![Orientation construct, initially developed by Saxe and Weitz (1982).
A number of empirical studies investigated sales-oriented and
customer-oriented behaviors, but we found no research that
explicitly examine how both dimensions of the SOCO scale affect
strategic account manager's performance. In fact, the existing
literature on strategic account management pays little attention to
the impact of these two constructs on the strategic account
manager's outcomes. In addition to this, given the emphasis placed
in the literature (e.g. Wotruba & Castleberry, 1993; Sengupta et al.,
2000) on the coordination role and integration tasks performed by
the strategic account manager, it is also important to understand
how this dimension might influence his/her outcomes. Therefore,
whereas, Schultz and Evans (2002) study focused on the individual-
level determinants of strategic account manager performance, we
also took into consideration – through the inclusion of the team
selling construct – the collective or organizational dimension of his/
her work. If customer orientation and selling orientation reflect the
strategic account manager's individual behaviors towards the
customer, team selling, by contrast, expresses the collective
dimension of the strategic account manager's position as it
represents his/her efforts to marshal and coordinate resources
from different people and departments in his/her organization.
2) As statedbySchultz and Evans(2002,pp. 29–30), oneof the limitations
of their work was that their model did not test causal relationships
between outcome constructs. Therefore, in order to enrich their
framework, we posit role performance and synergistic solutions as
antecedents of strategic account manager trust. From a theoretical
standpoint, this choice is based on literature on relationship marketing
and relationship selling. In fact, trust is probably the most important
outcome construct in relational literature, especially because it is the
building block for relationship commitment and has a long term
nature (e.g. Morgan & Hunt, 1994; Doney & Cannon, 1997; Weitz &
Bradford, 1999). For example, Berry (1996, p. 42) states that “trust [is]
perhaps thesinglemost powerful relationship marketingtool available
to a company” and, as highlighted by Gundlach and Murphy (1993, p.
41), “the variable most universally accepted as a basis of any human
interactionorexchangeistrust”.Similarly,trustisakeyconstructinthe
literature investigating relationship selling (e.g. Jolson, 1997; Wilson,
2000). In our context of analysis, customer trust develops thankstothe
capacity of the strategic account manager to cooperate with the buyer
(s) to accomplish joint solutions (i.e. synergistic solutions) and to fulfill
the different facets of his/her role (i.e. role performance). Moreover,
based on relational literature, relationship selling performance should
include both present and future indicators of the customer's relation-
ship with the salesperson. In this perspective, role performance and
synergistic solutions represent measures of present results reached
either individually or jointly. In other words, they assess the current
state of the strategic account manager's performance. Conversely,
customer trust provides a crucial insight into the future stability and
potential growth of the relationship. In fact, as highlighted by Swan,
Bowers and Richardson (1999), the concept of trust includes
expectations about the future and expresses the customer's will-
ingness to accept the risk of undesirable outcomes if the salesperson
lacks the ability or motivation to keep his/her promises.
The following paragraphs examine the conceptualization of each
construct in our model and develop hypotheses.
2.1. Strategic account managers' relational behaviors
The exogenous variables in the model can be defined as selling
behaviors or, as suggested by Crosby, Evans and Cowles (1990), as
behavioral tendencies exhibited by representatives of vendors affecting
their behavioral performance. One aspect of the strategic account
manager position is to act individually to manage the relationship with
the accounts. According to the literature, in dealing with customers, he
or she may adopt two distinct sets of behaviors referred as customer
and selling orientations (Saxe & Weitz,1982). However, it has also been
noted by manyauthors (Wotruba & Castleberry,1993; Weitz & Bradford,
1999) that the strategic account manager position is characterized by a
second aspect, reflecting the role of a coordinator of a sales team. Teams
are used because in a strategic account setting, the selling process
usually goes beyond the capabilities of any one individual (Workman
et al., 2003). As a result, strategic account managers are also in charge of
managing the activities of teams rather than simply managing their
ownpersonal activities (Weitz & Bradford,1999). We therefore included
in our study a third class of behaviors called team selling.
2.1.1. Customer orientation–selling orientation
Customer oriented selling can be interpreted as the implementation of
the marketing concept at the level of the individual salesperson. As such,
this selling approach is consistent with the building of long-lasting posi-
tive relationships between the buyer and the seller. Therefore, it is widely
recognized as an important class of relational selling behaviors (Flaherty,
Dahlstrom & Skinner, 1999; Keillor, Parker & Pettijohn, 2000; Martin &
Bush, 2003;Schultz &Good,2000),butquitesurprisinglyit hasnever been
investigatedina typicalrelationalcontextlikekeyaccountmanagement.It
is noteworthy that the SOCO scale incorporates two sub-scales, i.e. selling
orientation and customer orientation: although strongly (and negatively)
interrelated, these two factors have rarely been investigated separately. In
this study, we will develop and test opposing hypotheses for the two sub-
scales.Importantly,althoughSOCO has beenconceptualized inmany ways
(e.g. as an attitude, a measure of performance, etc.), it is best viewed as a
set of behaviors (Schwepker, 2003).
2.1.2. Team selling
Teamsellingimpliesthatmultiplecontactsareestablishedbetweenthe
supplier and the buyer (Cron & DeCarlo, 2006). It is important to underline
that such contacts usuallytake place and need tobe coordinated effectively
and efficiently even when formal sales teams do not exist in the selling
firm. In this study, this construct includes the strategic account manager's
ability to identify experts within his/her organization and deploy them to
help the customer. Although similar to the idea of entrepreneurial ability
offered by Sengupta et al. (2000), our conceptualization goes further and
incorporates the efforts made by the strategic account manager to
coordinate the resources mobilized. Team selling is usually adopted in
complex buyer–seller situations, where the customer requires special
treatment and the potential sale is large, i.e. typical conditions of
relationships with strategic and keyaccounts (Jackson, Widmier, Giacobbe,
&Keith,1999). As themembers of thesales teamusuallybelongtodifferent
departments, the necessity of a strong coordination needs to be
emphasized (Honeycutt, 1996). Therefore, strategic account managers are
typically described as captains or leaders of teams who are in charge of
selecting technicians and other specialists who then meet with specialists
on the buying side (Georges & Eggert, 2003).
2.2. Strategic account managers' relational outcomes
Consistently with the goal to extend the previous work by Schultz
and Evans (2002), our study explores three relational outcomes (role
performance, synergistic solutions and trust in the strategic account
manager) of long term buyer–seller interactions. Role performance
and synergistic solutions can be classified as pertaining to the
individual's effectiveness (i.e. role performance), on the one hand,
and the combined effort of the two parties (i.e. synergistic solutions),
on the other. As pointed out beforehand, trust in the strategic account
manager (i.e. a critical aspect of relationship success) is interpreted as
the final relational outcome variable.
2.2.1. Role performance
This construct describes the strategic account manager's individual
effectiveness in developing plans and providing expertise to the customer
302 P. Guenzi et al. / Industrial Marketing Management 38 (2009) 300–311](https://image.slidesharecdn.com/45888822791892897349-230525174359-3810c9a5/85/4_5888822791892897349-pdf-3-320.jpg)
![selection problems to obtain participation from dyads of respondents, the
survey respondents were all strategic account managers. A similar
approach was also adopted in recent studies on strategic account
management (see Sengupta et al., 2000; Schultz & Evans, 2002). In fact,
strategic account managers base their decisions and actions upon their
personal evaluations but also on their perceptions of the evaluations of
third parties (sales manager or customer). In keeping with Expectancy
Theory, strategic account managers will spend more effort on those
behaviors which, in their belief, lead to greater performance which, in
turn, should drive to greater rewards. In strategic account management,
performance is usually best interpreted in terms of relationship building.
Hence, both managers and academics need to understand what are the
strategic account managers’ beliefs and perceptions concerning the
linkages between behaviors and relational outcomes. This is why using
strategic account managers as respondents should be appreciated from
both a theoretical and a managerial perspective.
Strategic account managers were asked how often they perform
customer-oriented, selling-oriented and team selling behaviors. In
addition, for the different outcomes included in our model (i.e. role
performance, synergistic solutions and trust), we asked their percep-
tions about the evaluations of the strategic account.
Using self-reporting measures has some limitations. However
Spector (2006, p.225), based on the findings of many empirical
studies, claims that “evidence fails to support social desirability as a
general source of correlation inflating CMV when self-reports are
used”. In addition to this, Schultz and Evans (2002) argued that it is
appropriate to use the key account managers' self-report performance
assessment as a “proxy” interpretation of the key accounts' view
because in such close, long-term relationships the understanding
parties have of each other transcends the typical information available
in contexts characterized by single transactions (see Ickes, 1993).
Additionally, Schultz and Evans suggested that using self-reporting
measures is appropriate especially when (like in our case) this self-
assessment does not refer to objective criteria (e.g. quota, sales
volume, etc.), but rather to qualitative dimensions (e.g. trust). Finally,
it is important to underline that Schultz and Evans (2002, p.26) also
surveyed 40 key accounts in their study and found that “the customer
survey yielded virtually identical relationship results to those of the
vendor data”. Based on these arguments, we also conclude, just like
Schultz and Evans, that the use of key account managers to measure
the quality of the customer relationship is appropriate.
In addition to these considerations, consistently with the recom-
mendations made by Podsakoff et al. (2003), we used a number of
procedural and statistical remedies against the potential problems
associated with common method bias and single respondent bias.
First, we protected respondent anonymity by assuring complete
confidentiality. Second, we reduced item ambiguity by using well
established scales and by pre-testing the questionnaire. Third, we
used different scale formats and scale anchors (“never–always”;
“strongly disagree–strongly agree”; “very little–to a great deal”; etc.).
Fourth, we used Harman's one-factor test: a principal component
factor analysis on all the variables measured using the survey
instrument revealed six factors with eigenvalues greater than 1,
which together accounted for 73.8% of the total variance; also, the
first, largest factor did not account for a majority of the variance (28%).
Therefore, it can be concluded that common method bias is not likely
to be a serious concern in our study (Podsakoff & Organ, 1986).
To test our hypotheses, a cross-sectional survey was conducted in Italy.
Thequestionnairewasoriginallydeveloped inEnglish,asalltheconstructs
used in the research were based on studies in English-speaking countries.
The English version of the questionnaire was translated into Italian by one
expert translator and then translated back into English by a second; both
translators reconciled differences. Strategic account managers were
selected as respondents. The goal of the sampling process was to include
sales organizations fromvarious selling environments. Using a database of
former participants in executive education training programs in a major
Italian management school, 220 potential respondents were selected.
Although many differences arose in their job titles, they all stated they
were in charge of managing relationships with their company's strategic
accounts. They were informed by phone that they would be receiving the
questionnaire. Participants were also assured complete confidentiality.
Respondents provided data for a specific strategic account for which they
were responsible. If they handled several accounts, they were asked to
choose the one that best represented the above-cited definition of a
strategic account. Three weeks after the initial contact, non-respondents
were followed up by telephone. A total of 127 usable questionnaires were
obtained, thus achieving a response rate of 57.7%. In line with our
objectives, many different industries are represented, including industrial
goods (34%), consumer/durable goods (28.2%), services (29.1%), pharma-
ceuticals/medical (6.8%), and others (1.9%). Equally noteworthy is the fact
that the sample includes organizations which vary in size. The average
experience of the respondents participating in the study is 4 years in their
current position, just under 7 years with their current organization, and
10.5 years in a sales position. The average number of strategic accounts of
the selling firm is 37, and the average number of accounts managed per
strategic account manager is 7.7. Remuneration is mainly based on fixed
salary (79% of total). Commissions and bonuses are 7% and 12%
respectively. As for the characteristics of the working environment,
companies in the sample operate in highly competitive markets, where
the sales force is of paramount importance (whereas e-commerce is not
relevant), there is a wide variety of customers and sales trends are mainly
stable. Early and late respondents were compared to check for a possible
response bias (Armstrong & Overton,1977). A t-test of difference in means
on all the constructs used in the research model showed no significant
difference in means between earlyand late respondents (see Appendix A).
