What comprises a successful key account manager? Differences in the drivers of sales performance between key account managers and regular salespeople

.


Introduction
Key account managers (KAMs) are responsible for the most strategic customers of their company and have been noted for increasing both profits and revenues by as much as 15% (Yip & Bink, 2007). Although usually representing a small share of the sales force, KAMs have a disproportionally high influence on suppliers' sales outcomes as they manage customers responsible for a significant share of the supplier's production. Hence, the emergence of key account management reflected "one of the most important changes in selling" (Ryals, 2012).
Despite the high effort dedicated to identifying performance drivers of individual salespeople (e.g., Churchill Jr, Ford, Hartley, & Walker Jr, 1985;Churchill Jr, Ford, & Walker Jr, 1979;Verbeke, Dietz, & Verwaal, 2011), research investigating performance drivers of KAMs on the individual level remains limited. Prior research on key account management investigated the consequences of key account management programs for companies, the identification of relevant customers, and the structure and formalization of the key account management function and its operational practices (see Appendix A for an overview). However, only 5% of the KAM-related scientific articles between 2001 and 2015 focused on individual-level phenomena (Guesalaga, Gabrielsson, Rogers, Ryals, & Cuevas, 2018;Guesalaga & Johnston, 2010).
In light of the limited knowledge of individual-level drivers of KAM's sales performance, the key goal of this study is to provide a better understanding of how individual characteristics of KAMs influence their sales performance. In addition, we integrate KAM research with research on individual-level drivers of salesperson performance and investigate whether drivers of sales performance differ between KAMs and non-KAMs. The findings of our investigation will not only be relevant to research on key account management but will also offer important insights for selection processes and corporate training programs in key account management.
First, KAMs are responsible for customers that have strategic importance for the supplier company. As a loss of a key account would be detrimental to a supplier's performance, KAMs need to engage in creating and maintaining relationships with these key accounts including the establishment of internal networks in the customer company. Further, due to the size and importance of key accounts, KAMs need to put a high emphasis on planning all customer-oriented activities (Wang & Brennan, 2014).
Second, KAMs are oftentimes involved in complex sales deals that require them to rely on the support resources of their company (e.g., the engineering department). A KAM, therefore, has to be a cross-functional team manager who performs the project management of multiple stakeholders and integrates sales experts and tools to support the selling process (Davies & Ryals, 2013;Workman Jr et al., 2003).
Third, KAMs are responsible for the creation of customer value and the presentation and communication of this value to their customers' buying centers. Although the creation of customer value and the presentation of it to customers is common for both KAMs and regular salespeople, we aim to explore whether there are differences between KAMs and regular salespeople in how this important sales skill influences individual sales performance.
In addition, we incorporate two personality dimensions in our model. Our aim is to investigate their influence on KAM's sales performance and whether these are more or less important for KAMs than they are for regular salespeople. Thereby, we follow prior research on individual-level drivers of sales performance (Churchill Jr et al., 1985;Verbeke et al., 2011;Vinchur, Schippmann, Switzer III, & Roth, 1998) and focus on dominance orientation and need for achievement as personality dimensions (Bateman & Crant, 1993;Buss & Finn, 1987).
To investigate our conceptual model, we build on objective performance data and corresponding survey data of 7023 salespeople (1205 KAMs and 5818 non-KAM salespeople). Results of structural equation modeling identify KAM's project management skills, use of internal experts, their creation of customer value, and their need for achievement as positive drivers of sales performance, whereas KAMs' use of digital tools diminishes their sales performance. Further, we find significant differences between drivers of KAM's sales performance and drivers of regular salespeople's sales performance.
Our findings advance academic research in at least two important ways. First, our findings provide important insights to research on key account management by identifying individual-level drivers of KAM's sales performance. We thereby expand prior research on key account management that considered drivers of organizational sales performance (e.g., Davies & Ryals, 2014;Friend & Johnson, 2014;Tzempelikos & Gounaris, 2015), but remained silent on individual-level drivers of KAM's sales performance. Findings of our study reveal that KAM's project management, use of internal experts, and creation of customer value reflect crucial selling skills that influence KAM's sales performance. In addition, KAM's need for achievement is an important personality dimension that exerts a positive influence on KAM's sales performance whereas KAMs' use of digital tools has the potential to diminish KAM's sales performance.
Second, we contribute to research on individual-level drivers of sales performance by illuminating differences among drivers of sales performance between KAMs and non-KAM salespeople. Our findings indicate that prior knowledge on individual-level drivers of sales performance (e. g., Churchill Jr et al., 1985;Verbeke et al., 2011) is not unconditionally valid for key account management contexts as there are differences in the drivers of KAM's sales performance and the drivers of sales performance of other salespeople.
In addition, our study offers important implications for managers. Due to the identification of drivers of sales performance at the individual KAM-level, our findings provide important insights into how managers can identify the best candidates for effective key account management positions. Furthermore, our findings enable organizations to define training agendas to foster impactful selling skills.

Determinants of individual-level sales performance of KAMs
We integrate prior research on drivers of individual-level sales performance (Churchill Jr et al., 1985;Verbeke et al., 2011; and research on drivers of key account management performance (e.g., Abratt & Kelly, 2002;Sengupta et al., 2000;Tzempelikos & Gounaris, 2015) to conceptualize individual-level drivers of KAM's sales performance. We thereby particularly focus on individual-level characteristics that are relevant for the optimal selection of KAMs and reflect opportunities for KAM training programs: KAM's selling skills and personality dimensions. Fig. 1 depicts our conceptual framework. In the following, we conceptualize KAM's selling skills and personality dimensions and develop hypotheses for their effects on sales performance and for how these effects differ between KAMs and non-KAM salespeople.

