Capability management control and salesperson turnover: A double-edged sword in a product complexity scenario

Industrial Marketing Management 96 (2021) 100–112 Available online 15 May 2021 0019-8501/© 2021 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Research paper Capability management control and salesperson turnover: A double-edged sword in a product complexity scenario Bel´ en Bande a , * , Takuma Kimura b , Pilar Fern´ andez-Ferrín c , Fernando Jaramillo d a University of Santiago de Compostela, Facultad de Administraci´ on y Direcci´ on de Empresas, Avda. Alfonso X el Sabio, s/n, Lugo 27002, Spain b Hosei University, 2-17-1, Fujimi, Chiyoda-ku, Tokyo, Japan c University of País Vasco (UPV/EHU), Facultad de Economía y Empresa, Comandante Izarduy, 23, Vitoria-Gasteiz 01006, Spain d University of Texas at Arlington, College of Business, 76019 Arlington, TX, USA ARTICLE INFO Keywords: Management control Product complexity Work meaning Work overload Turnover Salesperson ABSTRACT Drawing on the Job Demand-Resources (JD-R) theory, the present study investigates the underlying mechanisms through which capability control, a type of behavior-based control, influences salesperson turnover. Using a sample of 145 industrial salesperson–supervisor dyads from different industries, this study's findings reveal that capability control contributes to decreasing salesperson turnover intentions, both directly and indirectly. Specifically, management capability control reduces work overload and increases work meaning, thus lowering salesperson turnover intention. The findings also confirm that these effects are contingent on the complexity of a product. When product complexity is low, capability control increases work overload and decreases work meaning, which has a positive effect on salesperson turnover intention. This study discusses these findings' theoretical and managerial implications. 1. Introduction Turnover represents an ongoing concern for employers. Profits fall dramatically after an employee leaves (Ton & Huckman, 2008), which reduces the remaining employees' levels of job satisfaction and performance (Argote, Insko, Yovetich, & Romero, 1995). Turnover rates for salespeople are such that they are the largest risk group (Katsikea, Theodosiou, & Morgan, 2015). For example, the annual turnover among US salespeople is twice the rate in the labor force (HBR, 2017). According to the The Bridge Group (2018), the median annual turnover sits at 39% among business-to-business salespeople. According to this report, the annual tenure of salespeople has dropped from 2.5 years in 2010 to 1.5 years in 2018. As to the impact of turnover, previous studies show that turnover among salespeople is particularly costly. In addition to the direct costs, the indirect costs associated with salesperson turnover may be even greater (Boles, Dudley, Onyemah, Rouzi` es, & Weeks, 2012). Salesperson turnover can significantly change customer relationships, which results in revenue losses, and investment losses in training and incentives (Sunder, Kumar, Goreczny, & Maurer, 2017). Additionally, turnover also leads to customer-directed and organizational-directed counterproductive behavior (Seriki, Nath, Ingene, & Evans, 2020). Concordant with these findings, sales managers identify sales representative retention as a particularly important issue – increasing its relevance as a managerial challenge by 83% in the last two years (The Bridge Group, 2018). Understanding salesperson turnover is therefore a particularly relevant topic for both researchers and practitioners (Rubenstein, Eberly, Lee, & Mitchell, 2018). Using precepts from Job Demands-Resources Theory (JD-R), this manuscript argues that capability control provides managers with an important tool to increase salesperson retention. Sales management control systems, conceptualized by Anderson and Oliver (1987) as “an organization's set of procedures for monitoring, directing, evaluating, and compensating its employees“(p. 76), are frequently used by managers to align employees with organizational goals (Anderson & Oliver, 1987; Jaworski, 1988; Oliver & Anderson, 1994). Given the flexibility in a salesperson's role, control is an essential part of sales management activity and “one of the most important determinants of the effective management of salespeople” (Flaherty, Pappas, & Allison, 2014, p. 305). Researchers have accumulated a deep understanding about sales management control (Anderson & Oliver, 1987) and how it influences employees' attitudes and behaviors (see * Corresponding author. E-mail addresses: [email protected] (B. Bande), [email protected] (T. Kimura), [email protected] (P. Fern´ andez-Ferrín), [email protected] (F. Jaramillo). Contents lists available at ScienceDirect Industrial Marketing Management journal homepage: www.elsevier.com/locate/indmarman https://doi.org/10.1016/j.indmarman.2021.05.002 Received 9 September 2020; Received in revised form 21 April 2021; Accepted 10 May 2021 Industrial Marketing Management 96 (2021) 100–112 101 recent meta-analyses, Katsikeas, Auh, Spyropoulou, & Menguc, 2018; Malek, Sarin, & Jaworski, 2018). This study focuses on capability control, a sales management control, which is a behavior-based control that prioritizes job-related abilities, designed to develop and reward salespeople's selling skills (Challagalla & Shervani, 1996). By prioritizing the development of sales skills, sales managers can motivate salespeople to stay in an organization that allows them to be productive (Pettijohn, Pettijohn, & Taylor, 2007). The JD-R theory views capability control as a critical resource “because it enables salespeople to achieve work goals and to cope with job demands with appropriate selling skills and knowledge” (Miao & Evans, 2013, p. 74). Ohiomah, Benyoucef, and Andreev (2020) provide meta-analytic evidence of this assertion with findings that capability control is associated with of sales success in business-to-business. However, as discussed below, the effects of capability control on turnover intention are complex and not well understood. This is in spite precepts from JD-R theory that salesperson resources affect employee engagement, attitudes, and strain (Miao & Evans, 2013), which are known drivers of turnover intentions. Considering the effect of capability control on turnover is critical because of the long-lasting debate regarding whether investments in human resource development ultimately decrease turnover (Baldauf & Cravens, 2003; Benson, Finegold, & Mohrman, 2004; Gerhart, 1990). Only a few studies have examined the influence of capability control on salesperson turnover so far, and they yielded mixed findings. For example, Baranchenko, Xie, Lin, Lau, and Ma (2020) found that a supervisor's investment in capability control increases turnover intentions. Conversely, Katsikea et al. (2015) confirmed that capability control decreases turnover intention among export sales managers. It is therefore theoretically and practically important to examine the boundary conditions that affect the relationship between capability control and salesperson turnover. As a boundary condition, this study investigates the effect of product complexity because of its prevalence and its implication for sales performance. As shown by Cuevas (2018), modern sales