cars,
computers and computer services, insurance and financial
services, and advertising products and services. We
obtained ASB correlations from 42 studies and CO correla-
tions from 40 studies, and 10 studies reported both ASB and
CO. The remaining studies provided one or more correla-
tions between gender, experience, performance, and/or
satisfaction.
The key data taken from each source were the pairwise
correlations between variables in the model. We recorded
measure reliability when it was reported or when it was
based on the Spearman–Brown prophecy formula for corre-
lations determined as the composite of related individual
measures (e.g., Nunnally 1978, Ch. 5). We also recorded the
values of the aforementioned moderator variables when
possible. We discussed the codings, and analyses are based
on consensus decisions.
Meta-Analytic Analysis Procedures
We took a random-effects perspective, which Hunter and
Schmidt (2000) recommend for routine use in meta-
analysis, when calculating mean correlations and perform-
ing tests of moderators. In this approach, the variability in
findings across studies is treated as arising from both the
sampling of studies, reflecting between-studies variance,
and the sampling of individuals within studies, reflecting
sampling error. In comparison, the fixed-effects perspective
treats studies as differing only in terms of the participants
examined, plus perhaps a few identifiable study characteris-
tics that are taken into account as moderators. The random-
effects approach simplifies to the fixed-effects approach
when the studies’ outcomes are homogeneous (i.e., when all
the between-studies variance is attributable to sampling
error from the selection of study participants). When the
assumptions of the fixed-effects approach are not met, the
random-effects approach produces more generalizable
results and is less subject to Type I errors in testing mean
effect sizes (e.g., Hunter and Schmidt 2000).
As Hunter and Schmidt (1990) and Schulze (2004) rec-
ommend, we analyzed the correlations in their raw form
rather than after applying Fisher’s r-to-z transformation.
Schulze’s extensive simulations that compare alternative
meta-analytic approaches indicate that Hunter and
Schmidt’s (1990) procedures are effective for estimating
average effect sizes, testing the significance of average
effects, and estimating the variance of the effect size in the
overall population. However, their homogeneity test
(Hunter and Schmidt 1990, p. 111) has relatively low power
to distinguish between fixed- and random-effects distribu-
tions. Therefore, we used (though we do not report it in
detail) a potentially more powerful approach that Shadish
and Haddock (1994, Equation 18-23, p. 275) discuss as an
additional test of homogeneity.
We used weighted regression procedures to test the mod-
erator variables. The weights are the inverse of the sum of
the standard error of the study effect and the variance of the
population effect size (if greater than zero). This approach
produces correct coefficient estimates, but the resulting
standard errors must be divided by the square root of the
mean square error to yield standard normal variates (z sta-
tistics) for accurate tests of significance (Hedges and Olkin
1985).
Structural Model Analysis
We analyzed the mean correlations obtained in the meta-
analysis with LISREL 8.54 to test the models that appear in
Figure 1. The harmonic mean of the correlations’ total sam-
ple sizes is specified as the sample size for the analyses
(Viswesvaran and Ones 1995). Because the different per-
formance indicators are likely to be related in ways that are
not fully accounted for by their common antecedents, we
allowed their structural errors in the LISREL psi matrix to
correlate. We assumed that gender and sales experience
were measured without error. With one exception, we set
the error terms for the other constructs equal to 1 less the
mean reliabilities obtained in the meta-analysis (ASB, .82;
CO, .84; self-rated performance, .84; manager-rated per-
formance, .86; objective performance, .69; job satisfaction,
.85). The exception is the objective-performance variable.
Objective performance is often measured with a single indi-
cator, for which no reliability estimate is available. When it
is a composite of just two or three indicators, such as sales
and quota achievement, reliability estimates will be low
unless the indicators are highly correlated. Therefore, we
used a higher value of .85 as the reliability of objective per-
formance. This value is consistent with Rich and col-
leagues’ (1999) meta-analysis, it is similar to the reliabili-
ties of the other performance measures, and it may avoid
overcorrecting for measure unreliability in estimating the
structural model relationships.
RESULTS
Mean Correlations
The 28 correlations that appear in Table 1 are based on
11.6 samples and 2338 salespeople on average. They range
from .01 to .35 in absolute value, and 19 are significant. The
heterogeneity variance, which shows the variance due to
sampling of studies (as opposed to sampling of research
participants), equals .000 for only three of the distributions.
This variance equals zero in the homogeneous case that is
assumed in fixed-effects models. Both Hunter and
Schmidt’s (1990) and Shadish and Haddock’s (1994; not
shown to conserve space) tests of homogeneity show that
the variance is significant for 20 of the relationships. Con-
sequently, a fixed-effects analysis, which assumes homoge-
neous effects, would inappropriately weight the study
results and exaggerate the statistical significance of the
mean correlations for a majority of the relationships in the
model.