Selling skills
Salespeople's selling skills are a learned proficiency at performing the necessary tasks reflected in the execution of selling-related activities in the performance of their job Churchill Jr et al., 1985). Selling skills that are important for sales performance are likely to vary between regular salespeople and KAMs (e.g., Davies & Ryals, 2013;Lacoste, 2018). Specifically, in comparison to regular salespeople, KAMs are responsible for a smaller number of customers that are crucial for the supplier company and that are involved in complex sales deals more often. We account for the differences between regular salespeople and KAMs and select selling skills that categorize among selling activities that are crucial for KAM's performance. Important selling skills for KAMs are the development and management of customer relationships, the management of internal resources, and the creation and communication of customer value.

Relationship building
An ability to build and maintain trust-based, strong, and long-lasting interpersonal relationships between suppliers and buyers has been recognized as an essential skill of KAMs (Guenzi et al., 2007;Iacobucci & Ostrom, 1996;Narayandas & Rangan, 2004). As business-to-business sales roles increasingly evolve into relationship managers that have a strong focus on long-term sales goals (Davies, Ryals, & Holt, 2010;Weitz & Bradford, 1999), KAMs need to be at the forefront of these changes. KAM's boundary spanning between the customer and the supplier company and between single stakeholders of both companies creates synergistic value (Wilson & Millman, 2003) and reflects a key selling skill for the establishment and management of successful key account relationships. Even in industries in which sales is governed by long-term contracts, strong sales relationships can accelerate information sharing between buyer and supplier organizations, help to avoid and resolve conflicts, and positively influence sales performance (Biong & Selnes, 1997). Thus, the more successful KAMs are in developing, managing, and maintaining relationships with customers, the higher should be their sales performance.
Furthermore, when compared to non-KAM salespeople, KAMs are usually responsible for a single or a small number of customer accounts. Therefore, KAM's ability to establish and develop relationships reflects a crucial selling skill for the establishment and maintenance of relationships with highly influential customers (Shi, Zou, White, McNally, & Cavusgil, 2005). KAM's relationship building skills are particularly important because when compared to regular salespeople it is almost impossible for them to compensate for lost customers by increasing their sales volume with new or existing customers. Therefore, KAM's relationship building skills should reflect an important determinant of KAM's sales performance and should be more important for KAMs than for regular salespeople: H 1a : KAMs' relationship building has a positive effect on their sales performance.
H 1b : The effect of relationship building on sales performance is more positive for KAMs than it is for non-KAM salespeople.

Account planning
Moreover, there is a growing importance of account planning in the process of relationship building. Planning customer accounts helps to synchronize supplier competencies with customer needs and facilitates the identification of new selling opportunities (Claycomb, Germain, & Dröge, 2000;Conant & White, 1999;Ryals & Rogers, 2007). Key account planning is a critical sales activity of KAMs that includes the identification of key accounts, the collection and analysis of market information, and the development of a shared strategy with customers and its implementation plans (e.g., Davies & Ryals, 2013). In light of the difficulty of being responsible for customers that purchase a relevant share of the supplier's production and that are oftentimes involved in selling interactions that are highly complex, KAMs need to assure that these customers are served in an efficient manner and that the respective products and services can be delivered in an acceptable time. Thus, to maintain customer relationships with key accounts and to avoid high serving costs due to inefficiency, KAMs need to engage in the proper planning of the accounts' orders, needs, and interactions with the supplier organization. Therefore, we suggest that KAM's account planning positively influences KAM performance.
Although regular salespeople are less likely to spend much time on the detailed planning of single accounts (e.g., Davies & Ryals, 2013), the planning of customer accounts might be more beneficial for them than for KAMs. As regular salespeople are responsible for a high number of customers, appropriate account planning enables them to schedule sales visits more accurately and to monitor their customer relationships more thoroughly. Further, smaller customers are oftentimes not as advanced as key account customers are in their planning of future demand. Therefore, salespeople that contribute to the planning of their accounts tend to achieve more favorable customer consequences with smaller customers because it creates higher value for them. Consequently, we suggest that account planning reflects a skill that might be more beneficial for regular salespeople than for KAMs, as regular salespeople might benefit from a higher efficiency and more favorable customer consequences.
H 2a : KAMs' account planning has a positive effect on their sales performance.
H 2b : The effect of account planning on sales performance is less positive for KAM's than it is for non-KAM salespeople.