forces operate in increasingly complex contexts. The 2005 CEB survey of over 1000 business-to-business salespeople reported that 70.2% of sales professionals perceive their jobs as moderately to overly complex. Product complexity increases coordination efforts and negatively affects performance (Mocker, Weill, & Woerner, 2014; Tatikonda & Rosenthal, 2000). This study investigates whether product complexity increases the importance of capability control for effective sales management. Another question underexplored in previous studies is why capability control affects employee turnover. To fill this research gap, this study aims to identify constructs that mediate the relationship between capability control and turnover. Building on the motivational and energy-depletion processes of the Job Demand-Resources (JD-R) theory, this study examines both the direct and indirect effects of capability control on salesperson turnover intention through work overload and work meaning. According to JD-R, two psychological processes affect job strain and motivation (Bakker & Demerouti, 2007). The first process involves job demands like work overload which deplete employees from their energy (Bakker & Demerouti, 2007). The second process is motivational (Bakker & Demerouti, 2007). Salesperson motivation derives from doing something that counts and makes the job meaningful (Jaramillo, Mulki, & Boles, 2013). When the job is meaningful, salespeople are less likely to quit their jobs and perform at higher rates (Jaramillo et al., 2013). Although previous studies have revealed the significant effects of both work overload and work meaning on employee attitudes and turnover intentions (see Jaramillo et al., 2013; Mulki, Lassk, & Jaramillo, 2008), these constructs have been underexplored in sales management studies. Previous studies have also not examined the influence of these constructs to reveal the critical research question; does capability control decrease or increase workers' turnover intention? To address this issue, this study develops a mediated moderation model. The proposed model contains work overload and work meaning as mediators, and product complexity as a boundary condition of the direct and indirect control–turnover relationship. Complexity is referred to by practitioners as the bane of the sales profession (Seriki et al., 2020). By considering the moderating effect of product complexity, this study can reveal the reason why previous studies show mixed results for the effect of behavioral control on job stress (Challagalla & Shervani, 1996; Katsikea et al., 2015; Piercy, Cravens, & Lane, 2001). An examination of a moderator also helps to clarify how much an organization's investment in time and effort to improve salespeople skills and abilities contributes to reducing their intention to leave an organization. Including mediating variables in the analysis of the relationship between turnover and its antecedents can contribute to organizational research since many previous studies only consider direct effects on turnover (Allen, Hancock, Vardaman, & Mckee, 2014). We organize the paper as follows. First, we present the theoretical logic of capability control and JD-R theory. Then, we develop a conceptual model and hypotheses about the effect of capability control on turnover and the mediating roles of work meaning and work overload. We also hypothesize that product complexity moderates the effects of capability control on work meaning, work overload, and turnover. Next, the manuscript uses dyadic data derived from salespeople and their managers to analyze the data and test model hypotheses. Finally, we discuss model results, theoretical conclusions, and managerial implications. 2. Literature review 2.1. Capability control and salesperson turnover intention Anderson and Oliver (1987) classified formal sales controls into behavioral control and outcome control, representing two different managerial philosophies with similar goals (Oliver & Anderson, 1994). Under outcome control, a sales manager focuses on outputs, which provides greater autonomy to salespeople who are under great pressure because they are rewarded based on achieving planned objectives (Kohli, Shervani, & Challagalla, 1998). Conversely, behavior-based control, or process control, according to Jaworski's (1988) conceptualization, involves a longer time perspective by focusing on the selling process instead of on results. Behavior-based control involves support, coaching, evaluating, and monitoring salesperson behaviors rather than outcomes (Miao, Evans, & Shaoming, 2007). Building on Anderson and Oliver's (1987) distinction, Challagalla and Shervani (1996) identified two types of behavioral control: activity control and capability control. This study focuses on capability control that emphasizes the improvement of salespeople's skills and abilities in sales presentations, closing or negotiations. Capability control attempts to achieve excellent performance by ensuring high levels of employee skills and abilities (Challagalla & Shervani, 1996). It provides salespeople with guidelines and evaluations about selling skills, emphasizing their development and applications. A capability-oriented supervisor is more of a coach (Albizu, Rekalde, Landeta, & Ferrín, 2019; Kohli et al., 1998) who contributes to enhancing salespeople's well-being and their intrinsic motivation (Miao et al., 2007). Drawing on Social Exchange Theory (Blau, 1964; Rousseau, 1995), it is expected that a supervisor's investment in a subordinate's growth and development of skills and capabilities will be returned in the form of commitment to the organization. As Oliver and Anderson (1994) state “employees feel committed and grateful to supervisors who use process control because it provides them a nurturing climate and reduced performance risk” (p. 54). According to JD-R theory, capability control is a resource that increases employee engagement, improves employee attitudes, and reduces strain perceptions (Miao & Evans). In view of this, capability control should also reduce turnover intentions. Capability control “sends a positive signal of the supervisor's B. Bande et al. Industrial Marketing Management 96 (2021) 100–112 102 concern, care, and support for the salesperson” (Atuahene-Gima & Li, 2002, p. 66), which contributes to creating a closer relationship between the manager and the salesperson (Anderson & Oliver, 1987). Under a capability control, employees perceive themselves as being valuable, skillful, and able to accomplish their career objectives within the current organization (Benson et al., 2004; Nauta, Van Vianen, Van der Heijden, Van Dam, & Willemsen, 2009). As Benson et al. (2004) show, employees who participate in development programs will be less likely to leave an organization. Furthermore, several empirical studies confirm that supervisory control impacts salesperson turnover (e.g. Brashear, Manolis, & Brooks, 2005; Katsikea et al., 2015; Piercy et al., 2001). Based on the arguments and findings above, the first proposed hypothesis is as follows: H1. Capability control is negatively related to salesperson turnover intention. 