|
...organization with pooling, each salesperson may have a primary responsibility for some product line but is also able to sell some subset of the other product lines. Two important factors that influence the performance of such a salesforce are pricing and experience. Product price will influence the demand for a product, while the salesperson's experience influences the likelihood of making a sale. Therefore, when sizing and structuring a salesforce, a firm must consider pricing and the impact of staffing decisions on the experience of the salesforce. Understaffing the sales team for a product may lead to lost sales and possibly to lost future market share. Overstaffing the sales team can be expensive since good sales people are typically well compensated. Assigning staff to product lines must also be done carefully. Some products are complex and sales success depends upon experience, while simpler products require little experience on the part of the salesforce.
In this paper, we examine the interactions among staffing, learning, and pricing in the management of a salesforce. We develop a model of a salesforce that receives sales leads with the sales volume generated from each lead depending upon the experience of the salesperson and the price of the product. The firm's goal is to maximize profit by adjusting product prices as well as how many salespeople are allocated to each product. Our model applies to firms with the following three attributes: (i) sufficient market power to have control over pricing; (ii) complex products that require experience to sell effectively; and (iii) large marketing efforts, such as advertising in the media and at tradeshows, so that most of the leads are generated by activities outside of the salesforce. Specifically, a firm selling high-technology industrial products in a mature market would satisfy all of these criteria (such a firm satisfies attribute (iii), for in a mature market there are few untapped sales leads, and its salesforce focuses its energy on following up requests by existing customers rather than unearthing new customers). After developing the model, we estimate its parameters using data collected from the salesforce of one such firm.
This paper makes use of the framework developed in Pinker and Shumsky (2000) for modeling learning in service systems. We combine a model of job tenure with a model of experience-based learning and apply it to the problem of salesforce design. The current paper, however, has four significant differences and extensions that lead to interesting results not previously seen in the literature. Firstly, the service level, defined as the throughput of sales leads, is determined endogenously while it is exogenously determined in Pinker and Shumsky (2000). Therefore, we are able to see how optimal staffing responds to changes to cost parameters. Given this additional degree of freedom, we also find that worker learning can significantly dampen the effect of rising costs (or decreasing margins) on staffing levels. For example, if learning is a significant factor, an increase or decrease in the cost of salesforce compensation has a relatively small impact on the optimal salesforce size. Secondly, we model a more complex routing of customers (sales leads) to workers (salespeople) that more accurately reflects a common practice in sales organizations. Given this routing, we arrive at the surprising result that, with learning, pooling of workers may lead to optimal staffing levels that are higher than when workers specialize. This contradicts the conventional wisdom that the economy of scale provided by pooling reduces staffing requirements. Thirdly, the model incorporates pricing decisions and their effect on demand. As a result we are able to study the relationship among staffing levels, job flexibility, and price levels. In particular, we find that the optimal salesforce size declines as price sensitivity increases. We also find that when learning is a significant factor in determining sales volume, a specialized (or "exclusive") salesforce leads to higher optimal prices than the optimal prices for a pooled salesforce. Fourthly, motivated by data collected from the salesforce of a large manufacturer, we propose a learning-curve model different from the model in Pinker and Shumsky (2000), and we demonstrate how parameters of the model may be estimated from the data.
In the next section, we present an overview of the relevant literature. In Section 3, we formulate our model by integrating a service process model with an employee tenure model, a model of experience-based learning, and a model of consumer demand. Section 4 contains analytical results that describe the impact of various parameters on the optimal price. Section 5 describes the analysis of industry sales data that provide baseline parameters for the numerical experiments of Section 6. These numerical experiments provide insights into how learning effects staffing and pricing. Section 7 summarizes our results and discusses possible extensions of the model.
2. Literature review
As noted above, this paper is most closely related to Pinker and Shumsky (2000). Other researchers have also considered parts of the problem addressed in this paper but we believe that ours is the first to integrate all of them into one model. Some researchers have studied the control problem of how to hire, fire and promote workers to maintain appropriate staff levels when career paths are stochastic, and these are listed in Pinker and Shumsky (2000). None of these studies consider the effect of pricing on staffing and therefore do not connect staffing and learning to sales, limiting their applicability to salesforce design.
The literature on salesforce management has focused on either salesforce incentives, see for example Basu et al. (1985), Lal and Srinivasan (1993), Joseph and Thevaranjan (1998) and Bhardwaj (2001), or salesforce sizing and mix issues. In this paper, we do not focus our attention on compensation issues. In particular, we assume that effort is perfectly observable and hence a fixed wage, forcing contract is optimal.
Montgomery and Urban (1969) and Lucas et al. (1975) use profit maximization models to solve for the optimal salesforce size. A limitation of their approach is that they ignore the presence of multiple products and/or territories. Lodish et al. (1988) use a more sophisticated approach to modeling the issue in the case of one particular firm. They showed, for this one firm, that adding salespeople and redeploying them would result in increased profits. Zoltners (1976), Lodish (1976, 1980), Rangaswamy et al. (1990), and Mantrala et al. (1992) consider the problem of finding the optimal allocation of salespeople to territories, products or customers. These studies use static frameworks that do not incorporate learning effects within the salesforce. Another issue that has not been addressed adequately in the literature is specialization and the effect it has on structuring the salesforce. Given that salespeople often specialize in particular products and that such specialists are scarce, there are instances when a non-specialist serves a customer, which may have an impact on sales.
Dewan and Mendelson (1990). Stidham (1992) and So and Song (1998) are examples of works in which both capacity and pricing are endogenous to the firm's decision problem. In all of these, the firm is modeled as a single-server queue and capacity is determined by the service rate. In this paper, we are explicitly modeling capacity as the staffing level in a multi-server queue. Furthermore, we consider the interaction of parallel queues serving different customer types. Finally, in our model price determines the sales quantity rather than the customer arrival process. As we mentioned in the Introduction, our formulation is appropriate for environments in which sales leads are "handed off" to the salesforce.
To summarize, we are studying a set of problems that have been looked at in isolation from various perspectives in the marketing and operations literature. We aim to integrate these diverse perspectives into a unified...
NOTE: All illustrations and photos
have been removed from this article.
More articles from IIE Transactions
Statistical Methods for Six-Sigma in R & D Manufacturing.(Book Review), December 01, 2004
Simulation optimization: a comprehensive review on theory and applicat..., November 01, 2004
Strategic inventory deployment in the steel industry., November 01, 2004
Development and evaluation of cross-training policies for manufacturin..., October 01, 2004
Staff scheduling in high volume service facilities with downgrading., October 01, 2004
Looking for additional articles?
Search our database of over 3 million articles.
Looking for more in-depth information on this industry?
Search our complete database of Industry & Market reports by text, subject, publication
name or publication date.
About Goliath
Whether you're looking for sales prospects, competitive information, company
analysis or best practices in managing your organization,
Goliath can help you meet your business needs.
Our extensive business information databases empower business
professionals with both the breadth and depth of credible,
authoritative information they need to support their business
goals. Whether it be strategic planning, sales prospecting,
company research or defining management best practices -
Goliath is your leading source for accurate information.
|
