Does choice of sales control conceptualization matter? An empirical comparison of existing conceptualizations and directions for future research

Abstract

Sales controls represent an important area of sales research. Unfortunately, knowledge to date tends to be fragmented in that prior work has followed one of the two notable conceptualizations which, however, take a different route to studying sales controls: the Oliver and Anderson (O&A) index and the Jaworski and Colleagues (J&C) measures. Consequently, important questions remain unaddressed: Does choice of sales control conceptualization matter to sales research? What is the degree of similarity between the two conceptualizations and how do they compare against measurement qualities and effects on performance? Drawing on a unique dataset that matches survey data to objective, time-lagged, firm financial performance data, we reveal that choice of sales control conceptualization matters. Surprisingly, unlike current assumptions, the O&A index, is equally related to J&C's output and process controls whereas it reflects both formal and informal types of control. Analyses show that at least some components of the O&A index are not identical to J&C's process or output controls and thus measures from each conceptualization cannot be substituted. Moreover, the size and nature of sales controls’ effects on sales force performance differ depending on the conceptualization employed, thus strengthening the view that the two conceptualizations should not be used interchangeably. Finally, results reveal an intriguing pattern of nonlinear effects such as that – beyond a certain point – process (but not behavioral) control may be detrimental to customer relationship performance. We conclude by contributing ideas on how to move forward with the development of a new, modern measure of sales control.

Keywords:

• sales controls
• salesperson
• performance
• measurement
• formative/reflective indicators

Acknowledgements

An earlier version of this paper won the Best in Sales Track Paper Award, 2014 Winter Marketing Educators' Conference. The first author would like to thank his 9 research assistants that aided with data collection as well the senior sales executives that participated in this research. The authors would also like to extend their appreciation to the Editor, Manfred Krafft, the two anonymous reviewers, and Sönke Albers for their many helpful comments.

Notes

1. Given that our goal in this study is to compare the two conceptualizations rather than develop a new one, our discussion focuses on their differences with respect to the ‘attribute’ step of this procedure that refers to ‘sales controls’ (the concrete singular ‘object’ – that is, the company – and the ‘rater entity’ – that is, managers – are the same in both conceptualizations). We are grateful to one anonymous Reviewer for this suggestion.

2. A comprehensive review of the literature on the two conceptualizations is available upon request.

3. We were able to collect firm financial performance data for a subset of the sample firms due to regulations allowing certain types of firms to not publicly disclose financial data. Additionally, some firms were merged or acquired while a few closed during the years of investigation. To explore whether availability bias was present we compared the firms for which we lacked objective financial data with those for which these data were available on several variables. Specifically, we did not detect any significant differences with respect to the number of employees, average number of sales calls done by a salesperson in a typical week, number of customers per salesperson, number of salespeople that quit each year, total annual gross pay for a salesperson, percentage of straight salary in compensation plan, and the number of hierarchical levels in the sales organization.

4. We are grateful to one anonymous Reviewer for this suggestion.

5. Our empirical analysis assumes and tests for a linear relationship between output control and outcome performance. In keeping with Iyengar's perspective on motivation (e.g., DeVoe and Iyengar 2004), one might envision that there may be a J-shaped relationship between the two constructs in that early the extrinsic motivation that comes from output control hurts outcome performance while later it facilitates performance. To test this possibility in our data, we ran the following hierarchical regression specification:

Results from this analysis show that the linear term is positive β₁ = .75; p < .01 while the second derivative is nonsignificant β₂ = −.058; p > .10). As such, there is no evidence for a J-shaped relationship between output control and outcome performance but rather for a continuously positive relationship. We thank one anonymous Reviewer for this suggestion.

6. We thank one anonymous Reviewer for this suggestion.

7. See Note 6.