2.2. Work meaning In addition to the social exchange process, other influential mechanisms of capability control exist. They are i) providing meaningfulness to work and ii) alleviating the adverse effect of job demand. These mechanisms can explain why capability control lowers turnover intention in some cases but increases it in others. Work meaning is defined as “the degree to which the employee experiences the job as one which is generally meaningful, valuable, and worthwhile” (Hackman & Oldham, 1975, p.162). Considered as “the engine of empowerment” (Spreitzer, Kizilos, & Nason, 1997, p. 681) – as well as a source of motivation at work (Barrick, Mount, & Li, 2013; Rosso, Dekas, & Wrzesniewski, 2010) – the meaning of work is found to be related to important work outcomes (Liden, Wayne, & Sparrowe, 2000), such as job satisfaction (Steger, Dik, & Duffy, 2012), well-being (Arnold, Turner, Barling, Kelloway, & McKee, 2007), organizational commitment (Steger et al., 2012), absenteeism (Steger et al., 2012) and individual performance (Wrzesniewski, 2003). Jaramillo et al. (2013) show that experienced meaningfulness reduces strain perceptions and reduces turnover intention. Although work meaning is an unexplored topic in the extant sales literature (Fock, Yim, & Rodriguez, 2010; Jaramillo et al., 2013), research reveals that finding meaning in work is positively related to effective sales performance (Ahearne, Mathieu, & Rapp, 2005) and negatively related to turnover intention (Jaramillo et al., 2013). In terms of JD-R model terminology (Demerouti, Bakker, Nachreiner, & Schaufeli, 2001), capability control works as a job resource for employees since it provides them with developmental opportunities and perceived competence (Challagalla & Shervani, 1996), and therefore contributes to high performance and positive work attitude (Miao et al., 2007). Job resources are important not just because they can reduce job demands and the associated costs, but because they help to achieve or protect other valued resources (Bakker & Demerouti, 2007). In addition, they can unleash a motivational process and lead to high work engagement and positive organizational outcomes. This study considers work meaning as a manifestation of job engagement because it reflects an employee's level of involvement in their job. Previous studies show a strong link between work meaning and work engagement (e.g. Britt, Adler, & Bartone, 2001; Hirschi, 2012). This study posits that capability control as a job resource plays an intrinsic motivational role because it promotes employee learning and it enhances employee development and growth, which contribute to the fulfilment of personal potential or self-realization. Cross-cultural research (e.g. Work Importance Study) has identified self-realization as an essential source as well as a component of the meaning of work (Super & ˇ Sverko, 1995). Moreover, in a capability control context, supervisors dedicate time and effort to assess their subordinate's capabilities and to guide their growth and improvement. Employees consequently perceive themselves as being valuable (Nauta et al., 2009) and when individuals feel useful and valuable, their sense of work meaning will increase (Rothmann & Olivier, 2007). Thus, capability control is expected to be positively related to work meaning. On the other hand, the lack of work meaning causes detachment from one's work (Thomas & Velthouse, 1990) and it is positively related to an employee's intentions to leave an organization (Leunissen, Sedikides, Wildschut, & Cohen, 2018; Steger et al., 2012). Therefore, the second hypothesis proposes: H2. Capability control is negatively related to salesperson turnover intention through salesperson work meaning. 2.3. Work overload and job demands Another mediator that is considered is work overload. Work overload is one of the major stressors in the workplace (Frone, 2008). Perceived work overload is a psychological stressor which measures the extent to which an individual perceives that they have too many tasks to perform in a given time (Greenglass, Burke, & Moore, 2003). According to JD-R theory, work overload is a job demand that depletes employees from their energy and can lead to employee strain (Bakker & Demerouti, 2007). Overworked salespeople feel that they cannot complete job tasks adequately and in time (Jaramillo, Mulki, & Boles, 2011). Work overload increases emotional exhaustion, reduces job satisfaction and commitment, and ultimately leads to higher turnover (Jaramillo et al., 2011). For salespeople, work overload is unavoidable due to the highpressure and competitive environment in which they operate (Mulki et al., 2008). Narayanan, Menon, and Spector (1999) showed that among sales positions, the unfavorable effects of work overload are stronger than other stressors such as role conflict or role ambiguity. Meta-analytic reviews confirm that work overload affects work outcomes, such as organizational commitment, job satisfaction and job performance (Gilboa, Shirom, Fried, & Cooper, 2008; ¨ Ortqvist & Wincent, 2006). Building on the conservation of resources theory (Hobfoll, 1989), this study posits that capability control increases employees' resources so they are more capable of meeting job demands; and it also prevents resource depletion. As Ahearne, Rapp, Hughes, and Jindal (2010) argued, behavior-based control can increase employees' sense of security and decrease their fear and reticence to fail. In a similar vein, Baldauf, Cravens, and Piercy (2005) found that behavior-based sales management control is positively related to professional competence. The evidence confirms that capability control augments salespeople's perception of competence (Challagalla & Shervani, 1996; Evans, Landry, Li, & Zou, 2007). When salespeople are confident in their abilities and skills, they can prioritize and allocate effort, which minimizes the perception of overload (Jaramillo & Mulki, 2008; Mulki et al., 2008). As Mulki et al. (2008) note, self-confident salespeople can “spend available resources to get the sales job done without becoming overwhelmed or frustrated” (p. 290). Regarding the relationship between work overload and turnover, one of the emotion-focused coping strategies that individuals can use to cope with a stressful situation is avoidance (Lazarus & Folkman, 1984). Thus, salespeople might consider leaving their current position to cope with perceived work overload (Jones, Chonko, Rangarajan, & Roberts, 2007). Meta-analytic studies (e.g. Griffeth, Hom, & Gaertner, 2000) confirm a positive relationship between work overload and turnover intention. Building on the above discussion, the following hypothesis states: H3. Capability control is negatively related to salesperson turnover intention through the reduction of work overload. 