The only significant, negative correlation is between gen-
der and experience (r = –.16; saleswomen have less selling
experience on average than salesmen). The largest positive
correlation is between manager-rated and objective per-
formance (r = .35); ten other correlations range from .20 to
.26. The correlational results are largely consistent with the
hypotheses and other effects that appear in Figure 1. The
exceptions to the hypotheses are H3 (CO is positively
related to self-rated performance but not to manager-rated
or objective performance), H6 (gender is uncorrelated with
Salesperson Adaptive Selling Behavior and Customer Orientation 697

Table 1
META-ANALYTIC CORRELATIONS

Relationship
Number of Heterogeneity
Hypothesis Variable 1 Variable 2 Estimates Total n Mean r z Qa Variance
H1 ASB Self-rated performance 26 5822 .261 11.32** 92.16** .010
H1 ASB Manager-rated performance 6 970 .089 2.46* 7.75 .002
H1 ASB Objective performance 14 2453 .149 3.99** 49.48** .014
H2 ASB Job satisfaction 3 652 .254 5.13** 5.47 .003
H3 CO Self-rated performance 15 3254 .194 4.62** 92.24** .022
H3 CO Manager-rated performance 3 542 .013 .39 1.86 .000
H3 CO Objective performance 15 3311 .021 .49 92.20** .023
H4 CO Job satisfaction 11 2452 .221 6.52** 33.92** .009
H5/H5r ASB CO 10 2155 .259 5.81** 48.84** .016
H6 Genderb ASB 5 1266 –.030 –.92 6.68 .001
H7 Genderb CO 7 1119 .165 2.52* 35.23** .024
H8 Experience ASB 10 2635 .137 4.90** 21.16* .004
H9 Experience CO 9 981 .040 1.58 5.61 .000
— Experience Self-rated performance 17 2978 .220 9.32** 30.95* .004
— Experience Manager-rated performance 2 426 .201 2.71** 5.06* .007
— Experience Objective performance 8 1597 .262 3.71** 73.35** .036
— Experience Job satisfaction 9 1745 .033 .76 28.93** .011
— Self-rated performance Job satisfaction 24 5584 .212 9.92** 66.85** .007
— Manager-rated performance Job satisfaction 13 2856 .139 4.87** 31.29** .006
— Objective performance Job satisfaction 19 4769 .126 4.11** 86.75** .014
— Gender Experience 14 2712 –.161 –4.69** 47.02** .012
— Gender Self-rated performance 11 2138 –.010 –.54 8.03 .000
— Gender Manager-rated performance 8 1475 –.034 –1.21 9.40 .001
— Gender Objective performance 9 1482 –.042 –.95 25.60** .011
— Gender Job satisfaction 12 2529 .041 1.37 27.50** .006
— Self-rated performance Manager-rated performance 9 1423 .246 8.07** 13.40 .003
— Self-rated performance Objective performance 13 2470 .263 6.29** 64.66** .018
— Manager-rated performance Objective performance 22 3674 .352 10.65** 113.68** .019
*p < .05.
**p < .01.
aQ is a chi-square statistic that indicates whether the heterogeneity variance is significantly greater than zero.
bGender is coded as male = 0, and female = 1.

ASB), and H9 (experience is uncorrelated with CO).
Experience is also unrelated to job satisfaction, contrary to
expectations.
Structural Model Analyses
The harmonic mean of the sample sizes for the estimates
in Table 1 is 1544, giving substantial power to detect poor
model fit and obtain significant parameter estimates in the
LISREL analysis. We report parameter estimates for three
alternative models in Table 2. The initial model is as it
appears in Figure 1, with ASB causing CO and performance
causing job satisfaction. The fit of this model is excellent:
χ2 = .576, d.f. = 6; root mean square error of approxima-
tion = 0; nonnormed fit index = 1.02; and standardized root
mean square residual = .003. The variance explained is
2.3% for ASB, 13.5% for CO, 15.7% for self-rated perform-
ance, 5.3% for manager-rated performance, 10.1% for
objective performance, and 15.9% for job satisfaction.
As with the correlational results, the total effects support
many of the hypothesized relationships shown in Figure 1.
Adaptive selling behavior increases all three performance
measures (H1), job satisfaction (H2), and CO (H5). Cus-
tomer orientation increases self-rated performance (H3) and
job satisfaction (H4), though contrary to H3, it has no effect
on manager-rated or objective performance. Gender has no
effect on ASB, contrary to H6, but it increases CO (H7).
Experience has a positive effect on ASB (H8) and CO (H9).
The influence on CO combines a nonsignificant direct
effect plus an indirect effect mediated by ASB to yield a
significant total effect, contrary to the correlational results.
Experience also increases performance, though it does not
increase job satisfaction. Both self-rated and manager-rated
performance increase job satisfaction. Finally, the mediat-
ing effect of CO leads to small but significant total effects
of gender on self-rated performance and job satisfaction.