3.2. Measures
Based on a review of the literature, we used existing scales for each
construct, with the exception of team selling. In fact, although
Workman, Homburg and Jensen (2003) have recently offered a new
three-item scale to measure “use of teams”, no generally accepted
measure of team selling exists in the sales literature. Therefore,
following our review of the literature on this topic, a new five-item
survey instrument was developed. To measure the customer orienta-
tion and selling orientation, the ten items of the SOCO scale, as
proposed by Thomas, Soutar and Ryan (2001), were used. As for role
performance and synergistic solutions, we used the two scales
developed by Schultz and Evans (2002). Finally, regarding trust we
pooled the items used by Sengupta et al. (2000), which only refer to
the customer's perceptions about the strategic account manager's
honesty/benevolence, together with those adopted by Schultz and
Evans (2002) which, in contrast, express the customer's perceptions
about the reliability and competence of the counterpart. By doing this,
our measure is consistent with the traditional definition of inter-
personal trust in buyer–seller relationship, i.e. a situation where “the
industrial buyer believes and feels that he can rely on what the
salesperson says or promises to do in a situation where the buyer is
dependent upon the salesperson's honesty and reliability (Swan &
Nolan, 1985, p.40). In fact reliability and benevolence are the two key
facets of this construct (Swan, Bowers & Richardson 1999). Our choice
to treat interpersonal trust as a unidimensional construct including
items that tap both the credibility and benevolence aspects is
consistent with the approach adopted by Doney and Cannon (1997,
p.43), who note that “Although credibility and benevolence could be
conceptually distinct […] they may be so intertwined that in practice
they are operationally inseparable”. Moreover, in keeping with the
suggestion made by Swan, Bowers and Richardson, (1999), our scale
incorporates items at different level of abstraction, i.e. items which
refer to specific behaviors of the strategic account manager (e.g. “could
be trusted to make emergency decisions, if he/she could not be
reached”), to attributes broader than a specific behavior and to general
304 P. Guenzi et al. / Industrial Marketing Management 38 (2009) 300–311](https://image.slidesharecdn.com/45888822791892897349-230525174359-3810c9a5/85/4_5888822791892897349-pdf-5-320.jpg)
![Because, one of our goals is to test a causal order between
relational outcomes in the model, we need to examine if synergistic
solutions and role performance act as mediating variables between
individual behaviors and customer trust. We address this issue by
comparing alternative model formulations to our baseline model (i.e.
Model 1). In Model 2 role performance and synergistic solutions were
excluded and the three strategic account manager's behaviors were
directly linked to customer trust. Model 3 includes both the direct and
indirect paths from behaviors to customer trust.
We will first discuss the results of Model 1 and then compare them
to those of Model 2 and 3.
4.1.2. Structural model and hypothesis testing
Table 3 reports the standardized parameters for Model 1, which are
obtained from bootstrap simulation (Chin, 1998a). T-values confirm
the significance of hypotheses H1 (B=0.30), H2 (B=0.30), H4 (B=
−0.17), H5 (B=0.18), H6 (B=0.22), H7 (B=0.26) and H8 (B=0.41). One
hypothesis (H3) is non-significant (B=−0.10).
The structural model demonstrates predictive power as the
variance explained (R2
) in key endogenous constructs was 0.20 for
role performance, 0.27 for synergistic solutions, 0.34 for customer
trust. The findings show that our model (i.e. Model 1) explains a large
part of the variance in the endogenous variables, with an average R2
of
0. 27. Communality and redundancy coefficients are also presented in
Table 4. They can be used essentially in the same way as the R2
, since
they reflect the relative amount of explained variance for latent and
manifest variables.
An important part of model evaluation is the examination of fit
indexes reflecting the predictive power of estimated inner and outer
model relationships. As pointed out by Tenenhaus et al. (2005, p.173),
“… differently from SEM-ML, PLS path modeling does not optimize
any scalar function so that it naturally lacks of an index that can
provide the user with a global validation of the model (as it is instead
with χ2
and related measures in SEM-ML). The GoF represents an
operational solution to this problem as it may be meant as an index for
validating the PLS model globally”.
A general criterion for evaluating goodness-of-fit (GoF) is to
calculate the geometric mean of the average communality and the
average R2
(Tenenhaus et al., 2005). According to the results in Table 4,
GoF=√[(0.73)×(0.27)]=0.44, which can be considered as satisfactory
(Tenenhaus et al., 2005;).
Then, the blindfolding approach proposed by Wold (1982) was
followed to calculate the cv-communality and cv-redundancy indexes
(see Table 5). The cv-communality index (H2
) measures the quality of
the measurement model, whereas the cv-redundancy index (i.e.
Stone–Geisser's Q2
, which Tenenhaus et al. (2005, p. 174) call F2
)
measures the quality of the structural model.
As shown in Table 5, the measurement model (H2
=0.54) shows a
better quality than the structural one (F2
=0.16). However, also
endogenous constructs are correctly estimated with a minimum F2
of 0.13 for role performance and an average F2
of 0.16.
4.1.3. The mediating role of synergistic solutions and role performance:
comparing alternative models
To test a causal order between relational outcomes, we compared
Model 1 to Model 2 and 3. In Model 2, as shown in Table 3, the direct
path between customer orientation and customer trust is significant
(B=0.47) as well as the direct path between selling orientation and
customer trust (B=−0.15). However, team selling has no significant
direct effect on customer trust (B=0.10). In comparison with our initial
model (i.e. Model 1), excluding role performance and synergistic
solutions results in a drop of R2
to 0.30 for customer trust (see Table 4).
The goodness of fit index (GoF) stays stable (GoF=0.45) as no major
changes can be observed in term of communality. Table 5 shows a
similar average cv-communality for both models (H2
=0.54 for Model
1 and H2
=0.54 for Model 2) but a higher average cv-redundancy for
Model 2 (F2
=0.16 for Model 1 and F2
=0.22 for Model 2). These results
indicate, on average, a better quality of the prediction of customer
trust for Model 2. However, it is important to keep in mind the lower
explained variance (R2
) in comparison with Model 1.
In Model 3 (see Table 4), just like in Model 1, hypotheses H1
(B=0.29), H2 (B=0.30), H4 (B=−0.16), H5 (B=0.18), H6 (B=0.22), H7
(B=0.17) and H8 (B=0.28) are significant, whereas hypothesis H3 is
non-significant (B=−0.10). In this model the direct path between
customer orientation and customer trust is significant (B=0.33),
whereas team selling and selling orientation have no significant direct
impact on customer trust (B=0.01 and B=−0.08, respectively).
As indicated in Table 4, this third model shows significant increases
in the explained variance for customer trust (R2
=0.44) as well as a
significant increase in average redundancy (F2
=0.47). In comparison
Table 3
Parameter estimation of the PLS Models by the Bootstrap method
Construct Model 1a
Model 2a
Model 3a
Dependent variable: Role performance
Customer orientation 0.30b
3.20c⁎ – – 0.29 3.02⁎
Selling orientation −0.10 1.07 – – −0.10 1.55
Team selling 0.18 2.03⁎ – – 0.18 2.04⁎
Dependent variable: Synergistic
solutions
Customer orientation 0.30 3.23⁎ – – 0.30 3.43⁎
Selling orientation −0.17 1.94⁎ – – −0.16 2.14⁎
Team selling 0.22 3.12⁎ – – 0.22 2.98⁎
Dependent variable: Customer trust
Role performance 0.26 2.74⁎ – – 0.17 1.89⁎
Synergistic solutions 0.41⁎ 4.69⁎ – – 0.28 3.04⁎
Customer orientation – – 0.47 4.22⁎ 0.33 3.00⁎
Selling orientation – – −0.15 2.18⁎ −0.08 1.04
Team selling – – 0.10 1.28 0.01 0.07
⁎Sig. if above 1.64 for 1-tailed test.
a
Model 1: original model; Model 2: model without synergistic solutions and role
performance; Model 3: full model including direct paths between customer orientation,
team selling and customer trust.
b
Standardized path coefficients.
c
T-values.
Table 4
Explained variance (R2
), communality, redundancy and goodness-of-fit index for
Models 1, 2, 3 and 4
Construct Model 1a
Model 2 Model 3 Model 4
Trust 0.34b
(0.83)c
0.14d
0.30 (0.83) 0.20 0.44 (0.83) 0.21 0.43 (0.83) 0.22
Role
performance
0.20 (0.71) 0.11 – (–) – 0.20 (0.71) 0.11 0.20 (0.71) 0.11
Synergistic
solutions
0.27 (0.85) 0.15 – (–) – 0.27 (0.85) 0.15 0.27 (0.85) 0.15
Customer
orientation
– (0.77) – – (0.76) – – (0.76) – – (0.76) –
Selling
orientation
– (0.56) – – (0.56) – – (0.56) – – (0.56) –
Team selling – (0.75) – – (0.75) – – (0.75) – – (0.75) –
Average 0.27 (0.73e
) 0.13 0.30 (0.68) 0.20 0.30 (0.73) 0.47 0.30 (0.73) 0.47
GoFf
0.44 0.45 0.47 0.47
NOTE:
a
Model 1: original model; Model 2: model without synergistic solutions and role
performance; Model 3: full model including direct paths between customer orientation,
team selling and customer trust; Model 4: final model.
b
Explained variance.
c
Communality: Communality coefficients are equal to the squared correlations
between manifest variables and their associated latent variables.
d
Redundancy: Redundancy coefficients reflect the joint predictive power of the
inner and outer model relationships.
e
Computed as a weighted average of the different communalities with the weights
being the number of manifest variables per each construct (see Tenenhaus et al., 2005,
p. 180).
f
GoF=-
= [(average communality) × (average R2
)]. Average communality is computed
as a weighted average of the different communalities with the weights being the
number of indicators per latent variable ( Tenenhaus et al., 2005).