2.4. The moderating effect of product complexity In the following, it is proposed that product complexity moderates the direct and indirect effect of capability control on turnover intention. B. Bande et al. Industrial Marketing Management 96 (2021) 100–112 103 There is a lack of clarity in the literature regarding the definition of product complexity (Trattner, Hvam, Forza, & Herbert-Hansen, 2019). This study follows Johnson and Sohi's (2014) conceptualization of product complexity as “the extent to which the products in the lines are technically complex and difficult to explain to customers” (p. 76). According to JD-R terminology, job demands refer to “those physical, psychological, social, or organizational aspects of the job that require sustained physical and/or psychological effort or skills” (Bakker & Demerouti, 2007, p. 312). Product complexity is considered a job demand, particularly in a sales context (Rivas, Quyen, & Rivas, 2017). The reason for this is that the high degree of technical complexity causes salespeople to spend more time in, getting to know the product properly, and guiding the consumer throughout the purchase process (Brown, Zablah, Bellenger, & Donthu, 2012; Friend, Johnson, & Ranjan, 2020). Product complexity reduces salespersons' capacity to apply cognitive schemas and is thus an important moderator in stressor-strain relationships (Johnson & Sohi, 2014). In Johnson and Sohi's (2014) study, product complexity increased the effect of number of product lines handled by a salesperson on role conflict. As to the first point, salespeople are required to have significant technical knowledge of the complex products they sell. This is to ensure that they can explain their products' characteristics and applications to buyers. When selling complex products, the selling proposition is usually more complex, which requires the seller to process a larger amount of information and to spend more time and effort to become an expert in the selling proposition (Johnson & Sohi, 2014; Jones, Dixon, Chonko, & Cannon, 2005). Regarding the second point, product complexity is also associated with the frequency of personal contacts and with an increase in the level of customer service (Campbell, 1985). Moreover, as the complexity of a product increases, consumers tend to perceive a high level of risk (Holak & Lehmann, 1990) and danger of being treated unethically by the salesperson (Sitzia & Zizzo, 2011). Product complexity can thus make interactions with clients emotionally demanding and increasingly stressful (Bakker & Demerouti, 2007). 2.4.1. Moderating effect on the indirect effect via work meaning According to the JD-R model, job demands and job resources interact, leading to specific outcomes. The evidence confirms that the presence of job resources leads salespeople to consider that job demands are more controllable, and eventually higher engagement (Miao & Evans, 2013). In a high product complexity scenario, it is expected that developing a salesperson's abilities and skills will have a stronger impact on enhancing their perceived level of competence, which will result in an increased work meaning. However, relying on self-determination theory (Deci & Ryan, 1985), this study proposes that capability control may work unfavorably on work meaning when product complexity is low. Capability control implies empowering managers to supervise and direct how salespeople perform sales tasks (Oliver & Anderson, 1994). With a high level of capability control, supervisory feedback becomes more exhaustive, and sales managers impose their ideas on their subordinates (Anderson & Oliver, 1987). When product complexity is low, salespeople will likely perceive managerial guidance as unnecessary and intrusive. As Peesker, Ryals, Rich, and Boehnke (2019) asserts, there is a “thin line between coaching and micromanagement” (p. 324). This situation also entails a risk for a salesperson's perceived autonomy that is more pronounced when product complexity is low. According to self-determination theory (Deci & Ryan, 2012), threats to satisfying autonomy needs result in weaker internal motivation. When the complexity of the product is low, risk perception is lower while selling tasks become monotonous and less autonomous. In this context, extensive supervisory feedback in the form of capability control may have a detrimental effect on internal motivation and the sense of work meaning. A previous study supports these arguments, showing that a low level of job autonomy leads to a loss of work meaning (Fried & Ferris, 1987). In addition to the loss of autonomy, capability control under the situation of low product complexity can make a job seem less challenging for salespeople. Supervisory capability orientation has a positive effect on salespeople's learning orientation (Kohli et al., 1998). Since “salespeople with a learning orientation enjoy pursuing challenging goals and tasks” (Kohli et al., 1998, p. 271), the lack of challenging tasks derived from a low complex product might negatively affect a salesperson's sense of work meaning. The empirical evidence confirms that employees react negatively when process control is employed in tasks requiring low procedural knowledge (Jaworski & MacInnis, 1989). Therefore, the following hypothesis states: H4. Product complexity moderates the indirect relationship between capability control and turnover intention through work meaning. The relationship between capability control and work meaning will be stronger (more positive) for higher levels of product complexity. 2.4.2. Moderating effect on the indirect effect via work overload According to Van den Broeck, Vansteenkiste, De Witte, Soenens, and Lens (2010), when employees face challenging job demands, it is easier for them to associate the effort needed to handle those demands with the likelihood of obtaining a favorable outcome. This study proposes that capability control helps to clarify this association, both by improving self-confidence and by developing sales skills. A salesperson with higher abilities and knowledge levels would have to exert less effort to meet their sales objectives. They can focus on designing smart selling techniques to accomplish sales tasks (Rouzies & Macquin, 2003). This study thus expects a negative relationship between capability control and overload when the complexity of a product is high. However, the contrasting effect of capability control under the situation of low product complexity can be detrimental. According to the theory of control and complexity (Frese, 1987; Speier & Frese, 1997), experienced control can reduce stress. The theory further suggests that lack of complexity leads to boredom and strain because it deprives employees of the opportunity to utilize their intellectual resources. According to JD-R theory, bored employees use behavioral strategies to change the design of the job itself. In ‘quiet day’ conditions, employees engage in playful work design and change job activities to make them more fun or competitive (Bakker, Hetland, Olsen, Espevik, & De Vries, 2020). When selling simple products, guidance from management on how to sell a product likely reduces the salespersons capacity to cope with boredom further exhausting individual job resources. Relying on this theory, it is likely that employees with a high level of capability control perceive there is a lack of opportunities to utilize their skills when product complexity is low. This perceived lack of opportunities leads to a sense of work overload irrespective of the actual workload level. Previous studies show that opportunities for skill utilization lower workers' burnout level (Jergensen, 2018; Leiter, 1990). Accordingly, this study hypothesizes that: H5. Product complexity moderates the indirect relationship between capability control and turnover intention through work overload. The relationship between capability control and work overload will be positive for high levels of product complexity but negative for low levels of product complexity. 