The second model in Table 2 allows for reciprocal paths
between ASB and CO and between performance and satis-
faction to help clarify the direction of the relationships
between variables (cf. Bagozzi 1980; Brown and Peterson
1993). Further modifications make the model identified and
more parsimonious. The paths from gender to ASB, experi-
ence to CO, and experience to satisfaction are omitted, and
the effects of self-rated, manager-rated, and objective per-
formance on satisfaction are constrained to be equal, as are
the effects of satisfaction on performance. The resulting
model fits the data well (χ2 = 2.37, d.f. = 9; root mean
square error of approximation = 0; nonnormed fit index =
1.02; and standardized root mean square residual = .006).
The direct and total effects in this model are mostly similar
in significance and magnitude to the initial results. The total
effect of ASB on CO is significant and positive (effect =
.39, t = 4.77), whereas the influence of CO on ASB is non-
significant (effect = –.10, t = –.82). Therefore, H5 is sup-
ported and H5r is not. However, the results show that satis-
faction leads to performance (effect = .16, t = 2.79) rather
than performance to satisfaction (effect = –.02, t = –.43).
 698

Table 2
DIRECT AND TOTAL EFFECTS FOR STRUCTURAL MODELS

Initial Model Reciprocal Model Final Model

Direct Total Direct Total Direct Total
Hypothesis Relationship Effect t Effect t Effect t Effect t Effect t Effect t
H1 ASB → self-rated performance 0.24 07.65**0 0.29 09.66** 0.20 05.66** 0.29 10.12** 0.20 06.26** 0.29 09.66**
H1 ASB → manager-rated performance 0.08 02.52**0 0.08 02.50** 0.04 01.14** 0.07 02.43** 0.04 01.28** 0.08 02.50**
H1 ASB → objective performance 0.15 04.67**0 0.14 04.65** 0.11 02.98** 0.13 04.21** 0.12 03.63** 0.14 04.65**
H2 ASB → job satisfaction 0.21 06.30**0 0.31 10.20** 0.26 06.54** 0.31 10.82** 0.25 07.85** 0.31 10.20**
H3 CO → self-rated performance 0.15 04.83**0 0.15 04.83** 0.12 03.58** 0.12 02.51** 0.12 03.87** 0.15 04.83**
H3 CO → manager-rated performance –.02 0–.60**0 –.02 0–.60** –.05 –1.48** –.03 0–.79** –.05 –1.50** –.02 0–.60**
H3 CO → objective performance –.03 –1.06**0 –.03 –1.06** –.06 –1.94** –.05 –1.33** –.05 –1.73** –.03 –1.06**
H4 CO → job satisfaction 0.17 05.52**0 0.18 05.99** 0.19 05.99** 0.15 03.08** 0.18 05.99** 0.18 05.99**
H5 ASB → CO 0.31 10.57**0 0.31 10.57** 0.40 03.74** 0.39 04.77** 0.31 10.57** 0.31 10.57**
H5r CO → ASB — — — — –.10 0–.78** –.10 0–.82** — — — —
H6 Gender → ASB –.01 0–.27**0 –.01 0–.27** — — –.02 0–.79** –.01 0–.27** –.01 0–.27**
H7 Gender → CO 0.19 07.28**0 0.19 06.96** 0.19 07.11** 0.19 07.12** 0.19 07.28** 0.19 06.96**
H8 Experience → ASB 0.15 05.32**0 0.15 05.32** 0.16 05.45** 0.15 05.34** 0.15 05.32** 0.15 05.32**
H9 Experience → CO 0.03 01.02**0 0.08 02.72** — — 0.06 03.43** 0.03 01.02** 0.08 02.72**
— Experience → self-rated performance 0.20 07.46**0 0.24 09.01** 0.20 07.59** 0.24 09.00** 0.20 07.59** 0.24 09.01**
— Experience → manager-rated performance 0.21 07.503** 0.22 08.02** 0.21 07.63** 0.22 08.05** 0.21 07.65** 0.22 08.02**
— Experience → objective performance 0.26 09.81**0 0.28 10.62** 0.26 09.89** 0.28 10.61** 0.26 09.90** 0.28 10.62**
— Experience → job satisfaction –.07 –2.41**0 0.04 01.48** — — 0.04 01.43** –.01 0–.39** 0.04 01.48**
— Self-rated performance → job satisfaction 0.12 03.57**0 0.12 03.57** –.02a 0–.42** –.02 0–.43** — — — —
— Manager-rated performance → job satisfaction 0.10 03.03**0 0.10 03.03** –.02a 0–.42** –.02 0–.43** — — — —
— Objective performance → job satisfaction 0.05 01.43**0 0.05 01.43** –.02a 0–.42** –.02 0–.43** — — — —
— Job satisfaction → self-rated performance — — — — 0.17b 02.64** 0.16 02.79** .15 04.94** 0.15 04.94**
— Job satisfaction → manager-rated performance — — — — 0.17b 02.64** 0.16 02.79** .16 04.91** 0.16 04.91**
— Job satisfaction → objective performance — — — — 0.17b 02.64** 0.16 02.79** .12 03.72** 0.12 03.72**
— Gender → self-rated performance — — 0.03 02.50** — — 0.02 02.50** — — 0.03 02.50**
— Gender →manager-rated performance — — –.00 0–.65** — — –.01 0–.78** — — –.00 0–.65**
— Gender → objective performance — — –.01 –1.03** — — –.01 –1.28** — — –.01 –1.03**
— Gender → job satisfaction — — 0.03 02.95** — — 0.03 03.02** — — 0.03 02.95**
*p < .05.