306 P. Guenzi et al. / Industrial Marketing Management 38 (2009) 300–311](https://image.slidesharecdn.com/45888822791892897349-230525174359-3810c9a5/85/4_5888822791892897349-pdf-7-320.jpg)
More Related Content
![Merkl-Davies, Doris M. and Brennan, Niamh M. [2007] Discretionary Disclosure ...](https://cdn.slidesharecdn.com/ss_thumbnails/0421merkl-daviesbrennandiscretionarydisclosurestrategiesincorporatenarratives-121116103141-phpapp02-thumbnail.jpg?width=560&fit=bounds)
Similar to 4_5888822791892897349.pdf (20)
![Merkl-Davies, Doris M., Brennan, Niamh M. and McLeay, Stuart J. [2011] Impres...](https://cdn.slidesharecdn.com/ss_thumbnails/0425merkl-daviesbrennanmcleayimpressionmanagementandretrospectivesense-makingincorporatenarrativesasocialpsychologyperspective-121116103444-phpapp02-thumbnail.jpg?width=560&fit=bounds)
4_5888822791892897349.pdf
- 1. The impact of strategic account managers' behaviors on relational outcomes: An empirical study Paolo Guenzi a, ⁎, Laurent Georges b , Catherine Pardo c a Department of Management – Institute of Marketing, Bocconi University and SDA Bocconi School of Management, Via Bocconi, 8, 20136 Milano, Italy b IUT-Tarbes, University Paul Sabatier Toulouse III, LCG Research Center, EA 2043, Toulouse, France c OCE – EM-Lyon, 23, Avenue Guy de Collongue BP 174, Lyon Ecully Cedex, France a r t i c l e i n f o a b s t r a c t Article history: Received 10 August 2006 Received in revised form 17 July 2007 Accepted 17 September 2007 Available online 2 April 2008 This paper explores the impact of some behaviors of strategic account managers on the relational outcomes of the relationships they are in charge of. Based on a review of literature on personal selling and strategic account management, we contribute to a greater understanding of the role of specific strategic account managers' behaviors in fostering synergistic solutions, role performance and customer trust. Results of the empirical study clearly show that the use of customer orientation has a strong influence on customer trust by increasing strategic account managers' role performance and stimulating synergistic solutions. Conversely, a selling orientation negatively affects synergistic solutions, thus decreasing customer trust. Finally, team selling has a positive impact on the attainment of synergistic solutions, thereby fostering customer trust. Based on these findings, managerial and research implications are discussed. © 2008 Elsevier Inc. All rights reserved. Keywords: Relational behaviors Strategic account managers Customer trust Synergistic solutions Team selling PLS 1. Introduction Strategic account management has recently been highlighted as a major area of academic research (Leigh & Marshall, 2001; Workman, Homburg & Jensen, 2003). Managers and academics alike increasingly recognize that suppliers need to develop long-lasting relationships with selected, high stake customers (Beverland, 2001; Homburg, Workman & Jensen, 2002; Workman et al., 2003). Despite this widespread interest in strategic account management in general, individual-level behaviors that should be adopted by those who are in charge of managing relationships with strategic accounts remain an under-developed topic in academic research. As pointed out by Sengupta, Krapfel and Pusateri (2000), many different terms have been used to label these selected clients: large accounts (Miller, Heiman & Tuleja, 1991), major accounts (Colletti & Tubridy, 1987), national accounts (Shapiro & Moriarty, 1984), key accounts (Pardo, 1997; Homburg et al., 2002; Workman et al., 2003) and strategic accounts (Napolitano, 1997). In our study we use the terms “strategic accounts” and “key accounts” interchangeably, to encompass all terms suggested to describe those customers who, for different reasons (e.g. their current and future quantity of purchases, their lead user function, their reputation, etc.), are considered worthy of special attention and treatment on the part of the selling company (Walter, Ritter & Gemünden, 2001). Similarly, we use the term “strategic account manager” or “keyaccount manager” interchangeably, to refer to the person who is in charge of managing the relationships with these strategic accounts. In previous research, this position has been given different labels, such as key account salesperson (Sengupta et al., 2000), key account representative (Schultz & Evans 2002), or simply salesperson (Boles, Barksdale & Johnson, 1996). Whatever the label used, there is a widespread recognition of the fact that these individuals contribute greatly to the implementation of a relational approach to strategic customers (Iacobucci & Ostrom, 1996; Mavondo & Rodrigo, 2001; Weitz & Bradford, 1999). Nevertheless, the majority of contributions to date have investigated company-level relational behaviors (e.g. Leuthesser, 1997) and drivers of key account relationships (e.g. Sengupta, Krapfel & Pusateri, 1997). Only a few exceptions have made empirical analysis of the individual-level behaviors that should be adopted by those who are in charge of managing relationships with strategic accounts. Platzer (1984), and Wotruba and Castleberry (1993), identified the most widespread and relevant activities performed by strategic account managers. Boles, Barksdale and Johnson (1996) listed eight classes of behavior, requested by decision-makers in large accounts. Sengupta et al. (2000) studied the determinants of key account salesperson effectiveness. More recently, Schultz and Evans (2002) developed and tested a model of collaborative communication by key account representatives. To sum up, the existing literature provides a first set of rich and interesting contributions but it has only scratched the surface of what affects strategic account managers' outcomes. Much more needs to be investigated if we want to fully understand how to manage a strategic accounts sales force. In other words, just like a lot of attention has been Industrial Marketing Management 38 (2009) 300–311 ⁎ Corresponding author. Tel.: +39 02 58366820; fax: +39 02 58366893. E-mail addresses: [email protected] (P. Guenzi), [email protected] (L. Georges), [email protected] (C. Pardo). 0019-8501/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.indmarman.2007.09.011 Contents lists available at ScienceDirect Industrial Marketing Management
- 2. focused on the individual-level determinants of salesperson performance, we argue that a similar effort should be placed on the antecedents of strategic account managers' outcomes. In light of these considerations, the purpose of this paper is to present and test a model of strategic account managers' contribution to fostering customer trust. This model integrates previous research on strategic account management and builds on the framework developed by Schultz and Evans (2002) by incorporating individual-level behaviors and relating them to outcome constructs which lead to customer trust. We decided to start from this framework because it has a lotof merit fordifferent reasons. First, as stated beforehand, it is one of the very few articles investigating individual-level behaviors of strategic account managers. Second, it is based on a well accepted relationship between an individual's behaviors and his/her performance (Plank & Reid,1994; Biong & Selnes,1996). Third, it incorporates a broad set of relational outcomes – synergistic solutions, role performance and trust in strategic account representative – that capture different facets of buyer–seller relational exchanges and can reasonably result in sales and favor the development of long term relationships. Despite its relevant contribution, Schultz and Evans' study has several limitations. First, it only investigates a limited set of drivers of relational outcomes in key account management, i.e. communication aspects (informality, bi- directionality, frequency and strategic content). Obviously, many other potentially relevant drivers deserve investigation. Our research examines a completely different set of key account manager behaviors – customer orientation, selling orientation and team selling – thus expanding existing knowledge and integrating recent research in key account management and salesperson's performance. Importantly, as we will explain later on, these behaviors cover both the “individual” and the “collective” dimen- sions of a key account manager's job. Second, the original model assumes that the three relational outcomes are at the same level. In contrast, based on the literature on relationship marketing and relationship selling, we posit that synergistic solutions and role performance act as mediating variables between individual behaviors and customer trust, therefore testing a causal order between relational outcomes. A third limitation of Schultz and Evans' study is that a key construct (trust) is measured by two items only that do not include the buyer's evaluation of the vendor's benevolence. Hence, we used a four item measure incorporating this relevant component, too. A fourth limitation of the study is that the model was tested in one company only. Our study focuses on key account relationships in many different business-to-business industries. The goal of our study is to address these limitations to increase existing knowledge on key account management. This paper is divided into four sections. The first part presents the model, clarifies the different constructs examined, and develops a set of hypotheses linking strategic account managers' behaviors to relational outcomes. Next, we explain the method used to test the model. The third section presents the results of the analysis. Finally, we discuss theoretical and managerial implications, outline limita- tions of the study and highlight future research opportunities. 2. Conceptual model, constructs and hypotheses Fig. 1 shows the conceptual framework that is proposed and tested in the study. It represents an attempt to identify structural characteristics of customer trust towards the strategic account manager in a business to business selling context and focuses on the point of view of the vendors. Collecting data from this perspective was thought to be crucial as strategic account managers are often in charge of relationships and set directions for their own behaviors. As underlined by Schultz and Evans (2002), strategic account managers' interpretations of their actual behaviors and outcomes will determine possible changes in their sales strategies as surveys of customers are not always available. Specifically, the model posits strategic account manager behaviors (customer orientation, selling orientation and team selling) as antecedents to role performance and synergistic solutions. Trust in the strategic account manager represents the final outcome of role performance and synergistic solutions. As stated beforehand, the model builds on the framework developed by Schultz and Evans (2002). However, it incorporates 1) a different set of relational behaviors as well as 2) a sequential path in the outcome variables. 1) Whereas, Schultz and Evans (2002) focused on collaborative communication, our model offers a different set of behaviors, drawn from literature on relationship selling and key account management. We focus our attention on three behaviors: customer orientation, selling orientation and team selling. The first two represent both dimensions of the Selling Orientation-Customer Fig. 1. A conceptual framework of strategic account managers' contribution to customer trust. 301 P. Guenzi et al. / Industrial Marketing Management 38 (2009) 300–311
- 3. Orientation construct, initially developed by Saxe and Weitz (1982). A number of empirical studies investigated sales-oriented and customer-oriented behaviors, but we found no research that explicitly examine how both dimensions of the SOCO scale affect strategic account manager's performance. In fact, the existing literature on strategic account management pays little attention to the impact of these two constructs on the strategic account manager's outcomes. In addition to this, given the emphasis placed in the literature (e.g. Wotruba & Castleberry, 1993; Sengupta et al., 2000) on the coordination role and integration tasks performed by the strategic account manager, it is also important to understand how this dimension might influence his/her outcomes. Therefore, whereas, Schultz and Evans (2002) study focused on the individual- level determinants of strategic account manager performance, we also took into consideration – through the inclusion of the team selling construct – the collective or organizational dimension of his/ her work. If customer orientation and selling orientation reflect the strategic account manager's individual behaviors towards the customer, team selling, by contrast, expresses the collective dimension of the strategic account manager's position as it represents his/her efforts to marshal and coordinate resources from different people and departments in his/her organization. 2) As statedbySchultz and Evans(2002,pp. 29–30), oneof the limitations of their work was that their model did not test causal relationships between outcome constructs. Therefore, in order to enrich their framework, we posit role performance and synergistic solutions as antecedents of strategic account manager trust. From a theoretical standpoint, this choice is based on literature on relationship marketing and relationship selling. In fact, trust is probably the most important outcome construct in relational literature, especially because it is the building block for relationship commitment and has a long term nature (e.g. Morgan & Hunt, 1994; Doney & Cannon, 1997; Weitz & Bradford, 1999). For example, Berry (1996, p. 42) states that “trust [is] perhaps thesinglemost powerful relationship marketingtool available to a company” and, as highlighted by Gundlach and Murphy (1993, p. 41), “the variable most universally accepted as a basis of any human interactionorexchangeistrust”.Similarly,trustisakeyconstructinthe literature investigating relationship selling (e.g. Jolson, 1997; Wilson, 2000). In our context of analysis, customer trust develops thankstothe capacity of the strategic account manager to cooperate with the buyer (s) to accomplish joint solutions (i.e. synergistic solutions) and to fulfill the different facets of his/her role (i.e. role performance). Moreover, based on relational literature, relationship selling performance should include both present and future indicators of the customer's relation- ship with the salesperson. In this perspective, role performance and synergistic solutions represent measures of present results reached either individually or jointly. In other words, they assess the current state of the strategic account manager's performance. Conversely, customer trust provides a crucial insight into the future stability and potential growth of the relationship. In fact, as highlighted by Swan, Bowers and Richardson (1999), the concept of trust includes expectations about the future and expresses the customer's will- ingness to accept the risk of undesirable outcomes if the salesperson lacks the ability or motivation to keep his/her promises. The following paragraphs examine the conceptualization of each construct in our model and develop hypotheses. 