2.4.3. Moderating effect on the direct effect of capability control In addition to the mediated moderation, this study also hypothesizes that product complexity positively moderates the direct relationship between capability control and turnover. According to JD-R theory, “on the days where work pressure is really high, employees have a very good reason to ask for social support from colleagues and coaching by the supervisor” (Bakker et al., 2020, p. 4). For subordinates, perception of a supervisor's support in terms of skills and abilities development in the face of job demands can work as a resource to cope with those demands (Miao & Evans, 2013). As empirical evidence for this argument, Bello and Gilliland (1997) confirmed a positive relationship between product B. Bande et al. Industrial Marketing Management 96 (2021) 100–112 104 complexity and the use of process or behavior control. Similarly, the interaction between customer purchase complexity and sales capabilities training was found to be positively related to turnover via salesforce morale (Panagopoulos, Hochstein, Baker, & Pimentel, 2018). This study thus expects a positive interactive effect of capability control and product complexity on reducing salesperson turnover intention. However, in a low product complexity scenario, capability control has less effect in reducing turnover. As Mitchell, Holtom, Lee, Sablynski, and Erez (2001) explain, employees' job knowledge, skills and abilities need to match job requirements. The authors confirm that the better the fit, the higher the chances of an employee remaining personally and professionally attached to an organization. In a situation of low product complexity, an employee may perceive a mismatch between job demands and job knowledge and skills because the selling skills and abilities developed through capability control are higher than needed. As a consequence of this discrepancy, the possibility of an employee leaving an organization to seek a position outside that matches their goals, knowledge, and job demands, increases. Although the relationship between capability control and salesperson turnover is expected to be generally negative due to the social exchange effect, it is also expected that the effect of capability control on turnover intention will become weaker in a low product complexity scenario. Accordingly: H6. Product complexity moderates the direct relationship between capability control and turnover intention. The relationship between capability control and turnover intention will be stronger (more negative) for higher levels of product complexity. 3. Method 3.1. Data collection To test the proposed model, data was collected from industrial salespeople and their direct supervisors who work for organizations in a variety of industries. A market research agency invited 151 organizations to participate in this study. These companies are located in Galicia, the northwest region of Spain. The sample included a representative sample of industries such as technology, manufacturing, construction, wholesale, and financial services. Invited salespeople engage in business-to-business sales. All participating firms operated for at least 10 years and had received commercial recognition from the local chamber of commerce. Of the companies initially contacted, six declined to participate in the study. A researcher personally contacted a sales manager and a randomly selected salesperson in each of the participating companies. The final sample consisted of 145 salespeople and 145 supervisors, yielding 145 supervisor–salesperson dyads. Most of the participating salespeople were male (72.4%), with an average age of 39.5 years (SD =7.7) and an average organizational tenure of 8 years (SD =7.7). Of the supervisors, 88.3% were male, the average age was 44.9 years (SD =9.3) and the average organizational tenure was 13.2 years (SD =8.9). These demographics were compared with census data and no differences were found in terms of employee age and experience. We also performed Standard Industrial Classification (SIC) code analyses to verify that the sample was representative of the northwest region of Spain. The dyadic nature of the sample and its composition is suitable for examining the proposed model. 3.2. Measures All measures used in this study are adapted from previous research. Participants responded to all items on a seven-point Likert scale anchored by strongly disagree (1) and strongly agree (7). Following commonly used back-translation procedures (i.e. Brislin, 1970), the scale items were translated into Spanish. Five items from Miao et al. (2007) were used to assess the information and rewards dimensions of capability control. Scale items were adapted to be self-assessed by the supervisor. This scale yielded good reliability indices (Cronbach's Alpha =0.84 and composite reliability =0.87). Work overload was assessed with the five-item scale developed by Roberts, Lapidus, and Chonko (1997). This measure captures the salesperson's perception that there are too many tasks to complete in a given time. Cronbach's Alpha for this scale was 0.91 and composite reliability was 0.84. Work meaning was measured using the three-items scale from Spreitzer (1995), which captures the extent to which a salesperson perceives their work as meaningful and important. This scale produced a good reliability indicator (Cronbach's Alpha =0.84; composite reliability =0.84). The technological complexity of the product was assessed by the supervisor with 3-items adapted from the scale by Homburg, Müller, and Klarmann (2011). This measure produced satisfactory reliability (Cronbach's Alpha =0.80; composite reliability =0.81). Turnover intention was assessed by the salesperson using a scale that captures a salesperson's propensity to leave an organization. This three-items scale is adopted from Fournier, Tanner Jr, Chonko, and Manolis (2010). The scale yielded good reliability (Cronbach's Alpha =0.94; composite reliability = 0.94). 3.2.1. Control variables Control variables comprise salesperson gender (1 =male, 2 =female) and salesperson organizational tenure (years). Prior research indicates that behavior control is more effective among less experienced employees (Katsikea et al., 2015). Tenure and gender have also been related to stress perceptions and turnover intention (e.g. Jaramillo, Grisaffe, Chonko, & Roberts, 2009). 3.3. Measurement validation Prior to testing the proposed hypotheses, a confirmatory factor analysis (CFA) was conducted to assess the validity of the multi-item scales within the conceptual model (Gerbing & Anderson, 1988). The parameters were estimated using the maximum likelihood method. Although the chi-square statistic is significant ( χ 2 =201.61; df =138; p <0.01), other fit indexes (RMSEA =0.57; TLI =0.95; CFI =0.96; IFI = 0.96) suggest an acceptable model fit. All factor loadings are greater than the 0.5 value and most of them surpass the 0.70 value. All constructs also exceed the threshold values of 0.70 for the composite reliability and 0.50 for the average variance extracted (AVE) (Bagozzi & Yi, 1988). All items and factor loadings are presented in Appendix 1. Discriminant validity was attained as the average variance extracted (AVE) of each construct exceeded its correlations with the other constructs (Fornell & Larcker, 1981). Descriptive statistics, Cronbach's Alpha, and correlations are shown in Table 1. The dyadic nature of our data ensures that the effects of capability control and product complexity on work meaning, work overload, and turnover intention are unlikely affected by common method variance. (See Fig. 1.) 