Project management
The management of key accounts entails substantial responsibility, implies high selling complexity, and requires KAMs to consistently offer high interaction quality to their customers. To fulfill these requirements, KAMs usually work in a team and use multiple organizational support functions (e.g., resources from marketing, IT, finance) (Barrett, 1986;Woodburn & McDonald, 2013). To orchestrate the interactions between cross-functional support resources, KAMs need to effectively manage the activities of multiple stakeholders and prioritize them on an ongoing basis to ensure that expert support is available and solutions are provided to customers in an acceptable time. KAMs, therefore, have the role of project managers that orchestrate the supplier organizations' resources to offer products, solutions, and services to their key accounts. Consequently, KAM's project management skills should reflect an important driver of their sales performance.
Whereas KAMs, thus, have to overcome the "lone wolf" mentality of regular salespeople (Weitz & Bradford, 1999), regular salespeople may rely more strongly on their capabilities to achieve sales success (e.g., Guenzi et al., 2007;Workman Jr et al., 2003). They are responsible for planning, preparing, and structuring their sales visits and customer interactions by themselves and have to manage the communication with their supplier organization. Furthermore, as they are usually responsible for a high number of customers, they have to efficiently manage their time to simultaneously acquire new customers while managing existing customers. Thus, based on the variety of tasks and the higher number of customers, we suggest that project management reflects a selling skill that is more important for regular salespeople than it is for KAMs.
H 3a : KAMs' project management has a positive influence on their sales performance.
H 3b : The effect of project management on sales performance is less positive for KAMs than it is for non-KAM salespeople.

Use of internal experts
In addition to the required selling skills that enable KAMs to establish and maintain customer relationships, KAMs are also required to have the skills to efficiently manage internal resources and interact with internal stakeholders. Due to the high complexity of KAM's selling processes, KAMs cannot only rely on their own expertise to provide solutions that are acceptable to their customers. Therefore, KAMs need to involve experts and orchestrate them within the customer organization (Steward, Wu, & Hartley, 2010). By strategically using expert resources to extend their own experience, KAMs are likely to create greater value for customers and should achieve more favorable customer outcomes. Thus, in line with prior research (Steward et al., 2010), we expect KAM's ability to build on internal expert resources during the sales process to reflect an important driver of KAM's sales performance. Further, due to the usually lower degree of complexity of regular salespeople's sales processes, we suggest that the use of internal experts reflects a less important driver of sales performance for regular salespeople.
H 4a : KAM's use of internal experts has a positive effect on their sales performance.
H 4b : The effect of the use of internal experts on sales performance is more positive for KAMs than it is for non-KAM salespeople.

Use of digital tools
Salespeople's use of digital tools has been identified as an important driver of customer service, salespeople's adaptability, and salespeople's effectiveness (e.g., Ahearne, Jones, Rapp, & Mathieu, 2008;Johnson & Bharadwaj, 2005). The use of digital tools enables KAMs to get access to important information more easily, provides them with further channels to communicate with their customers and with other organizational functions, and relieves them from spending time on administrative tasks (e.g., Limbu, Jayachandran, & Babin, 2014;Singh et al., 2019;Syam & Sharma, 2018). The use of digital tools thus allows KAMs to spend their time more efficiently and provide them with higher effectiveness in performing their tasks. Therefore, we propose that KAM's use of digital tools positively affects their sales performance.
Further, due to the higher number of customers, regular salespeople are more likely to achieve advantages in terms of efficiency and effectiveness when using digital technologies than KAMs. Furthermore, in key account management, personal relationships are more highly valued so that the use of non-personalized digital technologies may have unfavorable customer consequences. Thus, we expect the use of digital tools to reflect a more important driver of sales performance for non-KAM salespeople than for KAMs.
H 5a : KAM's use of digital sales tools has a positive effect on their sales performance.
H 5b : The effect of the use of digital sales tools on sales performance is less positive for KAMs than it is for non-KAM salespeople.

Creation and communication of customer value
Designing customer solutions, creating customer specific value propositions, and leveraging and effectively communicating these to the customer reflect core tasks for KAMs (Storbacka, 2012). KAM's creation of customer value begins with their accurate assessment of customer needs and then demands KAMs to adapt their offering to meet their customers' needs (e.g., Cannon & Perreault Jr, 1999;Homburg, Müller, & Klarmann, 2011;Homburg, Wieseke, & Bornemann, 2009). In industrial markets, KAMs usually individualize the offering to the customer and create a value proposition. A value proposition is a statement of the benefits the customer receives from the company and the price the company demands for it (Ballantyne, Frow, Varey, & Payne, 2011;Lanning & Michaels, 1988). To create successful value propositions, KAMs do not only have to consider the benefits that the offering of the company provides to their customers but, additionally, need to ensure that it is competitive in the market and that the costs are acceptable for the customer (Ballantyne et al., 2011;Lanning, 1998). Furthermore, the creation of customer value has to be accompanied by KAM's appropriate communication and presentation of the value that the offering provides to the customer (e.g., Ballantyne et al., 2011;Kassemeier, Alavi, Habel, & Schmitz, 2022;Terho, Haas, Eggert, & Ulaga, 2012). If KAMs perform these activities satisfactorily, they should provide customers with solutions that engender their satisfaction and that make them willing to purchase at the supplier organization (e.g., Brady & Cronin Jr, 2001;Brown, Mowen, Donavan, & Licata, 2002). Therefore, we propose that KAM's creation of customer value positively influences their sales performance.
Furthermore, we expect the creation of customer value to reflect an equally important selling activity for regular salespeople's sales performance and for KAM's sales performance. Whereas the creation of customer value might differentiate in its complexity between KAMs and regular salespeople, it is unlikely to differ in its impact on sales performance because both regular customers and key account customers tend to appreciate the value creation efforts of the salesperson.
H 6 : KAMs' creation of customer value has a positive effect on their sales performance.