2.1. Strategic account managers' relational behaviors The exogenous variables in the model can be defined as selling behaviors or, as suggested by Crosby, Evans and Cowles (1990), as behavioral tendencies exhibited by representatives of vendors affecting their behavioral performance. One aspect of the strategic account manager position is to act individually to manage the relationship with the accounts. According to the literature, in dealing with customers, he or she may adopt two distinct sets of behaviors referred as customer and selling orientations (Saxe & Weitz,1982). However, it has also been noted by manyauthors (Wotruba & Castleberry,1993; Weitz & Bradford, 1999) that the strategic account manager position is characterized by a second aspect, reflecting the role of a coordinator of a sales team. Teams are used because in a strategic account setting, the selling process usually goes beyond the capabilities of any one individual (Workman et al., 2003). As a result, strategic account managers are also in charge of managing the activities of teams rather than simply managing their ownpersonal activities (Weitz & Bradford,1999). We therefore included in our study a third class of behaviors called team selling. 2.1.1. Customer orientation–selling orientation Customer oriented selling can be interpreted as the implementation of the marketing concept at the level of the individual salesperson. As such, this selling approach is consistent with the building of long-lasting posi- tive relationships between the buyer and the seller. Therefore, it is widely recognized as an important class of relational selling behaviors (Flaherty, Dahlstrom & Skinner, 1999; Keillor, Parker & Pettijohn, 2000; Martin & Bush, 2003;Schultz &Good,2000),butquitesurprisinglyit hasnever been investigatedina typicalrelationalcontextlikekeyaccountmanagement.It is noteworthy that the SOCO scale incorporates two sub-scales, i.e. selling orientation and customer orientation: although strongly (and negatively) interrelated, these two factors have rarely been investigated separately. In this study, we will develop and test opposing hypotheses for the two sub- scales.Importantly,althoughSOCO has beenconceptualized inmany ways (e.g. as an attitude, a measure of performance, etc.), it is best viewed as a set of behaviors (Schwepker, 2003). 2.1.2. Team selling Teamsellingimpliesthatmultiplecontactsareestablishedbetweenthe supplier and the buyer (Cron & DeCarlo, 2006). It is important to underline that such contacts usuallytake place and need tobe coordinated effectively and efficiently even when formal sales teams do not exist in the selling firm. In this study, this construct includes the strategic account manager's ability to identify experts within his/her organization and deploy them to help the customer. Although similar to the idea of entrepreneurial ability offered by Sengupta et al. (2000), our conceptualization goes further and incorporates the efforts made by the strategic account manager to coordinate the resources mobilized. Team selling is usually adopted in complex buyer–seller situations, where the customer requires special treatment and the potential sale is large, i.e. typical conditions of relationships with strategic and keyaccounts (Jackson, Widmier, Giacobbe, &Keith,1999). As themembers of thesales teamusuallybelongtodifferent departments, the necessity of a strong coordination needs to be emphasized (Honeycutt, 1996). Therefore, strategic account managers are typically described as captains or leaders of teams who are in charge of selecting technicians and other specialists who then meet with specialists on the buying side (Georges & Eggert, 2003). 2.2. Strategic account managers' relational outcomes Consistently with the goal to extend the previous work by Schultz and Evans (2002), our study explores three relational outcomes (role performance, synergistic solutions and trust in the strategic account manager) of long term buyer–seller interactions. Role performance and synergistic solutions can be classified as pertaining to the individual's effectiveness (i.e. role performance), on the one hand, and the combined effort of the two parties (i.e. synergistic solutions), on the other. As pointed out beforehand, trust in the strategic account manager (i.e. a critical aspect of relationship success) is interpreted as the final relational outcome variable. 2.2.1. Role performance This construct describes the strategic account manager's individual effectiveness in developing plans and providing expertise to the customer 302 P. Guenzi et al. / Industrial Marketing Management 38 (2009) 300–311
- 4. (Frazier & Rody, 1991). It represents the extent to which he/she perceives that the relationship has been effective in achieving performance objectives. In this way, role performance synthesizes his/her individual effectiveness in solving the customer's problems, developing strategic plans, and enhancing outcomes (Schultz & Evans, 2002, p. 25). 2.2.2. Synergistic solutions In contrast with role performance, which covers the individual performance of the strategic account manager, this construct represents the combined effort of the two parties to be innovative and accomplish joint solutions that could not be obtained by either individual in the absence of the partner (Jap & Weitz, 1994; Schultz & Evans, 2002). Implementing a relational strategy requires interactive, two-way exchange processes between the seller and the buyer, aimed especially at developing interdependence and co-production of value (Möller & Wilson, 1995). Similarly, Wilson (1995) points out that maintaining the buyer–seller relationship over time relies on mutual goals and joint action, while Ganesan (1994) underlines that at the heart of relational exchanges there are joint synergies resulting from exploiting idiosyncratic assets and risk sharing. 2.2.3. Trust in the strategic account manager In the present research, we define trust as the customer's belief that the strategic account manager can be expected to show reliability and integrity (Morgan & Hunt, 1994) and forbearance from opportunism (Smith & Barclay,1997) in future interactions. Our conceptualization is limited to the customer's trust in a person and does not take into consideration trust in the selling organization. Adopting a long-term perspective emphasizes the vital importance of trust, a construct which synthesizes and represents both the output of past relations and the input for future ones (Anderson & Narus, 1990; Morgan & Hunt, 1994). In short, trust is a focal construct in the analysis of relational strategies (Ganesan, 1994; Doney & Cannon, 1997; Blois, 1999). 2.3. Hypotheses It is generally recognized that customer orientation increases performance (Saxe & Weitz, 1982; Swenson & Herche, 1994; Keillor et al., 2000; Boles, Babin, Brashear, & Brooks, 2001; Brown, Mowen, Donavan, & Licata, 2002) and effectiveness (Baldauf & Cravens, 1999). However, to our knowledge, empirical research on this topic to date has not investigated the impact of customer orientation on the specific outcomes we take into account in this research. We argue that there is a large body of evidence which links customer orientation to relational-like outcomes. Humphreys and Williams (1996) and Goff, Boles, Bellenger & Stojack (1997) have demonstrated that customer orientation generates higher levels of customer satisfaction with the salesperson, thus increasing his/her role performance. Jones, Busch and Dacin (2003) showed that customer orientation reduces a customer's propensity to switch to other suppliers. Boles, Barksdale and Johnson (1996), Beverland (2001) and Liu and Leach (2001) found that a salesperson's relational behaviors which are oriented toward the customer improve the customer's perception of the quality of the buyer–seller relationship. To sum up, we can conclude that customer orientation increases the selling company's relational outcomes by improving the quality of the customer's interpersonal relationship with the sales force (Langerak, 2001). As customer orientation reflects non-opportunistic behavior which stresses customer-focused solu- tions and mutual benefits (Schwepker, 2003), we argue that it should positively affect both the strategic account manager's role perfor- mance and provision of synergistic solutions, thereby ultimately increasing customer trust. Building on the above-cited theoretical and empirical foundations, we hypothesize that: H1. The strategic account manager's customer orientation is posi- tively related to his/her role performance. H2. The strategic account manager's customer orientation is posi- tively related to his/her provision of synergistic solutions. Interestingly, it has been pointed out that the SOCO scale consists of two distinct, though related, subscales (Thomas, Soutar and Ryan, 2001). Nevertheless, very few studies have examined simultaneously the impact of customer orientation on the one hand, and the impact of selling orientation on the other, on selected outcome variables. Among these studies, Goff et al. (1997) posits and partially supports the existence of opposing consequences of customer orientation and selling orientation on customer satisfaction. Tam and Wong (2001) demonstrates that customer orientation has a positive effect on both customer satisfaction and customer trust, while selling orientation has a negative impact on both outputs. Thus, we hypothesize that a selling orientation, compared to a customerorientation,willhaveanopposite(i.e. negative)effectonthetwo relational outcomes taken into account in this study: H3. The strategic account manager's selling orientation is negatively related to his/her role performance. H4. The strategic account manager's selling orientation is negatively related to his/her provision of synergistic solutions. Cross-functional selling teams are necessary in strategic account management since an individual salesperson does not possess the knowledge or intrafirm influence to propose and implement a program that has the potential for building a competitive advantage for the seller–buyer dyad (Weitz & Bradford, 1999). This statement clearly underlines the fact that synergistic solutions (i.e. a win–win outcome for both members of the dyad) often result from team selling. Based on this background we hypothesize: H5. The strategic account manager's adoption of team selling is positively related to synergistic solutions. The management of a business relationship with a customer encompasses a multiplicity of different tasks which require a wide range of knowledge, skills, and abilities. Therefore, the ability of the individual strategic account manager to be the selling team coordi- nator and to work closely with colleagues from other departments is widely recognized as a fundamental driver of his/her individual performance (e.g. Millman, 1996; Weilbaker & Weeks, 1997; Wilson & Millman, 2003). Hence we hypothesize that: H6. The strategic account manager's adoption of team selling is positively related to his/her role performance. Although there is no general consensus on the causal ordering of relational constructs, we argue that customer trust in the salesperson is one of the ultimate goals of the relationship-building process, as it leads to commitment (Morgan & Hunt, 1994). Therefore, in our conceptualization, synergistic solutions and role performance are most appropriately posited as predictors of trust. They stem from past behaviors and refer to the customer's evaluation of his/her relation- ship with the strategic account manager to date. By contrast, trust reflects the customer's expectations about future behaviors. Finally, customers will be more inclined to trust account managers who are effective in terms of performance and develop synergistic solutions. Based on this rationale, we hypothesize that: H7. The strategic account manager's provision of synergistic solutions is positively related to customer trust. H8. The strategic account manager's role performance is positively related to customer trust. 3. Method of the study 3.1. Data collection, questionnaire development and sample Although it could be argued that the best informants for the model tested in the study would be buyer–seller dyads, in light of the obvious 303 P. Guenzi et al. / Industrial Marketing Management 38 (2009) 300–311
- 5. selection problems to obtain participation from dyads of respondents, the survey respondents were all strategic account managers. A similar approach was also adopted in recent studies on strategic account management (see Sengupta et al., 2000; Schultz & Evans, 2002). In fact, strategic account managers base their decisions and actions upon their personal evaluations but also on their perceptions of the evaluations of third parties (sales manager or customer). In keeping with Expectancy Theory, strategic account managers will spend more effort on those behaviors which, in their belief, lead to greater performance which, in turn, should drive to greater rewards. In strategic account management, performance is usually best interpreted in terms of relationship building. Hence, both managers and academics need to understand what are the strategic account managers’ beliefs and perceptions concerning the linkages between behaviors and relational outcomes. This is why using strategic account managers as respondents should be appreciated from both a theoretical and a managerial perspective. Strategic account managers were asked how often they perform customer-oriented, selling-oriented and team selling behaviors. In addition, for the different outcomes included in our model (i.e. role performance, synergistic solutions and trust), we asked their percep- tions about the evaluations of the strategic account. Using self-reporting measures has some limitations. However Spector (2006, p.225), based on the findings of many empirical studies, claims that “evidence fails to support social desirability as a general source of correlation inflating CMV when self-reports are used”. In addition to this, Schultz and Evans (2002) argued that it is appropriate to use the key account managers' self-report performance assessment as a “proxy” interpretation of the key accounts' view because in such close, long-term relationships the understanding parties have of each other transcends the typical information available in contexts characterized by single transactions (see Ickes, 1993). Additionally, Schultz and Evans suggested that using self-reporting measures is appropriate especially when (like in our case) this self- assessment does not refer to objective criteria (e.g. quota, sales volume, etc.), but rather to qualitative dimensions (e.g. trust). Finally, it is important to underline that Schultz and Evans (2002, p.26) also surveyed 40 key accounts in their study and found that “the customer survey yielded virtually identical relationship results to those of the vendor data”. Based on these arguments, we also conclude, just like Schultz and Evans, that the use of key account managers to measure the quality of the customer relationship is appropriate. In addition to these considerations, consistently with the recom- mendations made by Podsakoff et al. (2003), we used a number of procedural and statistical remedies against the potential problems associated with common method bias and single respondent bias. First, we protected respondent anonymity by assuring complete confidentiality. Second, we reduced item