4. Discussion of results We employed conditional process analyses to test the proposed hypotheses. This approach is appropriate when the proposed conceptualization includes moderated mediated models that are tested simultaneously (e.g., Chai, Li, Tangpong, & Clauss, 2020). As Hayes (2018, p. 393) states, “conditional process analysis is used when the analytical goal is to describe and understand the conditional nature of the mechanism or mechanisms by which a variable transmits its effect on another.” Conditional process analyses provide a deeper understanding of the mechanism or mechanisms through which an effect operates and involves “describing and analyzing the when of the how” (Hayes, 2018, p. 7). In order to test the proposed model and conduct a conditional process analysis, Hayes' (2018) PROCESS macro (model 8; 5000 bootstrap samples) was used. This tool allows inference on direct and indirect B. Bande et al. Industrial Marketing Management 96 (2021) 100–112 105 effects, testing the moderation of mediation, calculating a confidence interval for the index of moderated mediation and facilitates the use of the pick-a-point procedure. As shown in Fig. 2, one model for each of the mediating variables was run: model A (work meaning) and model B (work overload). Salesperson gender and salesperson organizational tenure were included in both models as controls in the analyses. 4.1. Model A (work meaning as mediating variable) The proposed model is tested with moderated regression analyses. Relationships that contain dyadic data from a salesperson and her/his corresponding manager are frequently evaluated with this method (e.g., Gonzalez & Claro, 2019). The results of the moderated mediation analyses using ordinary least squares (OLS) and mean-centering capability control and product complexity show that capability control reduces turnover intention directly and indirectly through a mediating processes that involve work meaning. Also that this indirect effect is moderated by product complexity. As shown in Table 2A, capability control is positively related to work meaning (β =0.30; p <0.01; SE = 0.09) which in turn is negatively related to turnover intention (β = −0.58; p <0.001; SE =0.12). The interaction capability control X product complexity is positively related to work meaning (β =0.20; p < 0.001; SE =0.06) supporting the moderating effect. The confidence interval for the index of moderated mediation is entirely below zero (Index = − 0.112; SE (Boot) =0.05; 95% Boot CI = − 0.23; −0.03), which is evidence of moderation of the indirect effect of capability control on Table 1 Means, standard deviations and correlations among variables. Variable Mean S.D. 1 2 3 4 5 6 7 1. Capability control 5.64 0.86 (0.84) 2. Work meaning 5.67 0.98 0.21* (0.84) 3. Work overload 3.75 1.58 −0.17* −0.35** (0.91) 4. Product complexity 5.32 1.26 0.38** 0.09 0.12 (0.80) 5. Turnover intention 3.02 1.86 −0.23** −0.41** 0.63** 0.13 (0.94) 6. Salesperson tenure 11.57 8.46 −0.01 0.22** −0.17* 0.05 −0.24** – 7. Salesperson gender 1.28 0.44 0.06 −0.18* 0.17* 0.05 0.07 0.10 – Note: **p < 0.01, *p < 0.05. Alphas appear diagonally. Fig. 1. Proposed model. Model A Model B Capability control Turnover intention Work meaning Capability control Work overload Turnover intention Product complexity Product complexity Fig. 2. The moderated mediation models. B. Bande et al. Industrial Marketing Management 96 (2021) 100–112 106 turnover intention through work meaning. As shown in Table 3A, the indirect effect is negative, and it can be discarded as null for values equal or above the 25th percentile in the product complexity scale, considering that the bootstrapped 95% confidence intervals do not include zero (Aiken & West, 1991)., Hypothesis 2 is thus supported. As predicted, capability control reduces turnover intention via increased work meaning. These findings demonstrate that capability control works as a supervisor's control system as well as a subordinate's job resource. The feedback that salespeople receive from managers through capability control is an important resource that increases perceptions that the manager provides salespeople with feedback to improve selling skills and become more effective at selling. Managerial feedback provides salespeople with increased motivation and helps them handle their workload, giving them reasons to stay in the organization. However, as discussed below, the positive effects of capability control are contingent on product complexity. Table 3A shows the indirect conditional effects of capability control on salesperson turnover intention at different values of product complexity. Post hoc analyses of interaction terms using Johnson and Neyman's (1936) regions-of-significance procedure (Table 4) revealed that for salespeople dealing with levels of product complexity below 1.91, capability control is negatively related to work meaning. However, for levels of product complexity above 4.73, capability control is positively related to work meaning. These combined findings provide support for hypothesis 4. Capability control reduces turnover intention via increased work meaning when product complexity is high; however, when product complexity is low, capability control increases turnover intention by decreasing work meaning. Moreover, as shown in Table 2A, capability control is found to be directly and negatively associated with salesperson turnover intention (β = − 0.72; p <0.001; SE =0.17). Also consistent with our predictions, as shown by the significant effect of the interaction term (β = − 0.32; p < 0.01; SE =0.10), product complexity moderates the direct effect of capability control on salesperson turnover intention. In particular, the effect is stronger (more negative) as the complexity of a product increases. However, regions of significant calculations via Johnson–Neyman's procedure (Table 5) show that for low levels of product complexity (below 1.15), the effect of capability control on salesperson turnover is positive and significant. For values of product complexity above 4.15, the effect of capability control on salesperson intention to quit is also significant but negative. Thus, hypothesis 6 is supported. Capability control reduces turnover intention when product complexity is high; however, when product complexity is low, capability control increases turnover intention. Although not hypothesized, product complexity is positively related to turnover intention (β =0.36; p <0.01; SE =0.11). 