**p < .01.
Notes: Parameter estimates with the same superscript are constrained to be equal. Gender is coded as male = 0, and female = 1.
JOURNAL OF MARKETING RESEARCH, NOVEMBER 2006
Salesperson Adaptive Selling Behavior and Customer Orientation
Therefore, reversing the performance–satisfaction relation-
ship in the initial model leads to a preferable final model.
The initial and final models are equivalent in the sense
that they would fit any given data exactly the same (Mac-
Callum et al. 1993). The models also yield identical total
effects for all relationships, except between performance
and satisfaction, and therefore are interchangeable with
respect to the tests of hypotheses and effects of gender and
experience. The important difference is that in the final
model, job satisfaction has a significant, positive effect on
all three measures of performance, and performance has no
influence on job satisfaction. As with the initial model, the
final model accounts for 2.3% of the variance in ASB and
13.5% of the variance in CO. It accounts for 17.7%, 7.4%,
and 11.2% of the variance in self-rated, manager-rated, and
objective performance, respectively, and 12.5% of the vari-
ance in job satisfaction.
Moderator Analyses
We examined potential moderators for 13 correlations:
ASB and CO with each other, with their two antecedents
(gender and experience), and with four consequences (job
satisfaction and three measures of performance). The eight
moderators we considered were type of customer, type of
product sold, average selling experience, proportion of
males in the sample, publication year and source, and which
measures of ASB and CO were used. We tested each mod-
erator separately to maximize the number of usable obser-
vations. Of the 104 weighted regressions run, one is signifi-
cant at p < .01, and six others are significant at p < .05.
Three indicate that the relationship between ASB and self-
rated performance has declined over time, is lower in jour-
nal articles than other sources, and is lower for more experi-
enced salespeople. Another indicates that the relationship
between ASB and sales experience is lower for the original
ADAPTS scale than for alternative measures of ASB. How-
ever, when we eliminate one dissertation that reports unusu-
ally high correlations for ASB, all four moderators become
nonsignificant. The other significant moderators suggest
that the correlation between CO and objective performance
is higher in samples with greater proportions of women and
that over time, the relationship between CO and experience
has decreased, whereas the relationship between CO and
self-rated performance has increased. As with ASB, remov-
ing one study from each regression makes these CO moder-
ators become nonsignificant. Given that we found only a
few more significant moderators than would be expected
due to chance and that none of the results survived removal
of even a single study, the findings give little evidence of
moderating effects on relationships between ASB, CO, and
other variables.
DISCUSSION
The fixed-effects perspective on meta-analysis assumes
that differences across homogeneous studies are due solely
to random sampling of subjects. The more realistic random-
effects perspective acknowledges that observed effects will
vary as a result of the sampling of studies and the sampling
of subjects. Therefore, conclusions from a meta-analysis are
strengthened when the results are based on large samples of
studies and large numbers of subjects. Some of the current
findings should be robust; for example, the estimated rela-
699
tionship between ASB and self-rated performance is based
on 26 samples that include more than 5800 salespeople.
Others are more susceptible to change as new evidence
accumulates, such as the relationship between CO and
manager-rated performance based on three samples of 542
salespeople.
Tests of moderators, which compare effects across levels
of various predictor variables, are also heavily influenced
by the number of studies available for analysis. The results
do not show any strong evidence of effects varying over
time, product or customer type, or other study characteris-
tics. These negative findings could be due to limited statisti-
cal power, considering that only three tests had as many as
24 degrees of freedom for error, and most had fewer than
10. As further studies become available, the ability to detect
moderators will increase. In addition, further research may
perform insightful tests by examining within-study modera-
tors. For example, primary research could test for interac-
tive effects of ASB and CO, which was not possible in the
meta-analysis, or examine ASB × experience or CO × gen-
der with possibly hundreds of degrees of freedom.
Subject to these important caveats, this study updates and
extends several previous meta-analyses of sales force
research and provides new evidence on the antecedents and
consequences of adaptive selling and CO. As in prior inves-
tigations, the predictors do not account for large amounts of
variance in salesperson performance: less than 7% for any
single predictor and less than 18% for any structural model.