ambiguity by using well established scales and by pre-testing the questionnaire. Third, we used different scale formats and scale anchors (“never–always”; “strongly disagree–strongly agree”; “very little–to a great deal”; etc.). Fourth, we used Harman's one-factor test: a principal component factor analysis on all the variables measured using the survey instrument revealed six factors with eigenvalues greater than 1, which together accounted for 73.8% of the total variance; also, the first, largest factor did not account for a majority of the variance (28%). Therefore, it can be concluded that common method bias is not likely to be a serious concern in our study (Podsakoff & Organ, 1986). To test our hypotheses, a cross-sectional survey was conducted in Italy. Thequestionnairewasoriginallydeveloped inEnglish,asalltheconstructs used in the research were based on studies in English-speaking countries. The English version of the questionnaire was translated into Italian by one expert translator and then translated back into English by a second; both translators reconciled differences. Strategic account managers were selected as respondents. The goal of the sampling process was to include sales organizations fromvarious selling environments. Using a database of former participants in executive education training programs in a major Italian management school, 220 potential respondents were selected. Although many differences arose in their job titles, they all stated they were in charge of managing relationships with their company's strategic accounts. They were informed by phone that they would be receiving the questionnaire. Participants were also assured complete confidentiality. Respondents provided data for a specific strategic account for which they were responsible. If they handled several accounts, they were asked to choose the one that best represented the above-cited definition of a strategic account. Three weeks after the initial contact, non-respondents were followed up by telephone. A total of 127 usable questionnaires were obtained, thus achieving a response rate of 57.7%. In line with our objectives, many different industries are represented, including industrial goods (34%), consumer/durable goods (28.2%), services (29.1%), pharma- ceuticals/medical (6.8%), and others (1.9%). Equally noteworthy is the fact that the sample includes organizations which vary in size. The average experience of the respondents participating in the study is 4 years in their current position, just under 7 years with their current organization, and 10.5 years in a sales position. The average number of strategic accounts of the selling firm is 37, and the average number of accounts managed per strategic account manager is 7.7. Remuneration is mainly based on fixed salary (79% of total). Commissions and bonuses are 7% and 12% respectively. As for the characteristics of the working environment, companies in the sample operate in highly competitive markets, where the sales force is of paramount importance (whereas e-commerce is not relevant), there is a wide variety of customers and sales trends are mainly stable. Early and late respondents were compared to check for a possible response bias (Armstrong & Overton,1977). A t-test of difference in means on all the constructs used in the research model showed no significant difference in means between earlyand late respondents (see Appendix A). 3.2. Measures Based on a review of the literature, we used existing scales for each construct, with the exception of team selling. In fact, although Workman, Homburg and Jensen (2003) have recently offered a new three-item scale to measure “use of teams”, no generally accepted measure of team selling exists in the sales literature. Therefore, following our review of the literature on this topic, a new five-item survey instrument was developed. To measure the customer orienta- tion and selling orientation, the ten items of the SOCO scale, as proposed by Thomas, Soutar and Ryan (2001), were used. As for role performance and synergistic solutions, we used the two scales developed by Schultz and Evans (2002). Finally, regarding trust we pooled the items used by Sengupta et al. (2000), which only refer to the customer's perceptions about the strategic account manager's honesty/benevolence, together with those adopted by Schultz and Evans (2002) which, in contrast, express the customer's perceptions about the reliability and competence of the counterpart. By doing this, our measure is consistent with the traditional definition of inter- personal trust in buyer–seller relationship, i.e. a situation where “the industrial buyer believes and feels that he can rely on what the salesperson says or promises to do in a situation where the buyer is dependent upon the salesperson's honesty and reliability (Swan & Nolan, 1985, p.40). In fact reliability and benevolence are the two key facets of this construct (Swan, Bowers & Richardson 1999). Our choice to treat interpersonal trust as a unidimensional construct including items that tap both the credibility and benevolence aspects is consistent with the approach adopted by Doney and Cannon (1997, p.43), who note that “Although credibility and benevolence could be conceptually distinct […] they may be so intertwined that in practice they are operationally inseparable”. Moreover, in keeping with the suggestion made by Swan, Bowers and Richardson, (1999), our scale incorporates items at different level of abstraction, i.e. items which refer to specific behaviors of the strategic account manager (e.g. “could be trusted to make emergency decisions, if he/she could not be reached”), to attributes broader than a specific behavior and to general 304 P. Guenzi et al. / Industrial Marketing Management 38 (2009) 300–311
- 6. trust measures (e.g. “do the job with integrity”).The guidelines used were those outlined by Churchill (1979). First, all items were submitted to five marketing scholars and three strategic account managers to ensure content validity. Participants were asked to check the clarity of each item and its capacity to reflect the underlying construct. Second, the questionnaire was pre-tested with 20 strategic account managers participating in a Strategic Account Management Executive Program at a major Italian business school. After some minor adjustments, the resulting items were included in the final survey (see Appendix B for scale items). 3.3. Model estimation The structural equation model (Model 1), represented by the path diagram in Fig. 1, was estimated using a partial least square (PLS) latent path model. PLS can accommodate small samples and it provides measurement assessment which is crucial to our study, as we have a fairly limited sample size (Wold, 1982). In addition, it avoids some of the restrictive assumptions imposed by LISREL-like models (Fornell & Bookstein, 1982). In fact, as stated by O'Loughlin and Coenders (2004), ML-SEM assumes that observations are independent and follow a multivariate normal distribution. PLS-SEM uses non- parametric inference methods (such as bootstrapping or jackknifing) and is free of these assumptions. Due to our limited sample size (127 cases), we decided to use PLS. Another advantage of the PLS approach is that it provides measurement assessment which is critical to our study as we develop new measures (Dawes and Lee, 1996). PLS has also been chosen because it is ideal for the early stages of theory development, as is the case in this research (Hulland, 1999). Finally, according to Chin (1998b), using the resampling procedures (i.e. bootstrap and jackknife) packaged in the SmartPLS software (version 2.0), one can calculate the standard deviation and generate an approximate t-statistic. This overcomes the disadvantage of non- parametric methods of having no formal significance tests for the estimated parameters. Compared to jackknifing, the bootstrap technique offers two advantages: (1) the possibility to calculate confidence intervals other than those calculated from a normal distribution, and (2) the possibility to use a larger number of samples (Chin, 1998b). Due to these reasons, we adopted the bootstrap technique. Bootstrapping was used to draw at random a bootstrap set of 127 cases. The process was repeated 200 times to obtain stable standard errors and low differences between entire sample estimates and mean of subsamples (Léger, Politis & Romano, 1992). 4. Results 4.1. Data analysis Following standard procedures (Churchill, 1979), several steps were taken to ensure reliability and validity of the multi-item scales. In the first step, reliability analysis was carried out by calculating Cronbach's alphas. For all the constructs, Cronbach's alphas exceeded the 0.7 threshold (Nunnally, 1978). In the second step, we conducted principal component analysis on each construct to check for unidimensionality. One item (custo2) measuring customer orientation was dropped, due to low indicator loadings. The PLS analysis showed satisfactory reliability, convergent validity as well as discriminant validity among constructs (see Tables 1 and 2 for detailed results). The PLS results are interpreted in two stages: by assessment of its measurement model, and by assessment of its structural model (Fornell & Larcker, 1981). Moreover, a path model can be evaluated at three levels: the quality of the measurement model, the quality of the structural model, and each structural regression equation (Tenenhaus, Esposito Vinzi, Chatelin, & Lauro, 2005). 4.1.1. Measurement model Following Cohen (1988) the epistemic relationships were specified as reflective for all constructs, suggesting that the constructs are manifest in or give rise to their measures. The properties of the measurement model are provided in Table 1, and the means, standard deviations and correlations among the constructs are shown in Table 2. All factor loadings are higher than 0.65, suggesting satisfactory item reliability. As indicated in Table 1, all composite reliability indicators are satisfactory, since they are all above the 0.7 threshold recommended by Nunnally (1978). Convergent validity was confirmed as the average variance in manifest variables extracted by constructs (AVE) was at least 0.56, indicative that more variance was explained than unexplained in the variables associated with a given construct (Fornell & Larcker, 1981). One criterion for adequate discriminant validity is that the correlation of a construct with its indicators (i.e., the square root of the AVE) should exceed the correlation between the construct and any other construct (Fornell & Larcker 1981). The findings shown in Table 2 suggest discriminant validity: in fact all diagonal elements are greater than the off-diagonal elements in the corresponding rows and columns. Overall, these measurement results are satisfactory and suggest that it is appropriate to proceed with the evaluation of the structural model. Table 1 Properties of measurement Model 1 Construct Indicators Factor loadingsa Composite reliabilityb Average variance extractedc Customer orientation custo1 0.66 0.93 0.76 custo3 0.92 custo4 0.95 custo5 0.93 Selling orientation sello1 0.68 0.86 0.56 sello2 0.75 sello3 0.72 sello4 0.74 sello5 0.83 Team selling team1 0.86 0.90 0.75 team2 0.89 team3 0.86 team4 0.75 team5 0.86 Trust trust1 0.91 0.93 0.83 trust2 0.91 trust3 0.92 trust4 0.90 Synergistic solutions syne1 0.90 0.94 0.85 syne2 0.93 syne3 0.91 Role performance role1 0.76 0.86 0.71 role2 0.86 role3 0.84 a A loading is significant when above 0.55 (Falk and Miller 1992, p. 81). b Scale reliability is considered satisfactory when the composite reliability is above 0.70 (Nunnally, 1978). c Convergent validity is considered satisfactory when the AVE is above 0.50 (Fornell & Larcker, 1981). Table 2 Construct means, standard deviations and intercorrelations for Model 1 Mean SD 1 2 3 4 5 6 1. Customer orientation 5.78 1.69 0.87 2. Role performance 5.27 0.97 0.40 0.84 3. Selling orientation 2.89 1.29 −0.28 −0.22 0.75 4. Synergistic solutions 5.51 0.98 0.44 0.50 −0.30 0.92 5. Team selling 6.04 1.01 0.43 0.32 −0.21 0.39 0.86 6. Customer trust 5.62 0.80 0.55 0.46 −0.30 0.54 0.32 0.91 Note: Bold numbers on the diagonal show the square root of the AVE; numbers below the diagonal represent construct correlations. 305 P. Guenzi et al. / Industrial Marketing Management 38 (2009) 300–311
- 7. Because, one of our goals is to test a causal order between relational outcomes in the model, we need to examine if synergistic solutions and role performance act as mediating variables between individual behaviors and customer trust. We address this issue by comparing alternative model formulations to our baseline model (i.e. Model 1). In Model 2 role performance and synergistic solutions were excluded and the three strategic account manager's behaviors were directly linked to customer trust. Model 3 includes both the direct and indirect paths from behaviors to customer trust. We will first discuss the results of Model 1 and then compare them to those of Model 2 and 3. 4.1.2. Structural model and hypothesis testing Table 3 reports the standardized parameters for Model 1, which are obtained from bootstrap simulation (Chin, 1998a). T-values confirm the significance of hypotheses H1 (B=0.30), H2 (B=0.30), H4 (B= −0.17), H5 (B=0.18), H6 (B=0.22), H7 (B=0.26) and H8 (B=0.41). One hypothesis (H3) is non-significant (B=−0.10). The structural model demonstrates predictive power as the variance explained (R2 ) in key endogenous constructs was 0.20 for role performance, 0.27 for synergistic solutions, 0.34 for customer trust. The findings show that our model (i.e. Model 1) explains a large part of the variance in the endogenous variables, with an average R2 of 0. 27. Communality and redundancy coefficients are also presented in Table 4. They can be used essentially in the same way as the R2 , since they reflect the relative amount of explained variance for latent and manifest variables. An important part of model evaluation is the examination of fit indexes reflecting the predictive power of estimated inner and outer model relationships. As pointed out by Tenenhaus et al. (2005, p.173), “… differently from SEM-ML, PLS path modeling does not optimize any scalar function so that it naturally lacks of an index that can provide the user with a global validation of the model (as it is instead with χ2 and related measures in SEM-ML). The GoF represents an operational solution to this problem as it may be meant as an index for validating the PLS model globally”. A general criterion for evaluating goodness-of-fit (GoF) is to calculate the geometric mean of the average communality and the average R2 (Tenenhaus et al., 2005). According to the results in Table 4, GoF=√[(0.73)×(0.27)]=0.44, which can be considered as satisfactory (Tenenhaus et al., 2005;). Then, the blindfolding approach proposed by Wold (1982) was followed to calculate the cv-communality and cv-redundancy indexes (see Table 5). The cv-communality index (H2 ) measures the quality of the measurement model, whereas the cv-redundancy index (i.e. Stone–Geisser's Q2 , which Tenenhaus et al. (2005, p. 174) call F2 ) measures the quality of the structural model. As shown in Table 5, the measurement model (H2 =0.54) shows a better quality than the structural one (F2 =0.16). However, also endogenous constructs are correctly estimated with a minimum F2 of 0.13 for role performance and an average F2 of 0.16. 