Personality dimensions
Salespeople's personality has received much attention from prior sales research (e.g., Mahlamäki, Rintamäki, & Rajah, 2019;Verbeke et al., 2011). Especially the relationships between salespeople's personality dimensions and job performance have been the focus of multiple meta-analyses (e.g., Barrick & Mount, 1991;Churchill Jr et al., 1985;Tett, Jackson, & Rothstein, 1991). This high level of interest is not surprising, as salespeople's boundary-spanning role requires them to not only rely on their selling skills but also to have the right personality to achieve sales success. We focus on instrumental personality dimensions that describe the degree to which individuals influence their environment (Bateman & Crant, 1993;Buss & Finn, 1987). Two important facets that describe an individual's instrumental personality are an individual's need for achievement and dominance orientation (Bateman & Crant, 1993;Vinchur et al., 1998).

Need for achievement
KAM's need for achievement is defined as a salesperson's "striving for competence in one's work" (Hough, 1992, p. 144;Vinchur et al., 1998, p. 588). Individuals with a high need for achievement work hard, set themselves high standards, try to do a good job, and concentrate and persist in the completion of their tasks (Hough, 1992). Further, they are more likely to avoid the disproval of their competence and performance (Payne, Youngcourt, & Beaubien, 2007). They set themselves goals that are more challenging (Phillips & Gully, 1997), invest greater effort, and thereby achieve higher levels of performance (Vinchur et al., 1998).
Consequently, salespeople's need for achievement should positively affect sales performance. As it makes individuals perform well in multiple kinds of jobs, we do not expect it to differ in its influence on sales performance between KAMs and regular salespeople.
H 7 : KAMs' need for achievement has a positive effect on their sales performance.

Dominance orientation
Individuals with a high dominance orientation are assertive, forceful, and tend to take charge (Buss & Craik, 1980;Buss & Finn, 1987;Costa Jr & McCrae, 1986;Wiggins, 1979). They attempt to control their environment, aim to influence other people, enjoy the role of a leader, and may assume this role spontaneously (Jackson, 1967;Ray, 1981). A high dominance orientation tends to offer KAMs multiple advantages in managing the various stakeholders that are involved in key account selling. First, a high dominance orientation is usually intertwined with a high capability for leadership (e.g., Bono & Judge, 2004;Judge, Heller, & Mount, 2002). Leadership reflects an important job component of KAMs because they are responsible for the effective management of internal stakeholders to address customer demands. Consequently, KAMs with a high dominance orientation should be more capable to manage both internal stakeholders and external stakeholders which should positively affect the value they create for their customers and lead to higher sales performance. Second, KAM's dominance orientation enables them to achieve more positive outcomes in interactions with customers. For example, dominance orientation has been shown to reflect an important driver of negotiation outcomes (e.g., Carnevale, Pruitt, & Seilheimer, 1981;Weingart, Thompson, Bazerman, & Carroll, 1990). Consequently, KAMs with a high dominance orientation should achieve more positive negotiation outcomes.
In line with our reasoning on the favorable effects of KAM's dominance orientation on their leadership and their performance in interactions with customers, we suggest that it positively affects KAM's sales performance. Furthermore, as regular salespeople are less involved in managing internal stakeholders and teams, we expect that dominance orientation is less influential on their sales performance. Therefore, we suggest: H 8a : KAMs dominance orientation has a positive effect on their sales performance.
H 8b : The effect of dominance orientation on sales performance is more positive for KAMs than it is for non-KAM salespeople.

Data collection and sample
To test our hypotheses, we build on an extensive database of salespeople that was collected by a global strategy consulting company. The database includes matched survey data and objective performance data of individual salespeople from various companies of multiple industries in the Americas. Cross-industry samples strengthen the validity and generalizability of the research results as they reflect the reality of selling across a number of selling set-ups (e.g., Geyskens, Steenkamp, Scheer, & Kumar, 1996).
To collect the data, the consulting company created a questionnaire and shared it with various companies that employ the participating salespeople. The management of the respective companies was involved in announcing the sales survey prior to the launch, explaining its purpose and importance, and encouraging salespeople to respond by a given deadline. The participants in each company received a personalized survey link via email, which was then followed up by multiple reminder messages. The surveys typically stayed online for a period of about two weeks or until it was completed by the majority of the salespeople. The survey included questions on salespeople's skills, personality dimensions, and tenure in the supplier organization. The surveys did not differ between KAMs and non-KAM salespeople and had the same structure for all participating companies. In addition to the survey data, participating companies provided data on salespeople's objective sales performance, which was matched with the salesperson survey data via individual coding. The participating companies additionally provided whether the salesperson works in a KAM position or in a non-KAM position and whether the sales role is consultative or transactional.
The survey was sent to 16,439 salespeople of which 9220 salespeople responded resulting in a response rate of 56.09%. The management of the participating companies provided performance data of 7023 salespeople. 2197 salespeople who answered the survey had to be excluded because of missing values on the outcome variable sales performance. The final sample comprises 7023 salespeople of which 1205 (17.2%) are KAMs and 5818 (82.8%) are regular salespeople (e.g., Field Sellers). Salespeople's average tenure in the company was two years and the majority of salespeople worked in the financial services industry (45.7%). Table 1 provides information on salespeople's characteristics.