However, Dunlap’s (1994) common language effect size
indicator for bivariate normal correlations, or CLR, shows
that even small effects may be managerially relevant. For
two variables correlated r > 0, CLR = sin–1(r)/π + .5 (with
the sine in radians) can be interpreted as the probability that
between two randomly selected people, the one who is
higher on one of the variables will also be higher on the sec-
ond variable. When variables are uncorrelated, the probabil-
ity is simply 50%. The .261 correlation between ASB and
self-rated performance has an associated CLR of .584, sug-
gesting that a salesperson who is higher in ASB than
another salesperson has almost a 60% probability of also
being higher in self-rated performance. It is also possible to
interpret CLR for a single person relative to the average lev-
els of two measures, such that a salesperson who is above
average in ASB has a 58% probability (r = .254) of being
above average in job satisfaction. These are not large differ-
ences, but if the costs of selecting, training, and motivating
salespeople to practice more adaptive selling are relatively
small, they may be more than justified by the expected
benefits in terms of increased performance and job
satisfaction.
Both the correlations and the structural coefficients show
that ASB is related more to self-rated performance than to
manager-rated or objective performance. The higher corre-
lation between two self-rated attributes could result from
response tendencies and common method biases (e.g., Pod-
sakoff et al. 2003). More substantively, salespeople may
consider adaptiveness an aspect or cause of performance,
which would lead to positive correlations with self-ratings
of performance. The smaller correlations found with
manager- and self-rated performance could result if both
managers and salespeople believe that ASB contributes to
performance, but manager ratings differ from self-ratings of
ASB. If so, better communication between salespeople and
700
managers about these (and other performance factors)
might lead to more consistent performance ratings (cf. Rich
et al. 1999).
Adaptive selling behavior has a positive direct effect on
satisfaction and an indirect effect mediated by CO. Adap-
tive selling is less monotonous than repeating the same
message to all prospects, it takes more concentration while
interacting with prospects, and it may also require more
precall planning and information gathering. Therefore, part
of the effect of ASB on job satisfaction may be mediated by
effort, which tends to increase both performance and satis-
faction (Brown and Peterson 1994).
Customer orientation increases self-rated performance
and job satisfaction. The effects are not large, but as with
ASB, salespeople who are high in CO believe that they are
doing a better job and are more satisfied with their job than
salespeople who are low in CO. Apparently, though,
customer-oriented selling does not consistently lead to sales
or other results that managers value, because its effects on
manager-rated and objective performance are nonsignifi-
cant. Salespeople may believe that customer-oriented sell-
ing has long-term benefits, even if their short-term results
are not recognized by managers or reflected in objective
sales measures. In addition, the costs of implementing
customer-oriented selling may be higher than salespeople
realize. Another possibility is that salespeople are not good
judges of their CO or that they do not communicate their
CO effectively to customers and managers. The correlation
between CO and manager-rated performance used in the
structural analysis is based on 3 studies of more than 500
salespeople, and the results for CO and objective outcomes
are based on 15 studies of more than 3300 salespeople, so
the findings may be reasonably generalizable. Therefore,
the meta-analysis raises questions about how effectively
customer-oriented selling implements the marketing con-
cept at the salesperson–customer level (Saxe and Weitz
1982).
Empathetic salespeople may better identify and respond
to customers’ concerns, and Siguaw (1991) suggests that
adapting to different customers can lead salespeople to
empathize with them. In addition, women tend to be more
empathetic and relationship oriented than men (Feingold
1994). Therefore, as expected, saleswomen and salespeople
who are high in ASB tend to be higher in CO. However,
gender has no influence on ASB, possibly because women
tend to be lower in self-monitoring than men (Day et al.
2002). Self-monitoring, which reflects the degree to which
people monitor others’ responses to themselves and modify
their self-presentation to fit the social climate, is positively
linked to ASB (e.g., Spiro and Weitz 1990). Thus, lower
self-monitoring could reduce any tendency for saleswomen
to be more adaptive than salesmen.
It is conceptually plausible that CO influences ASB
rather than vice versa (Bodkin 1989), and this is empirically
supported when models equivalent to the ones in Table 2 are
analyzed. However, when reciprocal relationships are
allowed, the evidence supports ASB as a cause rather than
an effect of CO. If ASB increases effectiveness, “sales-
people with both high and low customer orientation will
adapt sales presentations to customer interests” (Saxe and
Weitz 1982, p. 344). That is, both customer and selling ori-
entations could lead to ASB but for different reasons, such
as to satisfy customer needs or to make an immediate sale
JOURNAL OF MARKETING RESEARCH, NOVEMBER 2006
regardless of the customer’s long-term satisfaction. Distin-
guishing between customer-oriented attitudes and behaviors
and examining the roles of empathy and other personal
characteristics may shed additional light on the relationship
between ASB and CO (cf. Stock and Hoyer 2005).