4.1.3. The mediating role of synergistic solutions and role performance: comparing alternative models To test a causal order between relational outcomes, we compared Model 1 to Model 2 and 3. In Model 2, as shown in Table 3, the direct path between customer orientation and customer trust is significant (B=0.47) as well as the direct path between selling orientation and customer trust (B=−0.15). However, team selling has no significant direct effect on customer trust (B=0.10). In comparison with our initial model (i.e. Model 1), excluding role performance and synergistic solutions results in a drop of R2 to 0.30 for customer trust (see Table 4). The goodness of fit index (GoF) stays stable (GoF=0.45) as no major changes can be observed in term of communality. Table 5 shows a similar average cv-communality for both models (H2 =0.54 for Model 1 and H2 =0.54 for Model 2) but a higher average cv-redundancy for Model 2 (F2 =0.16 for Model 1 and F2 =0.22 for Model 2). These results indicate, on average, a better quality of the prediction of customer trust for Model 2. However, it is important to keep in mind the lower explained variance (R2 ) in comparison with Model 1. In Model 3 (see Table 4), just like in Model 1, hypotheses H1 (B=0.29), H2 (B=0.30), H4 (B=−0.16), H5 (B=0.18), H6 (B=0.22), H7 (B=0.17) and H8 (B=0.28) are significant, whereas hypothesis H3 is non-significant (B=−0.10). In this model the direct path between customer orientation and customer trust is significant (B=0.33), whereas team selling and selling orientation have no significant direct impact on customer trust (B=0.01 and B=−0.08, respectively). As indicated in Table 4, this third model shows significant increases in the explained variance for customer trust (R2 =0.44) as well as a significant increase in average redundancy (F2 =0.47). In comparison Table 3 Parameter estimation of the PLS Models by the Bootstrap method Construct Model 1a Model 2a Model 3a Dependent variable: Role performance Customer orientation 0.30b 3.20c⁎ – – 0.29 3.02⁎ Selling orientation −0.10 1.07 – – −0.10 1.55 Team selling 0.18 2.03⁎ – – 0.18 2.04⁎ Dependent variable: Synergistic solutions Customer orientation 0.30 3.23⁎ – – 0.30 3.43⁎ Selling orientation −0.17 1.94⁎ – – −0.16 2.14⁎ Team selling 0.22 3.12⁎ – – 0.22 2.98⁎ Dependent variable: Customer trust Role performance 0.26 2.74⁎ – – 0.17 1.89⁎ Synergistic solutions 0.41⁎ 4.69⁎ – – 0.28 3.04⁎ Customer orientation – – 0.47 4.22⁎ 0.33 3.00⁎ Selling orientation – – −0.15 2.18⁎ −0.08 1.04 Team selling – – 0.10 1.28 0.01 0.07 ⁎Sig. if above 1.64 for 1-tailed test. a Model 1: original model; Model 2: model without synergistic solutions and role performance; Model 3: full model including direct paths between customer orientation, team selling and customer trust. b Standardized path coefficients. c T-values. Table 4 Explained variance (R2 ), communality, redundancy and goodness-of-fit index for Models 1, 2, 3 and 4 Construct Model 1a Model 2 Model 3 Model 4 Trust 0.34b (0.83)c 0.14d 0.30 (0.83) 0.20 0.44 (0.83) 0.21 0.43 (0.83) 0.22 Role performance 0.20 (0.71) 0.11 – (–) – 0.20 (0.71) 0.11 0.20 (0.71) 0.11 Synergistic solutions 0.27 (0.85) 0.15 – (–) – 0.27 (0.85) 0.15 0.27 (0.85) 0.15 Customer orientation – (0.77) – – (0.76) – – (0.76) – – (0.76) – Selling orientation – (0.56) – – (0.56) – – (0.56) – – (0.56) – Team selling – (0.75) – – (0.75) – – (0.75) – – (0.75) – Average 0.27 (0.73e ) 0.13 0.30 (0.68) 0.20 0.30 (0.73) 0.47 0.30 (0.73) 0.47 GoFf 0.44 0.45 0.47 0.47 NOTE: a Model 1: original model; Model 2: model without synergistic solutions and role performance; Model 3: full model including direct paths between customer orientation, team selling and customer trust; Model 4: final model. b Explained variance. c Communality: Communality coefficients are equal to the squared correlations between manifest variables and their associated latent variables. d Redundancy: Redundancy coefficients reflect the joint predictive power of the inner and outer model relationships. e Computed as a weighted average of the different communalities with the weights being the number of manifest variables per each construct (see Tenenhaus et al., 2005, p. 180). f GoF=- = [(average communality) × (average R2 )]. Average communality is computed as a weighted average of the different communalities with the weights being the number of indicators per latent variable ( Tenenhaus et al., 2005). 306 P. Guenzi et al. / Industrial Marketing Management 38 (2009) 300–311
- 8. with Models 1 and 2, results indicate for Model 3 a similar goodness- of-fit index (GoF=0.47). Regarding the overall quality of the measure- ment and prediction, no significant changes can be noticed (average cv-communality=0.54 and average cv-redundancy=0.23). Starting from these results, we evaluated the mediating role of both synergistic solutions and role performance. According to Baron and Kenny (1986), to establish mediation the following conditions must be satisfied: (a) customer orientation, selling orientation and/or team selling must have a significant effect on customer trust in Model 2, (b) customer orientation, selling orientation and/or team selling must have a significant effect on role performance and/or synergistic solutions in Model 3, (c) role performance and/or synergistic solutions must have a significant effect on customer trust in Model 3, and (d) the effect of customer orientation, selling orientation and/or team selling on cus- tomer trust must be lower in Model 3 than in Model 2. There is perfect mediation if the direct effect of customer orientation, selling orientation and team selling on customer trust is not significant in Model 3. Regarding the relationship between customer orientation and customer trust (see Table 3 for the path coefficients estimates of Models 2 and 3), we found that (a) customer orientation has a significant effect on customer trust in the absence of role performance and synergistic solutions (Model 2), (b) customer orientation has a significant effect on roleperformanceand synergistic solutions(Model3),(c)roleperformance and synergistic solutions have a significant effect on customer trust (Model 3), and (d) the effect of customer orientation on customer trust is reduced (but is still significant) in the presence of role performance and synergistic solutions (Model 3). Therefore, we can conclude that the relationship between customer orientation and customer trust is partially mediated by role performance and synergistic solutions. As for the relationship between selling orientation and customer trust, there is no mediating effect of role performance as condition (b) set by Baron and Kenny (1986) is not satisfied (i.e. there is no significant relationship in Model 3 between selling orientation and role perfor- mance). However, we found that all the four mediating conditions were satisfied for synergistic solutions. Specifically (see Table 3), it can be seen that (a) selling orientation has a significant effect on customer trust in the absence of synergistic solutions (Model 2), (b) selling orientation has a significant effect on synergistic solutions (Model 3), (c) synergistic solutions has a significant effect on customer trust (Model 3), and (d) the effect of selling orientation on customer trust is reduced in the presence of synergistic solutions (Model 3). Furthermore, the path from sellingorientationto customer trust is not significant (Model 3). Therefore, we can conclude that the relationship between selling orientation and customer trust is fully mediated by synergistic solutions. Fig. 2. Final model of strategic account managers' contribution to customer trust. Table 5 Blindfolding results: cv-communality (H2 ) and cv-redundancy (F2 ) for Models 1, 2, 3 and 4 Construct Model 1a Model 2 Model 3 Model 4 Customer trust 0.69b (0.14)c 0.69 (0.22) 0.69 (0.35) 0.69 (0.34) Role performance 0.41 (0.13) – (–) 0.41 (0.13) 0.41 (0.12) Synergistic solutions 0.65 (0.21) – (–) 0.65 (0.21) 0.65 (0.21) Customer orientation 0.56 (–) 0.56 (–) 0.56 (–) 0.56 (–) Selling orientation 0.32 (–) 0.33 (–) 0.33 (–) 0.33 (–) Team selling 0.59 (–) 0.60 (–) 0.60 (–) 0.60 (–) Average 0.54 (0.16) 0.54 (0.22) 0.54 (0.23) 0.54 (0.22) NOTE: The cv-communality index (H2 ) has been described as a cross-validated R2 between the block manifest variables and their own latent variable (Tenenhaus et al., 2005, p. 174). In other words, the cv-communality measures the capacity of the path model to predict the manifest variables directly from their own latent variable by cross-validation. It uses only the measurement model. The mean of the cv-communality indexes can be used to measure the global quality of the measurement model if they are positive for all blocks of variables. The quality of each structural equation is measured, instead, by the cv-redundancy index (i.e. Stone–Geisser's Q2 ). Tenenhaus et al. (2005, p. 174) call it F2 . They define it as a kind of cross-validated R2 between the manifest variables of an endogenous latent variable and all the manifest variables associated with the latent variables explaining the endogenous latent variable, using the estimated structural model. More specifically, it measures the capacity of the path model to predict the endogenous manifest variables indirectly from a prediction of their own latent variable using the related structural relation, by cross-validation (Tenenhaus et al., 2005, p.181). The mean of the various cv- redundancy indexes (F2 ) related to the endogenous blocks can be used to measure the global quality of the structural model, if they are positive for all endogenous blocks. The model has predictive relevance if F2 N0 and, if F2 N1, the observed endogenous variables can be perfectly predicted by the model (Fornell & Bookstein, 1982). a Model 1: original model; Model 2: model without synergistic solutions and role performance; Model 3: full model including direct paths between customer orientation, team selling and customer trust; Model 4: final model. b Cv-communality (H2 ). c Cv-redundancy (F2 ). 307 P. Guenzi et al. / Industrial Marketing Management 38 (2009) 300–311
- 9. Finally, the relationship between team selling and customer trust is not mediated by role performance nor by synergistic solutions as con- dition (a) set by Baron and Kenny (1986) is not satisfied (i.e. the path between team selling and customer trust is not significant in Model 2). Fig. 2 depicts the final model (Model 4), as well as the path coefficients, t-tests and R2. Additional results are provided in Tables 4 and 5. Our final model is identical to Model 3, except for the fact that the non significant paths are not included. 5. Conclusion By means of an empirical research on 127 firms, this article explored how strategic account managers can contribute to the development of customer trust in a relational context. Our study provides insights into the complex interrelationships between some of the strategic account manager's behaviors and relevant relational outcomes. More specifically, our findings show that strategic account managers can positively affect trust in two ways. First, they can provide synergistic solutions by successfully improving the combined effort of the two parties to accomplish innovative joint solutions. Secondly, they can perform their task as expected by the strategic account, thus attaining a good role performance. As shown in Fig. 2, both causal paths are significant, with synergistic solutions having a stronger impact (B=0.30) than role performance (B=0.17). Together, with the direct impact of customer orientation, they explain almost half of the variance in customer trust (R2 =0.43), indicating that we identified key drivers of customer trust (see Fig. 2). To enhance the attainment of synergistic solutions, strategic account managers should increase their customer orientation and promote team selling (B=0.30 and 0.22 respectively). Our findings suggest that adopting a selling orientation significantly decreases the provision of synergistic solutions (B=−0.16). As indicated in Fig. 2, taken together, customer orientation, selling orientation and team selling explain nearly one third of the variance in synergistic solutions (R2 =0.27). To improve their role performance, strategic account managers should be customer oriented and reinforce team selling, as both paths are significant and positive (B=0.32 and B=0.19). No significant impact on the strategic account manager's role performance emerged for selling orientation (H3 is not supported). Because the variance explained in role performance is quite limited (19%), additional research is needed to better identify its drivers. Our results also highlight the mediating effect of role performance and synergistic solutions. Prior studies, such as the one developed by Schultz and Evans (2002), have neither conceptually investigated nor formally tested this mediating role. We extended prior research by testing this role. Interestingly, we found differences. While the role performance construct plays a (partially) mediating role between customer orientation and customer trust, it does not play this role between selling orientation or team selling and customer trust. Furthermore, synergistic solutions fully mediate the relationship between selling orientation and customer trust and partially mediate the one between customer orientation and customer trust. However, synergistic solutions do not play a mediating role between team selling and customer trust. 5.1. Research contribution Homburg, Workman and Jensen (2000) identified strategic account management as a highly important subject for academic research and highlighted the scarcity of studies on this area. However, with very few notableexceptions(Boles,Barksdale&Johnson,1996;Abratt&Kelly,2002) the specific topic of strategic account managers' behaviors is almost unexplored. In addition to addressing some important limitations in the study by Schultz and Evans (2002), the theoretical contribution of our research to current knowledge can be summarized as follows. First, this research confirms the general assumption that strategic account managers can significantly contribute to the creation of trust in business relationships. More specifically, this study identifies two relational outcomes which are antecedents of customer trust, and some of their behavioral drivers. Findings of our study clearly show that the adoption of some specific classes of strategic account managers' behaviors (i.e. customer orientation, team selling) can contribute to the creation of strong and long-lasting positive relationships with customers, by increasing relational outcomes, thus finally fostering customer trust. Our study also suggests that selling orientation significantly decreases synergistic solutions and hence customer trust. Second, our research is one of the very few empirical studies on selling behaviors carried out in non-English speaking countries. This contribution is appreciable, as cross-cultural validation of constructs has sometimes led to controversial findings (e.g. Herche, Swenson & Verbeke, 1996). Third, our study tests separately distinct hypotheses for the two dimensions of the SOCO scale, and it finds evidence of opposite impacts for each of them. Finally, we developed and tested a new construct (team selling) which is particularly relevant and consistent with the relational approach, as well as with literature on strategic account management. Although the association between team selling and relational outcomes has already been postulated by anecdotal evidence and theoretical articles, the importance of its empirical validation should not be underestimated. 