Survey development, scale development, and measures
To assess salespeople's selling skills and personality dimensions, a survey was developed in cooperation with the global strategy consulting company. The survey development procedure followed four steps. In a first step, managers of the consulting company collaborated with two academics to identify and define relevant selling skills and personality dimensions. Second, the managers relied on academic literature on selling skills (e.g., Brown, Cron, & Slocum Jr, 1997;Homburg, Workman Jr, & Jensen, 2002) and personality dimensions (e.g., Mahlamäki et al., 2019) and studies from business practice to derive items for the measures (Delmulle, Grehan, & Sagar, 2015;Gimenez, Matrullo, & Maddens, 2019;McCrory, Paulowsky, Valdivieso De Uster, & Viertler, 2016). In a third step, the managers collaborated with two academics to develop scales for KAM's and non-KAM salespeople's creation and communication of customer value, need for achievement, and dominance orientation. In a fourth step, the managers integrated the survey items from previous literature and from the scale development in the questionnaire. Thereby they adapted the survey items so that these are in a similar format and are easily understandable by salespeople. After the initial pool of items was created, items were reviewed by multiple experts and managers of the company to ensure that the measures assess the key dimensions comprehensively, are grammatically sound, and the language used is understandable for salespeople. In the survey, KAM's and non-KAM salespeople's selling skills were assessed by asking participants how well they perform in the respective selling activities in comparison to their colleagues (from 1 -"I'm below the average" to 5 -"I'm one of the best"). To assess their personality dimensions, they were asked about the extent to which they agree or disagree that the respective statements describe their personality (from 1 -"definitely disagree" to 5 -"definitely agree").
In the following, we describe the measures that were derived from extant literature and measures for sales performance and tenure and then describe the scale development process for the measures of KAM's and non-KAM salespeople's creation and communication of customer value, need for achievement, and dominance orientation. Appendix B provides a list of the measurement items.

Measures
For relationship and network building, account planning, project management, use of internal experts, and use of digital tools, measures were derived from extant literature. To assess relationship building skills, we adapted the measure of Shu, Ren, and Zheng (2018). Account planning and project management were measured by adapting the scales of Brown et al. (1997). The use of internal experts was assessed by adapting the measure of Homburg et al. (2002). Further, we adapted three items from Schillewaert, Ahearne, Frambach, and Moenaert (2005) to assess the use of digital tools by asking KAMs and non-KAMs about their usage of and expertise regarding digital sales tools and the CRM system. Salespeople's tenure in the supplier organization was measured on a scale from 1 to 8 -with 1 indicating tenure below six months, 2 indicating tenure between six and twelve months, 3 indicating one to two years in the organization, until, finally, 8 referred to a tenure of above five years.
Further, sales performance was assessed from the organization's HR systems and matched to the survey data. Each company provided information about the sales performance realized by the individual salespeople, typically expressed as the achievement of a revenue target or a revenue quota and defined in percentages (e.g., if from a revenue target of USD 2 ′ 000'000 a salesperson generated USD 1 ′ 000'000 of revenue, the target achieved used in the analysis would be 50%). There was no difference in the operationalization of sales performance between KAMs and regular salespeople to allow the comparison between both sales roles.