Weitz, Sujan, and Sujan’s (1986) framework for adaptive
selling implies that ASB increases with experience, and our
results support this. Experience gives salespeople the
chance to encounter a wider variety of selling situations,
develop a broader repertoire of selling strategies, and apply
more information-acquisition skills. They also have had
more time to observe a positive effect of ASB on perform-
ance. However, the CLR for the experience–ASB relation-
ship is small (only .544). Thus, although the more experi-
enced salesperson will be higher in ASB on average, there
will be many exceptions to this pattern in most sales forces.
The same is true with the effect of experience on perform-
ance. The correlations across the three performance types
range from .20 to .26, and associated CLR values range
from .564 to .584. These correlations are similar to the
value of .20 that Ford and colleagues (1987) find for objec-
tive data and manager or peer ratings in studies dating as far
back as 1918, indicating considerable stability across stud-
ies and periods.
Sales experience has a nonsignificant correlation with
and direct effect on CO, but it has a small positive total
influence. Therefore, the results support H9, though the
rationale given for the hypothesis does not account for the
mediating effect of ASB. Experience also has a nonsignifi-
cant effect on job satisfaction. Extremely dissatisfied sales-
people should eventually switch to more fulfilling occupa-
tions, which by itself should lead to a positive experience–
satisfaction relationship. Therefore, the nonsignificant
effect suggests that some previously satisfied salespeople
become less satisfied over time, perhaps because of disap-
pointment with promotional opportunities or financial
recognition (Churchill, Ford, and Walker 1976).
In sales force research, the relationship between perform-
ance and satisfaction is generally small and sometimes non-
significant, depending on what antecedent variables are
taken into account (e.g., Brown and Peterson 1993, 1994).
The direction of the relationship is also controversial,
though the majority view is that performance leads to satis-
faction (Yilmaz 2002). However, the final model in the cur-
rent study suggests that satisfaction influences performance
rather than the reverse. The effects are similar for self-rated,
manager-rated, and objective performance measures, which
implies that failing to distinguish between types of perform-
ance has not been a limitation of previous performance–
satisfaction investigations. However, self-rated and
manager-rated performance may reflect dimensions of
extra-role performance that objective measures do not cap-
ture. If so, despite their consistent relationship to job satis-
faction, the three performance measures may show different
relationships with other outcomes, such as organizational
commitment and turnover (cf. MacKenzie, Podsakoff, and
Ahearne 1998).
It is worth noting that the small gender–performance
relationship found in the literature through 1982 (Ford et al.
1987) has largely disappeared in studies from 1979 to the
present. The total effects in the structural model show that
saleswomen have significantly higher self-rated perform-
ance and job satisfaction than salesmen, but the coefficient
Salesperson Adaptive Selling Behavior and Customer Orientation
for each is only .03. Therefore, the findings suggest that
men and women are equally suited for sales careers.
CONCLUSION
The results provide current empirical generalizations
about several widely studied variables in sales force
research. A key finding is that ASBs have stronger effects
than customer-oriented selling on salesperson performance
and satisfaction, though the strength of the effect depends
on the performance measure used. Sales experience
increases performance but not job satisfaction, and gender
has no important influence on either. Satisfaction increases
all three measures of performance. Because some variables
show consistent relationships across performance measures
whereas others do not, the findings support Churchill and
colleagues’ (1985, p. 117) recommendation “that the use of
multiple indicators of different types to assess the perform-
ance construct is desirable.”
Further research could extend the findings in several
ways. Longitudinal and experimental designs could help
justify the causal interpretation and clarify the directions of
the relationships between ASB, CO, and other variables.
Measuring variables we did not include in the meta-
analysis, such as empathy, effort, role conflict and ambigu-
ity, and in-role versus extra-role performance, could reveal
how the relationships between certain variables are medi-
ated or moderated by other variables. Comparing the effects
of self-rated, manager-rated, and customer-rated CO on per-
formance could have valuable implications for sales force
training and management. Within-study analyses could bet-
ter measure and test potential moderators, such as the nature
of the buying task, the strength of the customer relationship,
years of selling experience, or interactive effects of ASB
and CO. Replications of prior research and reports of corre-
lations between variables in addition to any other relevant
findings would strengthen the empirical foundations for
some of the relationships we studied in this meta-analysis.
In many cases, authors of prior studies could make a useful
contribution to the literature simply by reanalyzing existing
data sets and presenting their findings in additional detail.