4.2. Model B (work overload as mediating variable) The moderated mediation analyses show that capability control is negatively related to work overload (β = − 0.58; p <0.001; SE =0.16), which in turn is positively related to turnover intention (β = − 0.59; p < 0.001; SE =0.08) (Table 2B). As shown in Table 3B, the indirect effect is also significant and the bootstrapped 95% confidence intervals do not include zero for 25th, 50th, 75th, and 90th percentiles in the product complexity scale (Aiken et al., 1991). The findings show that capability control reduces turnover intention via reduced work overload. Thus, hypothesis 3 is supported. Regarding the moderating effect of product complexity, as shown in Table 2B, the interaction between capability control and product complexity is negatively related to work overload (β = − 0.31; p <0.001; SE =0.09). The depiction of this interaction (see Fig. 3) shows that capability control is negatively related to work overload when product complexity presents high values; on the contrary, when product complexity is low, capability control is positively related to work overload. In addition, the indirect effect of capability control on turnover intention through work overload is contingent on product complexity. The indirect conditional effects of capability control on salesperson turnover intention at different values of the moderator are shown in Table 3B. In addition, the confidence interval for the index of moderated mediation is entirely below zero (Index = − 0.18; SE (Boot) =0.07; 95% Boot CI = − 0.39; −0.08). Capability control thus reduces turnover intention via reduced work overload when product complexity is high; however, when product complexity is low, capability control increases turnover intention by increasing work overload. Overall, hypothesis 5 is supported. These findings suggests that under conditions of low product complexity, capability control increases job demands and is no longer perceived by salespeople as a resource. Moreover and similarly to model A, capability control is found to be directly and negatively related to salesperson intention to quit (β = −0.53; p <0.01; SE =0.15) (see Table 2B). In addition, and in coherence with this study's predictions, product complexity moderates the direct effect of capability control on salesperson turnover intention (β = − 0.24; p <0.01; SE =0.09). As concluded in model A, the effect is stronger (more negative) as the complexity of a product increases. Thus, hypothesis 6 is again supported. The results obtained, as a whole, strongly support hypothesis 1, which established that capability control and turnover intention are negatively related. The results indicate that this relationship is direct and indirect (through work meaning and work overload) and dependent on product complexity. Table 2A Model coefficients for the conditional process analysis (model A: Work meaning). Antecedents Consequences M (Work meaning) Y (Turnover intention) Coeff. SE p Coeff. SE p Constant 5.65 0.27 <0.001 7.04 0.97 <0.001 Capability control 0.30 0.09 <0.01 −0.72 0.17 <0.001 Product complexity 0.06 0.06 0.3369 0.36 0.11 <0.01 Work meaning – – – −0.58 0.15 <0.001 Capability control X product complexity 0.20 0.05 <0.001 −0.32 0.10 <0.01 Salesperson tenure 0.02 0.00 0.1285 −0.03 0.01 <0.05 Salesperson gender −0.26 0.17 0.1285 −0.13 0.30 0.6463 R 2 =0.19, F(5, 131) = 6.33, p =0 <0.001 R 2 =0.35, F(6,130) = 12.00, p =0 <0.001 Table 2B Model coefficients for the conditional process analysis (model B: Work overload). Antecedents Consequences M (Work overload) Y (Turnover intention) Coeff. SE p Coeff. SE p Constant 3.47 0.44 <0.001 1.64 0.50 <0.01 Capability control −0.58 0.16 <0.001 −0.53 0.15 <0.01 Product complexity 0.19 0.10 0.07 0.18 0.10 0.0699 Work overload – – – 0.59 0.08 <0.001 Capability control X product complexity −0.31 0.09 <0.001 −0.24 0.09 <0.01 Salesperson tenure −0.02 0.01 0.0730 −0.03 0.01 <0.05 Salesperson gender 0.53 0.28 0.0632 −0.27 0.27 0.3154 R 2 =0.19, F(5,132) = 6.36, p ≤0.001 R 2 =0.48, F(6,131) = 20.24, p ≤0.001 B. Bande et al. Industrial Marketing Management 96 (2021) 100–112 107 5. Conclusions 5.1. Theoretical contributions This study contributes to the literature in several ways. First, while previous studies on sales control have mainly focused on capability control aspects of information and reinforcement, this study has theorized and revealed its role as a job resource for subordinates. Manuscript findings bring support to precepts from JD-R theory that under certain conditions capability control operates as a supervisor's control system and a subordinate's job resource (Miao & Evans, 2013). This viewpoint can be a response to research calls to analyze the social exchange that arises when employees perceive that their organization is providing them with development opportunities (Benson et al., 2004). Second, the study expands the knowledge of the consequences of sales management control systems in an industrial selling context by revealing that capability control contributes to a reduction in salesperson turnover intentions, both directly and indirectly. The results show the direct negative effect of capability control on a salesperson's turnover intention. This finding is consistent with the unfolding turnover model, which states that employee growth and development of skills and abilities reduces the probability of an employee's turnover intention (Lee & Maurer, 1997). On the contrary, this study's findings contradict human capital theory (Becker, 1965) that suggests that companies will be unwilling to develop their employees' skills because this development is likely to increase their intention to leave the company. Although company investments in salesperson skills increase the attractiveness of a salesperson in the job market, they also provide the salesperson with resources to effectively cope with job demands while increasing work meaning, ultimately providing a reason for salespeople to stay in the organization. This study's model also proposes two mediating variables that help with understanding the mechanisms through which this behavior-based control reduces turnover intentions. It represents an advance over most previous empirical studies on turnover, which have mostly focused on analyzing direct relationships (Allen et al., 2014). The analyses reveal Table 3A Direct and indirect effects of capability control on salesperson turnover intention (model A: Work meaning). Conditional Indirect effects of Capability control on salesperson turnover through WORK MEANING at values of Product complexity (model A) Product complexity* Indirect effect BootSE BootLLCI BootULCI 3.33 0.02 0.09 −0.17 0.21 4.66 −0.11 0.07 −0.29 −0.00 5.66 −0.22 0.08 −0.43 −0.08 6.00 −0.25 0.09 −0.49 −0.10 6.66 −0.32 0.12 −0.61 −0.12 Conditional Direct effects of Capability control on salesperson turnover intention at values of Product complexity (model A) Product complexity Direct effect SE LLCI ULCI 3.33 −0.08 0.20 −0.49 0.32 4.66 −0.39 0.15 −0.69 −0.08 5.66 −0.62 0.17 −0.95 −0.28 6.00 −0.69 0.18 −1.06 −0.32 6.66 −0.84 0.22 −1.29 −0.40 * Values are for the 10th, 25th, 50th, 75th, and 90th percentiles. Table 3B Direct and indirect effects of capability control on salesperson turnover intention (model B: Work overload). Conditional Indirect effects of Capability control on salesperson turnover through WORK OVERLOAD at values of Product complexity (model B) Product complexity* Indirect effect BootSE BootLLCI BootULCI 3.33 0.02 0.15 −0.19 0.40 4.66 −0.22 0.11 −0.46 −0.01 5.66 −0.41 0.13 −0.73 −0.19 6.00 −0.47 0.15 −0.83 −0.23 6.66 −0.60 0.19 −1.06 −0.30 Conditional Direct effects of Capability control on salesperson turnover intention at values of Product complexity (model B) Product complexity Direct effect SE LLCI ULCI 3.33 −0.07 