Scale development for KAM's and non-KAM's creation and communication of customer value, need for achievement, and dominance orientation
To collect the data with the collaborating global strategy consulting company, we had to develop measures for KAM's and non-KAM's creation and communication of customer value, need for achievement, and dominance orientation. We, therefore, followed the scale development procedure proposed .
In the first step, we generated initial item pools for the three measures by deriving items (1) from academic literature, (2) from discussions with managers from the consulting company, and (3) from discussions with academics. First, we relied on academic literature on salesperson's value creation for customers (e.g., Habel et al., 2020;Kassemeier et al., 2022;Zablah, Franke, Brown, & Bartholomew, 2012) and personality dimensions (e.g., Bateman & Crant, 1993;Buss & Finn, 1987) to derive items for the three measures. We thereby particularly focused on literature for deriving items for KAM's creation and communication of customer value that examined salespeople's value-creating selling approaches (e.g., Behrman & Perreault Jr, 1982;Terho et al., 2012;Mullins, Menguc, & Panagopoulos, 2020). KAM's and non-KAM's creation and communication of customer value reflects salespeople's design of customer solutions and creation of customerspecific value propositions as well as the effective communication of the created value to the customer. The measure, therefore, needs to asses salespeople's capability to design solutions that meet customer needs, their crafting of value propositions, and their communication of value to customers.
For deriving items for need for achievement and dominance orientation, we focused on literature investigating personality dimensions of salespeople (e.g., Barrick & Mount, 1991;Churchill Jr et al., 1985;Tett et al., 1991). Need for achievement defines as an individual's "striving for competence in one's work" (Hough, 1992, p. 144;Vinchur et al., 1998, p. 588). Our measure of need for achievement needs to capture salespeople's strivings to work hard and their aim to achieve high levels of performance in their work (e.g., Hough, 1992). Finally, dominance orientation reflects the tendency to behave in assertive, forceful, and self-assured ways (Buss & Craik, 1980;Wiggins, 1979). The measure for dominance orientation, therefore, needs to assess salespeople's motivation to lead and to take control (Jackson, 1967;Ray, 1981). Second, we had several discussions with the managers of the consulting company in which the managers suggested items that assess the three concepts. Third, we consulted three academics from different institutions to suggest items that assess the three concepts. After consolidating items from the three sources in a long list of items, we dropped items that were redundant (De Vellis, 2003).
In a second step, we collaborated with the management of the consulting company and rated the item's comprehensiveness, logic, and relevance (Bearden, Netemeyer, & Teel, 1989;Tian & McKenzie, 2001). We evaluated (1) if salespeople have the necessary insights to respond to the items, (2) the item's comprehensibility and clarity compared to other items of the scale, and (3) the item's general ability to measure the variables. Furthermore, we used the definitions of the variables to rate whether the items are representative to measure them.
In a third step, we tested the construct validity and convergent validity of our measures. We relied on the crowd-sourcing internet platform Prolific to collect data of 125 salespeople to test the coefficient alpha and to conduct exploratory and confirmatory factor analyses. To recruit salespeople as participants, we screened for panelists who work in customer-facing roles. We administered the survey in English and screened for participants in the United States. In our survey, we asked participants to refer to their sales job when answering questions on their selling skills and personality. We excluded six participants who failed to respond to two attention checks correctly ("Please select "I'm above the average of my peers""; "Please select "Disagree""). The final sample comprises 119 salespeople. The average age of salespeople was 34.85 years (SD = 10.16), the average tenure was 3 years (SD = 2.53), and 52.1% of salespeople were female (SD = 0.50). The Cronbach's alpha values of creation and communication of customer value (0.96), need for achievement (0.86), and dominance orientation (0.92) indicate a sufficient internal consistency. Furthermore, results of exploratory factor analysis examined with principal axis factoring and Promax rotation show that one factor with an eigenvalue >1 was extracted for each concept (e.g., Kaiser, 1960). In addition, we conducted a confirmatory factor analysis to test for composite reliability and average variance extracted. Table 2 presents the results. Composite reliabilities for all three measures were not lower than 0.75 and average variances extracted were not lower than 0.57 thereby exceeding the recommended thresholds (Bagozzi & Yi, 1988). Therefore, these analyses provide support for the construct validity and convergent validity of the scales of KAM's and non-KAM's creation and communication of customer value, need for achievement, and dominance orientation. Table 3 presents descriptive statistics, psychometric properties and correlations of the study's variables. Results of confirmatory factor analysis indicate that the model fits the data well (RMSEA = 0.038; SRMR = 0.037; CFI = 0.935; TLI = 0.928). Furthermore, all Cronbach's alpha values are higher than 0.696, no composite reliability is lower than 0.707, and no average variance extracted is lower than 0.547.

Psychometric properties of measurement variables
Moreover, all square roots of the average variances extracted exceed the correlations between the respective variables thereby indicating that all recommended thresholds for reliability and validity are met or exceeded (Bagozzi & Yi, 1988;Cronbach, 1951;Fornell & Larcker, 1981).
To ensure that our model and its measures are interpreted in a conceptually similar manner by KAMs and by non-KAM salespeople, we followed the recommendations of Brown (2014) and Van de Schoot, Lugtig, and Hox (2012) to test for measurement invariance. Results of confirmatory factor analyses for both groups indicate that the model fits the data well and that configural variance can be established for the group of KAMs (RMSEA = 0.057; SRMR = 0.049; CFI = 0.90, TLI = 0.89) and the group of non-KAM salespeople (RMSEA = 0.041; SRMR = 0.057; CFI = 0.92, TLI = 0.91). Furthermore, results of model comparison tests demonstrate that metric invariance (ΔRMSEA <0.001; ΔSRMR = 0.008; ΔCFI = 0.003) and scalar invariance (ΔRMSEA <0.001; ΔSRMR = 0.001; ΔCFI = 0.005) were established across the groups of KAMs and non-KAMs as the models fitted the data well and the changes in model fit between the different models fall below the recommended thresholds (Steenkamp & Maydeu-Olivares, 2020). Therefore, both KAMs and non-KAM salespeople attributed the same meaning to our measures.

Analytical approach
To estimate our model for both KAMs and non-KAM salespeople we relied on a multigroup structural equation modeling approach (Kline, 2015), in which we estimated the effects of the selling skills and of the personality dimensions on sales performance simultaneously for KAMs and non-KAMs. We relied on a structural equation modeling approach because it enables us to estimate the effects of the independent variables on sales performance in one model while accounting for measurement error when estimating our model. Further, to account for the nested data structure (salespeople nested in different industries) we estimated our model by using a maximum-likelihood estimator that is robust against non-normality and non-independence of observations (Muthén & Muthén, 2018). To investigate differences between the effects of selling skills and personality dimensions on sales performance between KAMs and non-KAMs, we calculated the absolute differences between these effects across the two groups and tested their significance.

Table 2
Results of the scale validation analysis for creation and communication of customer value, need for achievement, and dominance orientation.  (Hough, 1992, p. 144;Vinchur et al., 1998, p. 588 (Jackson, 1967;Ray, 1981 Notes: N = 119, Standardized factor loadings are results of the confirmatory factor analysis, CR = composite reliability, AVE = Average variance extracted. Notes: ** p < .01, * p < .05; Cronbach (1951) internal consistency reliability reported on the diagonal. Model fit: RMSEA = 0.038; SRMR = 0.037; CFI = 0.935; TLI = 0.928. a Tenure in the company. b Consultative (= 1) or transactional (= 0). Table 4 presents the results. We find positive and significant effects of KAM's project management (γ KAM's project management → sales performance = 0.572; p < .05), KAM's use of internal experts (γ KAM's use of internal experts → sales performance = 4.876; p < .01), and KAM's creation and communication of customer value (γ KAM's creation and communication of customer value → sales performance = 4.755; p < .01) on sales performance. Thereby our results provide support for H 3a , H 4a , and H 6 and show that KAM's project management, use of internal experts, and creation and communication of customer value reflect skills that significantly contribute to KAM's sales performance.