5.2. Managerial contribution This study provides sales managers with some evidence of those behaviors strategic account managers should adopt to successfully fulfill their role as contributors in building long-term relationships. The general managerial implication is that companies willing to build and foster long term relationships with their accounts should facilitate and stimulate the adoption of these behaviors on the part of their strategic account managers. Our findings highlight the importance of team selling. Companies may facilitate the adoption of team selling by improving cross functional relationships in the selling firm. This implies increasing inter-functional integration through appropriate changes at four different levels: company culture, organization structure, integration mechanisms and personal characteristics of individuals (Rouziés et al., 2005). Hence, companies should first of all carefully select candidates for strategic account manager positions, investigating their skills and attitudes (e.g. team playing). Secondly, companies should design training programs specifically aimed at helping strategic account managers to develop those skills, abilities and competencies (e.g. conflict handling) which are necessary to successfully interact with colleagues from different departments. Then, team-based rewardsandincentives should beadopted in ordertoincreasecooperation in teams(Yilmaz & Hunt, 2001). Finally,strategic account managersshould be supported by integration mechanisms aimed at increasing cross- functional collaboration and information exchange (Kahn and Mentzer, 1998). Our findings also show the importance of synergistic solutions in a strategic account management setting. Developing synergistic solu- tions implies both understanding the customer's value system (which in turn typically requires an investment in information systems) and finding creative solutions to customer problems. In addition to critical characteristics of the individual strategic account manager such as creativity and entrepreneurship (Wilson & Millman, 2003) the latter can be fostered by a corporate culture which stimulates risk taking and empowerment (Anderson & Huang, 2006). From a strategic standpoint, investing in the building of long term relationships may imply the sacrifice of short-term outcomes. This has important implications both for firms and individual key account managers. Companies should modify their goal setting procedures as well as their performance evaluation systems. Findings of our research suggest that when designing reward schemes for strategic account managers, sales managers may, at least in part, evaluate their behavioral performance and measure indicators of relational performance, such as 308 P. Guenzi et al. / Industrial Marketing Management 38 (2009) 300–311
- 10. customer retention rate or customer satisfaction (Sharma,1997). Also, the shift from short term to long term goals typically implies accepting a higher level of risk, because investments on customer relationships often pay off only in the long run. To better manage risk, companies and key account managers should invest on customer portfolio analysis and management to optimizethe allocationof scarceresources. Increasingrole performance, building synergistic solutions and customer trust is costly, and the return on such investments in relationships with customers is often difficult to predict and hard to measure. Hence, the ability of key account managers to invest discretionary resources on selected customers with high potential and low risk becomes increasingly crucial (Pardo, Salle & Spencer, 1995; Ryals & Knox, 2005). To sum up, the emphasis on a trust-building, long-term oriented relationship selling approach suggests the need to achieve greater control over the actual implementation of relational behaviors on the part of their strategic account managers (see for example Cravens et al., 1993; Grant & Cravens, 1996). In turn, this may call for radical changes in the organizational structure of the sales department, as well as in sales force control systems. For example, companies may design strategic account management structures or formal sales teams. Similarly, firms may shift from an independent to an employee sales force (Dishman, 1996; Ross, Dalsace & Anderson, 2005), or from outcome-based to behavior-based sales force control systems. 5.3. Limitations and directions for future research This study has several limitations. First, the relatively small sample size can be regarded as a limitation. In fact we did not take into account differences in strategic account management typologies. Using larger samples should allow researchers to explore the potential impact of the adoption of different approaches in implementing strategic account management programs (Homburg et al., 2002). By definition, however, strategic account relationships are not numerous, which makes large-scale research in this field difficult. Instead of neglecting empirical research and relying on conceptual frameworks only, we recommend the application of statistical methods that are particularly well suited for small samples (e.g. PLS and the bootstrap method). In this way, complex models can still be estimated in a stable manner. Second, we focused on strategic account managers as key informants and used self-reporting measures. Although we used a number of procedural and statistical remedies against the potential problems associated with common method bias and single respondent bias, future studies on the topic should ideally use dyads of respondents. Third, as this study is cross-sectional, assigning cause must be done with caution. Since it is focused on a dynamic phenomenon (i.e. relationships), a longitudinal study would be more appropriate (Frankwick, Porter & Crosby, 2001). Finally, the measurement scales adopted in this study, although used successfully in previous research on the topic, may not fully capture all the facets of the underlying constructs. Hence, in future research, more comprehensive measures would be welcome. Many research implications can be derived from this study. Future research on the topic should broaden our framework by including other classes of relational behaviors and possibly “transactional” behaviors (e.g. hard selling techniques) which may decrease customer trust (Hawes, Strong & Winick, 1996). The simultaneous consideration of opposing behaviors could allow the comparison of their different impacts on relevant relational outcomes. Similarly, different measures of performance could also be considered, e.g. by comparing the impact of relational behaviors on long-term versus short-term performance indicators. In fact, relational behaviors may pay off only in the long run, and may even be detrimental to immediate sales. In our study we focused on “soft” outcomes. This could be regarded as a limitation. However, it should be noted that hard measures may not be very meaningful in KAM relationships, because vendors very often invest on such relationships for other reasons (e.g. strengthening their image as suppliers of top firms in the market, fostering innovation by acquiring their customer's knowledge and capabilities, etc.) and because in many cases in these relationships economic performance may be evaluated only in the long run. Wengler, Ehret and Saab (2006) empirically found that the vast majority of companies adopting KAM do not use hard measures of performance neither to set goals nor to evaluate performance. Findings of this study clearly show that performance evaluation is almost exclusively basedonsoftoutcomeslikecustomertrustandcustomersatisfaction.Using hard measures of performance in KAM research is very difficult and may even be misleading. Future studies on the topic willing to incorporate hard measures of performance should probablyalso investigate the goals behind the adoption of KAM in the specific sample of vendors used in the analysis. Moreover, there is a need to understand more clearly the organizational and personal factors supporting the adoption of relational behaviors on the part of strategic account managers. In terms of organizational variables, we suggest focusing the attention on the impact of sales force control systems and training programs on strategic account managers' behaviors. As for personal variables, the examination of the influence of personality traits and skills of strategic account managers would be beneficial. Finally, we stress the importance of making international comparisons, by means of replications and extensions of research in different cultural contexts. Acknowledgments The authors wish to thank the anonymous reviewers for their helpful comments and suggestions. Appendix A T-test of difference in means of the constructs between early and late respondents Construct T Sig⁎ Customer orientation −0.91 0.39 Selling orientation −0.65 0.52 Team selling −0.29 0.77 Customer trust −0.63 0.54 Role performance 0.80 0.44 Synergistic solutions −0.36 0.72 Appendix B Scale Items Construct Measure description Customer orientationa For each statement, please indicate how often you act as described with the strategic account you selected… I try to figure out the strategic account's needs (custo1) I have the strategic account's best interest in mind (custo2)⁎ I take a problem solving approach in selling products or services to the strategic account (custo3) I recommend products or services that are best suited to solving problems (custo4) I try to find out which kinds of products or services would be most helpful to the strategic account (custo5) Selling orientationa For each statement, please indicate how often you act as described with the strategic account you selected… I try to sell as much as I can, rather than satisfying the strategic account (sello1) I find it necessary to stretch the truth in my sales presentations (sello2) I try to sell as much as I can to convince the strategic account to buy, event if it is more than wise customers would buy (sello3) I paint too rosy a picture of the products or services to make them (continued on next page) (continued on next page) 309 P. Guenzi et al. / Industrial Marketing Management 38 (2009) 300–311
- 11. Appendix B (continued) Construct Measure description sound as good as possible (sello4) I make recommendations based on what I can sell and not on the basis of the strategic account's long-term satisfaction (sello5) Team sellingb Please indicate how much you agree with the following statements… I help this strategic account to get in touch with the different specialists of my firm when needed (team1) I place different experts from my organization at the disposal of this strategic account (team2) I organize visits and meetings between the different departments of both companies (supplier and strategic account) (team3) I share information about this strategic account with my colleagues from other departments (team4) I spend time coordinating the different employees of my firm involved in the relationship with this strategic account (team5) Trustb The strategic account you selected would say about you that… You can be trusted to get the job done right (trust1) You could be trusted to make emergency decisions, if nobody could be reached (trust2) You do your job with integrity (trust3) You will not undertake any actions to harm this strategic account interests (trust4) Synergistic solutionsb The combined results of this person and me have resulted in… Innovative solutions to problems (syne1) Ways to get better results for both companies (syne2) Accomplishments that neither of us could have achieved individually (syne3) Role performancec In your point of view, how much would the strategic account you selected say you have helped him/her… To solve problems (role1) To develop strategic plans that enable both parties to win (role2) To enhance outcomes (role3) ⁎Item deleted based on refinement procedures described in the text. a Measured on a 9 point scale ranging from “Never” to “Always”. b Measured on a 7 point scale ranging from “Strongly disagree” to “Strongly agree”. c Measured on a 7 point scale ranging from “Very little” to “A great deal”. References Abratt, R., & Kelly, P. M. (2002). Customer-supplier partnerships. Perceptions of a successful key account management program. Industrial Marketing Management, 31 (5), 467−476. Anderson, R. E., & Huang, W. (2006). Empowering salespeople: Personal, managerial, and organizational perspectives. Psychology and Marketing, 23(2), 139−159. Anderson, J. C., & Narus, J. A. (1990 January). A model of distributor firm and manufacturer firm working partnerships. Journal of Marketing, 54, 42−58. Armstrong, J. S., & Overton, T. S. (1977 August). Estimating nonresponse bias in mail surveys. Journal of Marketing Research, 14, 396−402. Baldauf, A., & Cravens, D. W. (1999 January). Improving the effectiveness of field sales organizations. Industrial Marketing Management, 28, 63−72. Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychology research: conceptual, strategic and statistical considerations. Journal of Personality and Social Psychology, 51(6), 1173−1182. Berry, L. L. (1996 Spring). Retailers with a future. Marketing Management, 5, 39−46. Beverland, M. (2001). Contextual influences and the adoption and practice of relationship selling in a business-to-business setting: An exploratory study. Jour- nal of Personal Selling and Sales Management, 21(3), 207−216. Biong, H., & Selnes, F. (1996). The strategic role of the salesperson in established buyer– seller relationships. Journal of Business-to-Business Marketing, 3(3), 39−78. Blois, K. J. (1999). Trust in business to business relationships: An evaluation of its status. Journal of Management Studies, 36(1), 197−215. Boles, J. S., Barksdale, H. C., Jr., & Johnson, J. T. (1996). What national account decision makers would tell salespeople about building relationships. Journal of Business and Industrial Marketing, 11(2), 6−19. Boles, J. S., Babin, B. J., Brashear, T. G., & Brooks, C. (2001). An examination of the relationships between retail work environments, salesperson selling orientation–customer orientation and job performance. Journal of Marketing Theory & Practice, 9(3), 1−13. Brown, T. J., Mowen, J. C., Donavan, D. T., & Licata, J. W. (2002 February). The customer orientation of service workers: Personality trait effects on self and supervisor performance ratings. Journal of Marketing Research, 39, 110−119. Chin, W. W. (1998). The partial least squares approach for structural equation modeling. In G. A. Marcoulides (Ed.), Modern Methods for Business Research London: Laurence Erlbaum Associates. Chin, W. W. (1998 March). Issues and opinion on structural equation modeling. MIS Quarterly, 22, 7−16. Churchill, G. A. (1979 February). A paradigm for developing better measures of marketing constructs. Journal of Marketing Research, 16, 64−73. Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Erlbaum 1988. Colletti, J. A., & Tubridy, G. S. (1987 August). Effective major account sales management. Journal of Personal Selling and Sales Management, 7, 1−10. Cravens, D. W., Ingram, T. N., LaForge, R. W., & Young, C. E. (1993). Behaviour-based and outcome-based salesforce control systems. Journal of Marketing, 57(4), 47−59. Cron, W., & DeCarlo, T. (2006). Sales Management (9th ed.). Hoboken, NJ: Wiley & Sons. Crosby, L. A., Evans, K. R., & Cowles, D. (1990 July). Relationship quality in services selling: An interpersonal influence perspective. Journal of Marketing, 54, 68−81. Dawes, P. L., & Lee, D. Y. (1996). Communication intensity in large-scale organizational high technology purchasing decisions. Journal of Business-to-BusinessMarketing3 (3),pp. 3–34. Dishman, P. (1996 September). Exploring strategies for companies that use manufacturers' representatives as their sales force. Industrial Marketing Management, 25, 453−461. Doney, P. M., & Cannon, J. P. (1997 April). An examination of the nature of trust in buyer– seller relationships. Journal of Marketing, 61, 35−51. Falk, R. F., & Miller, N. B. (1992). A Primer for Soft Modeling. Akron, Ohio: The University of Akron Press. Flaherty, T. B., Dahlstrom, R., & Skinner, S. J. (1999 Spring). Organizational values and role stress as determinants of customer-oriented selling performance. Journal of Personal Selling and Sales Management, 19, 1−18. Fornell, C., & Bookstein, F. L. (1982). A comparative analysis of two structural equation models: LISREL and PLS applied to market data. In C. Fornell (Ed.), A Second Generation of Multivariate Analysis. New York: Praeger. Fornell, C., & Larcker, D. F. (1981 February). Evaluating structural equation models with unobservable variables and measurements errors. Journal of Marketing Research, 18, 39−50. Frankwick, G., Porter, S. S., & Crosby, L. A. (2001 Spring). Dynamics of relationship selling: A longitudinal examination of changes in salesperson–customer relation- ship status. Journal of Personal Selling and Sales Management, 21, 135−147. Frazier, G. L., & Rody, R. C. (1991 January). The use of influence strategies in interfirm relationships in industrial products channels. Journal of Marketing, 55, 52−69. Ganesan, S. (1994 February). Determinants of long-term orientation in buyer–seller relationships. Journal of Marketing, 58, 1−19. Georges, L., & Eggert, A. (2003). Key account manager's role within the value creation process of collaborative relationships. Journal of Business-to-Business Marketing, 10(4), 1−22. Goff, B. G., Boles, J. S., Bellenger, D. N., & Stojack, C. (1997 Summer). The influence of salesperson selling behaviors on customer satisfaction with products. Journal of Retailing, 73, 171−183. Grant, K., & Cravens, D. W. (1996 September). Examining sales force performance in organizations that use behavior-based sales management processes. Industrial Marketing Management, 25, 361−371. Gundlach, G. T., & Murphy, P. E. (1993). Ethical and legal foundations of relational marketing exchanges. Journal of Marketing, 57(4), 35−46. Hawes, J. M., Strong, J. T., & Winick, B. S. (1996 September). Do closing techniques diminish prospect trust? Industrial Marketing Management, 25, 349−360. Herche, J., Swenson, M. J., & Verbeke, W. (1996 July). Personal selling constructs and measures: Emic versus Etic approaches to cross-national research. European Journal of Marketing, 30, 83−97. Homburg, C., Workman, J. P., & Jensen, O. (2000 Fall). Fundamental changes in marketing organization: The movement toward a customer-focused organizational structure. Journal of the Academy of Marketing Science, 28, 459−478. Homburg, C., Workman, J. P., & Jensen, O. (2002 April). Configurational perspective on key account management. Journal of Marketing, 66, 38−60. Honeycutt, E. D., Jr. (1996 September). Introduction to the special issue on selling and sales management. Industrial Marketing Management, 25, 323−325. Hulland, J. (1999). Use of Partial Least Squares (PLS) in strategic management research: A review of four recent studies. Strategic Management Journal, 20, 195−204. Humphreys, M., & Williams, M. R. (1996 Summer). Exploring the relative effects of salesperson interpersonal process attributes and technical product attributes on customer satisfaction. Journal of Personal Selling and Sales Management, 16, 47−57. Iacobucci, D., & Ostrom, A. L. (1996 February). Commercial and interpersonal relationships: Using the structure of interpersonal relationships to understand individual-to-individual, individual-to-firm, and firm-to-firm relationships in commerce. International Journal of Research in Marketing, 13, 53−72. Ickes, W. (1993). Empathic accuracy. Journal of Personality, 61, 587−610. Jackson, D. W., Jr., Widmier, S. M., Giacobbe, R., & Keith, J. E. (1999 March). Examining the use of team selling by manufacturers' representatives. Industrial Marketing Management, 28, 155−164. Jap, S. D., & Weitz, B. A. (1994). Functional, relational, and strategic long-term, buyer– supplier relationships. Working Paper. Jolson, M. A. (1997 Fall). Broadening the scope of relationship selling. Journal of Personal Selling and Sales Management, 17, 75−88. Jones, E., Busch, P., & Dacin, P. (2003 April). Firm market orientation and salesperson customer orientation: Interpersonal and intrapersonal influences on customer service and retention in business-to-business buyer–seller relationships. Journal of Business Research, 56, 323−340. Kahn, K. B., & Mentzer, J. T. (1998). Marketing's integration with other departments. Journal of Business Research, 42(1), 53−62. Keillor, B. D., Parker, R. S., & Pettijohn, C. E. (2000). Relationship-oriented characteristics and individual salesperson performance. Journal of Business and Industrial Market- ing, 15(1), 7−22. Langerak, F. (2001 September). Effects of market orientation on the behaviors of salespersons and purchasers, channel relationships, and performance of manu- facturers. International Journal of Research in Marketing, 18, 221−234. Léger, C., Politis, D., & Romano, J. P. (1992 November). Bootstrap technology and applications. Technometrics, 34, 378−398. 310 P. Guenzi et al. / Industrial Marketing Management 38 (2009) 300–311
- 12. Leigh, T. W., & Marshall, G. W. (2001 Spring). Research priorities in sales strategy and performance. Journal of Personal Selling and Sales Management, 21, 83−93. Leuthesser, L. (1997 May). Supplier relational behavior: An empirical assessment. In- dustrial Marketing Management, 26, 245−254. Liu, A. H., & Leach, M. P. (2001 Spring). Developing loyal customers with a value-adding sales force: Examining customer satisfaction and the perceived credibility of consultative salespeople. Journal of Personal Selling and Sales Management, 21, 147−157. Martin, C. A., & Bush, A. J. (2003). The potential influence of organizational and personal variables on customer-oriented selling. Journal of Business and Industrial Marketing, 18(2), 114−132. Mavondo, F. T., & Rodrigo, E. M. (2001 May). The effect of relationship dimensions on interpersonal and interorganizational commitment in organizations conducting business between Australia and China. Journal of Business Research, 52, 111−121. Miller, R. B., Heiman, S. E., & Tuleja, T. (1991). Successful Large Account Management. London: Kogan Page. Millman, T. F. (1996 December). Global key account management and systems selling. International Business Review, 5, 631−645. Möller, K., & Wilson, D. T. (1995). Business marketing: An Interaction and network perspective. Boston, MA: Kluwer. Morgan, R. M., & Hunt, S. D. (1994 July). The commitment-trust theory of relationship marketing. Journal of Marketing, 58, 20−38. Napolitano, L. (1997 Fall). Customer–supplier partnering: A strategy whose time has come. Journal of Personal Selling and Sales Management, 17, 1−8. Nunnally, J. C. (1978). Psychometric Theory. New York: McGraw-Hill. O'Loughlin, C., & Coenders, G. (2004 November/December). Estimation of the European customer satisfaction index: Maximum likelihood versus partial least squares; Application to postal services. Total Quality Management, 15, 1231−1255. Pardo, C. (1997 Fall). Key account management in the business to business field: The key account's point of view. Journal of Personal Selling and Sales Management, 17, 17−26. Pardo, C., Salle, R., & Spencer, R. (1995). The Process of key accountization of the firm – A case study. Industrial Marketing Management, 22(2), 123−134. Plank, R. E., & Reid, D. A. (1994). The mediating role of sales behaviors: an alternative perspective of sales performance and effectiveness. Journal of Personal Selling & Sales Management, 14(2), 43−56. Platzer, L. C. (1984). Managing National Accounts. Conference Board Report. New York: The Conference Board Inc. Podsakoff, P. M., & Organ, D. W. (1986). Self-reports in Organizational Research: Problems and Prospects. Journal of Management, 12(4), 531−544. Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879−903. Ross, W. T., Dalsace, F., & Anderson, E. (2005). Should you set upyourown sales force or should you outsource it? Pitfalls in the standard analysis. Business Horizons, 48(1), 23−36. Rouziès, D., Anderson, E., Kohli, A. K., Michaels, R. E., Weitz, B. A., & Zoltners, A. A. (2005). Sales and marketing integration: A proposed framework. Journal of Personal Selling and Sales Management, 25(2), 113−122. Ryals, L. J., & Knox, S. (2005). Measuring risk-adjusted customer lifetime value and its impact on relationship marketing strategies and shareholder value. European Journal of Marketing, 39(5/6), 456−472. Saxe, R., & Weitz, B. A. (1982 August). The SOCO scale: A measure of the customer orientation of salespeople. Journal of Marketing Research, 19, 343−351. Schultz, R. J., & Good, D. J. (2000). Impact of consideration of future sales consequences and customer-oriented selling on long-term buyer–seller relationships. Journal of Business and Industrial Marketing, 15(4), 200−215. Schultz, R. J., & Evans, K. R. (2002 Winter). Strategic collaborative communication by key account representatives. Journal of Personal Selling and Sales Management, 22, 23−31. Schwepker, C. H., Jr. (2003 Spring). Customer-oriented selling: A review, extension, and directions for future research. Journal of Personal Selling and Sales Management, 23, 151−171. Sengupta, S., Krapfel, R. E., & Pusateri, M. A. (1997 Fall). Switching costs in key account relationships. Journal of Personal Selling and Sales Management, 17, 9−16. Sengupta, S., Krapfel, R. E., & Pusateri, M. A. (2000 Fall). An empirical investigation of key account salesperson effectiveness. Journal of Personal Selling and Sales Management, 20, 253−261. Shapiro, B. P., & Moriarty, R. T. (1984). Support systems for national account management programs: Promises made, promises kept.Marketing Science Institute Report(84–102) 38 pages. Sharma, A. (1997 Spring). Customer satisfaction-based incentive systems: Some managerial and salesperson considerations. Journal of Personal Selling and Sales Management, 17, 61−70. Smith, J. B., & Barclay, D. W. (1997). The effects of organizational differences and trust on the effectiveness of selling partner relationships. Journal of Marketing, 61(1), 3−21. Spector, P. E. (2006). Method variance in organizational research. Truth or urban legend? Organizational Research Methods, 9(2), 221−232. Swan, J. E., Bowers, M. R., & Richardson, L. D. (1999). Customer trust in the salesperson: An integrative review and meta-analysis of the empirical literature. Journal of Business Research, 44(2), 93−107. Swan, J. E., & Nolan, J. J. (1985). Gaining customer trust: A conceptual guide for the salesperson. Journal of Personal Selling and Sales Management, 5(4), 39−48. Swenson, M. J., & Herche, J. (1994). Social values and salesperson performance: An empirical examination. Journal of the Academy of Marketing Science, 22(3), 283−289. Tam, J. L. M., & Wong, Y. H. (2001). Interactive selling: A dynamic framework for services. Journal of Services Marketing, 15(5), 379−396. Tenenhaus, M., Esposito Vinzi, V., Chatelin, Y. M., & Lauro, C. (2005). PLS path modelling. Computational Statistics & Data Analysis, 48, 159−205. Thomas, R. W., Soutar, G. N., & Ryan, M. M. (2001). The selling orientation-customer orientation (S.O.C.O.) scale: A proposed short form. Journal of Personal Selling and Sales Management, 21(1), 63−69. Walter, A., Ritter, T., & Gemünden, H. G. (2001). Value creation in buyer–seller relationships – Theoretical considerations and empirical results from a supplier's perspective. Industrial Marketing Management, 30(4), 365−377. Weilbaker, D. C., & Weeks, W. A. (1997). The evolution of national account management: A literature perspective. Journal of Personal Selling and Sales Management, 17(4), 49−59. Weitz, B. A., & Bradford, K. D. (1999). Personal selling and sales management: A relationship marketing perspective. Journal of the Academy of Marketing Science, 27(2), 241−254. Wengler, S., Ehret, M., & Saab, S. (2006). Implementation of key account management: Who, why, and how? An exploratory study on the current implementation of Key account management programs. Industrial Marketing Management, 35(1), 103−112. Wilson, D. T. (1995). An integrated model of buyer–seller relationships. Journal of the Academy of Marketing Science, 23(4), 335−345. Wilson, D. T. (2000). Deep Relationships: The case of the vanishing salesperson. Journal of Personal Selling and Sales Management, 20(1), 53−61. Wilson, K., & Millman, T. F. (2003). The global account manager as political entrepreneur. Industrial Marketing Management, 32(2), 151−158. Wold, H. (1982). Soft modeling: The basic design and some extensions. In K. G. Jöreskog & H. Wold (Eds.), Systems under indirect observation: causality, structure, prediction. Amsterdam: North Holland Publishing. Workman, J. P., Homburg, C., & Jensen, O. (2003). Intraorganizational determinants of key account management effectiveness. Journal of the Academy of Marketing Science, 31(1), 3−21. Wotruba, T. R., & Castleberry, S. B. (1993). Job analysis and hiring practices for national account marketing. Journal of Personal Selling and Sales Management, 13(3), 49−65. Yilmaz, C., & Hunt, S. D. (2001). Salesperson cooperation: The influence of relational, task, organizational, and personal factors. Journal of the Academy of Marketing Science, 29(4), 335−357. Paolo Guenzi is an Associate Professor of Marketing at Bocconi University and SDA Bocconi School of Management, Milan, Italy. His main research interests include sales management and relationship marketing. He has published in international journals such as Journal of Business Research, European Journal of Marketing, International Journal of Service Industry Management, Journal of Marketing Management, Interna- tional Journal of Sport Marketing & Sponsorship. Laurent Georges is an Associate Professor of Marketing at IUT-Tarbes, Toulouse, France. His main research interest is key account management and relationship marketing. He has published in international journals such as Industrial Marketing Management and Journal of Selling and Major Account Management. Catherine Pardo is an Associate Professor of Business-to-Business Marketing at EM-Lyon, France, member of the OCE – EM Lyon (Organisations, Careers and new Elites) research centre. Her research interests include key account management and marketing organization. 311 P. Guenzi et al. / Industrial Marketing Management 38 (2009) 300–311