0.23 −0.53 0.38 4.66 −0.51 0.17 −0.85 −0.17 5.66 −0.84 0.18 −1.21 −0.46 6.00 −0.94 0.20 −1.35 −0.54 6.66 −1.16 0.25 −1.66 −0.66 * Values are for the 10th, 25th, 50th, 75th, and 90th percentiles. Table 4 Conditional effects of capability control on work meaning at values of product complexity. Product complexity Effect S.E. p LLCI ULCI 1.00 −0.57 0.23 0.01 −1.04 −0.09 1.30 −0.50 0.22 0.02 −0.95 −0.06 1.60 −0.44 0.20 0.03 −0.86 −0.03 1.90 −0.38 0.19 0.04 −0.77 −0.00 1.91 −0.38 0.19 0.05 −0.76 0.00 2.20 −0.32 0.18 0.07 −0.68 0.03 2.50 −0.26 0.16 0.11 −0.59 0.06 2.80 −0.20 0.15 0.18 −0.50 0.09 3.10 −0.14 0.14 0.31 −0.42 0.13 3.40 −0.08 0.12 0.52 −0.33 0.17 3.70 −0.02 0.11 0.86 −0.25 0.21 4.00 0.04 0.10 0.70 −0.17 0.25 4.30 0.10 0.10 0.31 −0.10 0.30 4.60 0.16 0.09 0.09 −0.03 0.35 4.73 0.19 0.09 0.05 0.00 0.38 4.90 0.22 0.09 0.02 0.03 0.41 5.20 0.28 0.09 0.00 0.09 0.47 5.50 0.34 0.10 0.00 0.14 0.54 5.80 0.40 0.10 0.00 0.19 0.62 6.10 0.46 0.11 0.00 0.24 0.69 6.40 0.53 0.12 0.00 0.28 0.77 6.70 0.59 0.13 0.00 0.32 0.86 7.00 0.65 0.14 0.00 0.35 0.94 Table 5 Conditional effects of capability control on turnover intention at values of product complexity. Product complexity Effect S.E. p LLCI ULCI 1.00 0.99 0.43 0.02 0.13 1.86 1.30 0.86 0.40 0.03 0.05 1.67 1.60 0.77 0.39 0.05 0.00 1.54 1.90 0.73 0.38 0.05 −0.02 1.48 1.91 0.60 0.35 0.09 −0.10 1.30 2.20 0.47 0.32 0.15 −0.17 1.12 2.50 0.34 0.30 0.26 −0.25 0.94 2.80 0.21 0.27 0.45 −0.34 0.76 3.10 0.07 0.25 0.75 −0.42 0.58 3.40 −0.05 0.23 0.82 −0.51 0.41 3.70 −0.18 0.21 0.40 −0.60 0.24 4.00 −0.31 0.19 0.11 −0.70 0.08 4.30 −0.38 0.19 0.05 −0.76 0.00 4.60 −0.44 0.18 0.01 −0.81 −0.07 4.73 −0.57 0.17 0.00 −0.92 −0.22 4.90 −0.70 0.17 0.00 −1.05 −0.35 5.20 −0.83 0.17 0.00 −1.18 −0.48 5.50 −0.96 0.18 0.00 −1.32 −0.60 5.80 −1.09 0.19 0.00 −1.48 −0.71 6.10 −1.22 0.21 0.00 −1.64 −0.81 6.40 −1.35 0.22 0.00 −1.81 −0.90 6.70 −1.48 0.24 0.00 −1.98 −0.99 7.00 −1.62 0.27 0.00 −2.15 −1.08 B. Bande et al. Industrial Marketing Management 96 (2021) 100–112 108 that sales management capability control reduces job stress (work overload) and increases job engagement (work meaning), which in turn contributes to lowering salesperson turnover intention. This brings support to precepts from JD-R theory that job resources reduce strain perceptions and increase motivation, ultimately leading to important organizational outcomes like reduced turnover (Bakker & Demerouti, 2007). The finding on the negative effect of capability control on work overload is consistent with previous findings on other forms of role stress, such as role ambiguity and role conflict (Katsikea et al., 2015). It also confirms the reasoning that greater formalization, which is characteristic of a behavior-based control, reduces role stress (Oliver & Anderson, 1994). Moreover, by examining the mediating effect of work meaning in the supervisor–employee context, this study responds to a research call that researchers should “pay more attention to the influence of dyadic interpersonal relationships on the meaning of work” (Rosso et al., 2010, p. 102). This investigation also contributes to an under-researched topic of how work meaning is formed and developed (Fock et al., 2010). Third, we delve into the relationship between sales management behavior-based control and job stress – the relationship between behavioral control and job stress is unclear in the literature. While some studies show a negative relationship between behavioral control and job stress (Katsikea et al., 2015; Piercy et al., 2001), other studies report a non-significant relationship (Challagalla & Shervani, 1996; Katsikea et al., 2015). This study examines the complexity of a product as a moderator in the relationship between capability control and work overload because mixed findings have indicated the existence of boundary conditions. When the complexity of a product is considered, the effects of capability control are dramatically changed. This confirms Jaworski's (1988) argument regarding the need to connect the control instrument and the environmental context to attain the desired organizational results. The negative effect of capability control on turnover intention, both directly and indirectly through work meaning and work overload, only takes place when the complexity of a product is high. These findings align with JD-R theory (Demerouti et al., 2001) where high job demand (high product complexity) and job resource (capability control) are related to higher levels of motivation and increased personal growth. On the contrary, when the complexity of a product is low, capability control increases work overload and decreases work meaning, while having a positive effect on salesperson turnover intention. When the complexity of the product is low, capability control is no longer a valuable resource but rather a work demand that increases work overload and decreases motivation. Under low product complexity, managerial attempts to monitor, evaluate, and coach salespeople on how to sell are counterproductive. When selling simple products, salespeople may see these managerial behaviors as a sign of micromanagement (Peesker et al., 2019) and a threat to their job autonomy. Salespeople are willing to quit their jobs when facing unwanted managerial controls and advise. The goal of sales management control systems is to influence employee behavior and attitudes to achieve organizational goals (Baldauf et al., 2005). However, the study confirms that behavior-based control can be counterproductive depending on the characteristics of the sales context. The results support the notion that consequences of control mechanisms are contingent on the specific context in which the control system is used (Jaworski, 1988) and they align with the notion that employees' negative reactions to formal supervisory controls depend on contextual factors (Ramaswami, 1996). This study's findings on the moderating effect of product complexity can also contribute to clarifying the relationship between improving employees' employability and their turnover intention; other studies have obtained mixed results. While some studies identify that the improvement of employees' skills has a positive influence on their intention to quit their organization (Lynch, 1991), other studies report a non-significant relationship between both variables (Loewenstein & Spletzer, 1997). This study's results suggest that product complexity affects a company's return on investment in employees' employability. These results are also consistent with recent studies on transactional leadership and its influence on complex sales contexts. Transactional leadership, which involves providing subordinates with detailed guidance on how to perform their tasks and achieve work objectives, has proven to be counterproductive in certain sales situations because it reduces task autonomy (Schmitz, Lee, & Lilien, 2014). Furthermore, these findings reveal the importance of developing work meaning in human resource management. One stream of research posits that organizations are morally obliged to help employees experience work meaning (Michaelson, 2005). This study's findings suggest that capability control can help organizations to fulfill these ethical Fig. 3. Interactive effects of product complexity and capability control on work overload. B. Bande et al.