Effects of KAM's selling skills and KAM's personality dimensions on sales performance
Further, contrary to our expectations we do not find a significant and positive but a significant and negative effect of KAM's use of digital tools on their sales performance (γ KAM's use of digital tools → sales performance = − 4.574; p < .01). Consequently, we find no support for H 5a . In addition, we do not find support for H 1a and H 2a as KAM's relationship building skills (γ KAM's relationship building → sales performance = 0.093; n. s.) and account planning skills (γ KAM's account planning → sales performance = − 1.140; n. s.) do not have a significant impact on KAM's sales performance. KAM's relationship building skills might relate to favorable customer outcomes like satisfaction or loyalty but might not always directly impact their sales performance. This is in line with prior research on customeroriented selling activities (e.g., Franke & Park, 2006). Further, KAM's account planning might contribute to profit measures but might not directly relate to sales revenue, which was incorporated in this analysis.
Furthermore, we tested H 7 and H 8a by investigating how KAM's need for achievement and KAM's dominance orientation influence their sales performance. Whereas we find that KAM's need for achievement reflects a significant and positive driver of KAM's sales performance (γ KAM's need for achievement → sales performance = 4.231; p < .01), we do not find a significant effect of KAM's dominance orientation on their sales performance (γ KAM's dominance orientation → sales performance = 0.858; n. s.).

Differences in the drivers of KAM's and non-KAM's sales performance
In addition to identifying drivers of KAM's sales performance, we investigated whether and how drivers of KAM's sales performance differ from drivers of regular salespeople's sales performance. Our results show that project management skills, the use of internal experts, and the use of digital tools differ significantly in their effects on sales performance between KAMs and non-KAMs. Specifically, as project management skills are more important for regular salespeople's sales performance than for KAM's sales performance (Δγ Project management → sales performance = 1.076; p < .05), our results provide support for H 3b . Further, we find that KAM's use of internal experts has a more positive effect on sales performance than non-KAM's use of internal experts (Δγ Use of internal experts → sales performance = 3.355; p < .01). Thereby our results provide support for H 4b and show that the selling skill use of internal experts is more important for KAM's sales performance than for non-KAMs sales performance. In addition, we find that the use of digital tools has a negative influence on KAM's sales performance, whereas it has no significant impact on non-KAM's sales performance. The effect of the use of digital tools on sales performance varies significantly between KAMs and non-KAMs (Δγ Use of digital tools → sales performance = 4.484; p < .01) and indicates that the use of digital tools exerts a less detrimental influence on non-KAM's sales performance than on KAM's sales performance.
Furthermore, in line with our reasoning we do not find significant differences between KAMs and non-KAMs for the effects of creation and communication of customer value (Δγ Creation and communication of customer value → sales performance = 2.323; n. s.) and need for achievement (Δγ Need for achievement → sales performance = 0.061; n. s.) on sales performance. Thus, the personality dimension need for achievement reflects a positive driver of sales performance for KAMs and non-KAMs. Furthermore, the positive effect of creation and communication of customer value on sales performance is significant for both non-KAMs (γ Non-KAM's creation and communication of customer value → sales performance = 2.432; p < .05) as well as KAMs and does not vary significantly between the two types of salespeople.
In addition, although we find a significant difference between KAMs and non-KAMs for the effect of account planning on sales performance, which offers some support for H 2b (Δγ Account planning → sales performance = 3.583; p < .05), we do not find that account planning reflects a selling skill that significantly influences KAM's and non-KAM's sales performance. Further, we do not find significant differences between the effect of KAM's and non-KAM's relationship building on sales performance (Δγ Relationship building → sales performance = 0.072; n. s.). Consequently, we do not find support for H 1b . Finally, the effects of KAM's and non-KAM's dominance orientation on sales performance do not vary significantly (Δγ Dominance orientation → sales performance = 1.912; n. s.). Moreover, similarly to KAMs, dominance orientation does not significantly affect non-KAM's sales performance (γ Non-KAM's Dominance orientation → sales performance = − 1.054; n. s.). Thereby our results do not provide support for H 8b. Table 5 presents the results of the hypothesis tests of this study. Although individual-level drivers of sales performance are crucial for the appropriate selection and training of KAMs, knowledge regarding individual-level drivers of KAM's sales performance is limited. This study identifies KAM's project management skills, their use of internal experts, and their creation and communication of customer value as positive drivers of KAM's sales performance. In addition, KAM's use of digital tools exerts a negative impact on their sales performance. Salespeople's need for achievement reflects a personality dimension that complements these selling skills and exerts a positive influence on KAM's sales performance.

Discussion
Furthermore, this study provides initial insights on whether and how drivers of sales performance differentiate between regular salespeople and KAMs. Specifically, our findings show that the selling skill project management has consequences that are more beneficial for regular salespeople's sales performance than for KAM's sales performance. Further, salespeople's skill to use expert resources to support the sales process is a stronger driver of KAM's sales performance than it is for regular salespeople's sales performance. Finally, whereas the use of Notes: ** p < .01, * p < .05 (one-tailed); we report unstandardized coefficients; H = Hypothesis.
digital tools has undesired consequences for KAM's sales performance, it does not reflect a significant driver of the sales performance of regular salespeople.