Further research could also examine alternatives to the
causal sequences we investigated here. For example, adap-
tive salespeople may be better able to use feedback from
their manager or customers to improve their performance. If
so, performance outcomes of selling efforts could influence
and be influenced by ASB. Another modification to the
final model in Table 2 could make job satisfaction a cause
of CO or both CO and ASB. Satisfaction logically appears
to derive directly from working to adapt to customers and
meet their needs, but Siguaw (1991) argues that job satis-
faction increases organizational commitment, effort, empa-
thy, consideration, and other traits that motivate salespeople
to be concerned with meeting the needs of their organiza-
tion and customers. Therefore, job satisfaction could be a
cause of or both a cause and effect of ASB and CO.
Meta-analysis is a way of taking stock of the available
knowledge in a research area. Although further research
may enrich or contradict the results we obtained in this
study, the extant evidence raises questions about the influ-
ence of self-perceived CO on important performance out-
comes. Instead, the literature indicates that adaptive and
experienced salespeople are likely to outperform their col-
leagues on average.
701
REFERENCES
Bagozzi, Richard P. (1980), “Performance and Satisfaction in an
Industrial Sales Force: An Examination of Their Antecedents
and Simultaneity,” Journal of Marketing, 44 (Spring), 65–77.
Bodkin, Charles D. (1989), “The Adaptive Behavior Construct and
Its Effect on Salesperson Performance,” doctoral dissertation,
Department of Marketing, Virginia Tech University.
Boorom, Michael L., Jerry R. Goolsby, and Rosemary P. Ramsey
(1998), “Relational Communication Traits and Their Effect on
Adaptiveness and Sales Performance,” Journal of the Academy
of Marketing Science, 26 (Winter), 16–30.
Brown, Steven P. and Robert A. Peterson (1993), “Antecedents and
Consequences of Salesperson Job Satisfaction: Meta-Analysis
and Assessment of Causal Effects,” Journal of Marketing
Research, 30 (February), 63–77.
——— and ——— (1994), “The Effect of Effort on Sales Perfor-
mance and Job Satisfaction,” Journal of Marketing, 58 (April),
70–80.
Brown, Tom J., John C. Mowen, D. Todd Donavan, and Jane W.
Licata (2002), “The Customer Orientation of Service Workers:
Personality Trait Effects on Self- and Supervisor Performance
Ratings,” Journal of Marketing Research, 39 (February),
110–19.
Churchill, Gilbert A., Jr., Neil M. Ford, Steven W. Hartley, and
Orville C. Walker Jr. (1985), “The Determinants of Salesperson
Performance: A Meta-Analysis,” Journal of Marketing
Research, 22 (May), 103–18.
———, ———, and Orville C. Walker Jr. (1976), “Organizational
Climate and Job Satisfaction in the Salesforce,” Journal of Mar-
keting Research, 13 (November), 323–32.
Day, David V., Deidra J. Schleicher, Amy L. Unckless, and Nathan
J. Hiller (2002), “Self-Monitoring Personality at Work: A Meta-
Analytic Investigation of Construct Validity,” Journal of Applied
Psychology, 87 (April), 390–401.
Dunlap, William P. (1994), “Generalizing the Common Language
Effect Size Indicator to Bivariate Normal Correlations,” Psycho-
logical Bulletin, 116 (November), 509–511.
Feingold, Alan (1994), “Gender Differences in Personality: A
Meta-Analysis,” Psychological Bulletin, 116 (November),
429–56.
Ford, Neil M., Orville C. Walker Jr., Gilbert A. Churchill Jr., and
Steven W. Hartley (1987), “Selecting Successful Salespeople: A
Meta-Analysis of Biographical and Psychological Selection
Criteria,” in Review of Marketing 1987, Michael J. Houston, ed.
Chicago: American Marketing Association, 90–131.
Hedges, Larry V. and Ingram Olkin (1985), Statistical Methods for
Meta-Analysis. San Diego: Academic Press.
Hunter, John E. and Frank L. Schmidt (1990), Methods of Meta-
Analysis: Correcting Error and Bias in Research Findings.
Newbury Park, CA: Sage Publications.
——— and ——— (2000), “Fixed Effects Vs. Random Effects
Meta-Analysis Models: Implications for Cumulative Research
Knowledge,” International Journal of Selection and Assess-
ment, 8 (December), 275–92.
Keillor, Bruce D., R. Stephen Parker, and Charles E. Pettijohn
(1999), “Sales Force Performance Satisfaction and Aspects of
Relational Selling: Implications for Sales Managers,” Journal of
Marketing Theory and Practice, 7 (Winter), 101–115.
———, ———, and ——— (2000), “Relationship-Oriented
Characteristics and Individual Salesperson Performance,” Jour-
nal of Business & Industrial Marketing, 15 (1), 7–22.
MacCallum, Robert C., Duane T. Wegener, Bert N. Uchino, and
Leandre R. Fabrigar (1993), “The Problem of Equivalent Mod-
els in Applications of Covariance Structure Analysis,” Psycho-
logical Bulletin, 114 (July), 185–99.