Theoretical implications
The findings of this study contribute to academic research on key account management and sales research in at least two important ways. First, our findings contribute to research on key account management by identifying individual-level drivers of KAM's sales performance. Prior research on drivers of KAM's performance examined how the identification of relevant account information, the set up and formalization of the key account management function, and operational practices influence the sales performance of KAMs and the KAM function (e.g., Davies & Ryals, 2014;Friend & Johnson, 2014;Tzempelikos & Gounaris, 2015). This study advances current knowledge of drivers of KAM's sales performance by investigating whether and how KAM's selling skills and personality influence their individual-level sales performance. In particular, this study reveals that KAM's use of internal experts, their project management skills, their creation and communication of customer value, and their need for achievement reflect important drivers of individual sales performance whereas KAMs' use of digital tools has a detrimental influence on KAM's sales performance. By identifying selling skills and salespeople's need for achievement as important drivers of KAM's sales performance the current study adds an individual-level perspective to prior research on drivers of KAM's sales performance that explains inter-individual performance differences between KAMs.
In addition, future research might provide further explanations for the counterintuitive finding that the use of digital sales tools negatively influences KAM's sales performance. Two potential explanations might reflect the starting point for further research to investigate whether the use of digital sales tools unconditionally mitigates KAM's sales performance. First, as implementations of digital sales tools oftentimes fail (Bush, Moore, & Rocco, 2005;Gohmann, Guan, Barker, & Faulds, 2005;Robinson Jr, Marshall, & Stamps, 2005), it hinders the sales process and makes salespeople lose valuable selling time. Second, most of the standardized digital tools are designed for simpler sales processes, but not for the practice of selling to key accounts. As the latter often entails more complicated interactions with a higher number of stakeholders that place high importance on trust and business interest alignment, standardized digital tools might have a low fit to KAM's sales processes (Speier & Venkatesh, 2002) and might have undesired consequences on KAM's sales performance.
Second, we contribute to research on individual-level drivers of sales performance by illuminating differences in the drivers of sales performance between KAMs and non-KAM salespeople. Whereas extant sales literature has identified individual-level drivers of sales performance (e. g., Churchill Jr et al., 1985;Verbeke et al., 2011), it remains unclear whether these drivers are unconditionally valid for KAMs. Recent research on key account management identified differences between the relevant competencies of KAMs and non-KAMs conceptually (Lacoste, 2018) and compared how KAMs and non-KAMs differ in their attitudes and behavior (Davies & Ryals, 2013). We link this emergent stream of literature with research on individual-level drivers of sales performance and investigate how drivers of sales performance differ between KAMs and non-KAM salespeople. Our findings indicate that prior knowledge of individual-level drivers of sales performance is not unconditionally valid for the key account management context. By identifying differences in the drivers of sales performance between KAMs and non-KAMs, we provide important insights into the key account management literature and reveal that the key account management context requires a different set of selling skills than traditional sales jobs to successfully engender sales performance. This could reflect a starting point for future research to shed light on how the key account management context differs from the traditional selling contexts and how these differences affect performance outcomes. Future research could for example investigate how leadership mechanisms, sales controls, or negotiation styles differ between KAMs and non-KAMs in their influence on performance outcomes.

Managerial implications
Our findings provide important insights for practitioners by Table 5 Results of hypothesis tests.
identifying individual-level selling skills and personality dimensions that drive KAMs' sales performance. Furthermore, we equip managers with the knowledge of how individual-level drivers of KAM's sales performance differentiate from individual-level drivers of regular salespeople's sales performance. The findings of this study help organizations to select KAMs more appropriately and to create impactful capability-building programs for KAMs.
First, our findings help organizations to make more suitable KAM appointments. The findings of this study equip managers with metrics to select ideal KAMs. If KAMs are sourced externally, companies can test applicants' sophistication on skills that contribute to KAM's performance. For instance, by asking applicants to perform live sales pitch presentations, or by using case studies to sense if, when, and how the potential new KAMs engage in the management of projects and leverage expert resources in the sales process.
Second, the findings of this study help managers design KAM capability-building programs. Each year, suppliers in the United States spend USD 1 ′ 459 per salesperson solely on sales training and capability building (Cespedes & Lee, 2017). This number is 20% higher for salespeople than for other employees (Cespedes & Lee, 2017) and, thus, reflects a priority for many companies. Our research provides guidance for organizations when designing capability-building sales training for their KAMs. In these programs, elements such as the knowledge about the supplier's and competitive offerings, tailoring of the offering to customer needs, clever use of expert resources at various stages of the sales process, role-plays related to value proposition delivery, or crafty ways to approach the pricing of value propositions should take priority.

Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of Competing Interest
None.

Appendix A
Literature overview on KAM-related publications.

Appendix B
Measures. Notes: Selling skills were assessed on a scale ranging from 1 -"I'm below the average" to 5 -"I'm one of the best". Personality dimensions were assessed on a scale ranging from 1 -"Definitely disagree" to 5 -"definitely agree". Standardized factor loadings are standardized factor loadings of confirmatory factor analysis.