MacKenzie, Scott B., Philip M. Podsakoff, and Michael Ahearne
(1998), “Some Possible Antecedents and Consequences of In-
Role and Extra-Role Salesperson Performance,” Journal of
Marketing, 62 (July), 87–98.
702 JOURNAL OF MARKETING RESEARCH, NOVEMBER 2006
Nunnally, Jum C. (1978), Psychometric Theory, 2d ed. New York:
McGraw-Hill.
O’Hara, Bradley S., James S. Boles, and Mark W. Johnston
(1991), “The Influence of Personal Variables on Salesperson
Selling Orientation,” Journal of Personal Selling & Sales Man-
agement, 11 (Winter), 61–67.
Pettijohn, Charles E., Linda S. Pettijohn, and A.J. Taylor (2002),
“The Influence of Salesperson Skill, Motivation, and Training
on the Practice of Customer-Oriented Selling,” Psychology &
Marketing, 19 (September), 743–57.
Podsakoff, Philip M., Scott B. MacKenzie, Jeong-Yeon Lee, and
Nathan P. Podsakoff (2003), “Common Method Biases in
Behavioral Research: A Critical Review of the Literature and
Recommended Remedies,” Journal of Applied Psychology, 88
(October), 879–903.
Rich, Gregory A., William H. Bommer, Scott B. MacKenzie,
Philip M. Podsakoff, and Jonathan L. Johnson (1999), “Apples
and Apples or Apples and Oranges? A Meta-Analysis of Objec-
tive and Subjective Measures of Salesperson Performance,”
Journal of Personal Selling & Sales Management, 19 (Fall),
41–52.
Saxe, Robert (1979), “The Customer Orientation of Salespeople,”
doctoral dissertation, Anderson School of Management, Univer-
sity of California, Los Angeles.
——— and Barton A. Weitz (1982), “The SOCO Scale: A Mea-
sure of the Customer Orientation of Salespeople,” Journal of
Marketing Research, 19 (August), 343–51.
Schulze, Ralph (2004), Meta-Analysis: A Comparison of
Approaches. Cambridge, MA: Hogrefe and Huber.
Schwepker, Charles H., Jr. (2003), “Customer-Oriented Selling: A
Review, Extension, and Directions for Future Research,” Jour-
nal of Personal Selling & Sales Management, 23 (Spring),
151–71.
Shadish, William R. and C. Keith Haddock (1994), “Combining
Estimates of Effect Size,” in The Handbook of Research Synthe-
sis, Harris Cooper and Larry V. Hedges, eds. New York: Russell
Sage Foundation, 261–81.
Siguaw, Judy A.J. (1991), “The Relationships of Customer Orien-
tation, Adaptive Selling and Selected Antecedents: A Causal
Model,” doctoral dissertation, College of Administration and
Business, Louisiana Tech University.
——— and Earl D. Honeycutt Jr. (1995), “An Examination of
Gender Differences in Selling Behaviors and Job Attitudes,”
Industrial Marketing Management, 24 (January), 45–52.
Spiro, Rosann L. and Barton A. Weitz (1990), “Adaptive Selling:
Conceptualization, Measurement, and Nomological Validity,”
Journal of Marketing Research, 27 (February), 61–69.
Stock, Ruth Maria and Wayne D. Hoyer (2005), “An Attitude-
Behavior Model of Salespeople’s Customer Orientation,” Jour-
nal of the Academy of Marketing Science, 33 (Fall), 536–52.
Sujan, Harish, Barton A. Weitz, and Mita Sujan (1988), “Increas-
ing Sales Productivity by Getting Salespeople to Work
Smarter,” Journal of Personal Selling & Sales Management, 8
(August), 9–19.
Viswesvaran, Chockalingam and Deniz S. Ones (1995), “Theory
Testing: Combining Psychometric Meta-Analysis and Structural
Equations Modeling,” Personnel Psychology, 48 (Winter),
865–85.
Weitz, Barton A. (1981), “Effectiveness in Sales Interactions: A
Contingency Framework,” Journal of Marketing, 45 (Winter),
85–103.
———, Harish Sujan, and Mita Sujan (1986), “Knowledge, Moti-
vation, and Adaptive Behavior: A Framework for Improving
Selling Effectiveness,” Journal of Marketing, 50 (October),
174–91.
Williams, Michael R. and Jill S. Attaway (1996), “Exploring
Salespersons’ Customer Orientation as a Mediator of Organiza-
tional Culture’s Influence on Buyer-Seller Relationships,” Jour-
nal of Personal Selling & Sales Management, 16 (Fall), 33–52.
Yilmaz, Cengiz (2002), “Salesperson Performance and Job Atti-
tudes Revisited: An Extended Model and Effects of Potential
Moderators,” European Journal of Marketing, 36 (11–12),
1389–1414.