|
Article Excerpt
We quantified the net unit and profit impact of each promotion offered in 2003 by CVS, a leading U.S. drug retail chain, and analyzed the key drivers of variation in this net impact. We used this analysis to identify the least effective promotions and conducted a controlled field test to demonstrate the impact of eliminating them before chainwide implementation. Our key findings are as follows. First, approximately 45% of the gross lift from promotions is incremental for CVS. Further, for every unit of gross lift, 0.16 unit of some other product is purchased elsewhere in the store. Still, more than 50% of promotions are not profitable because the lower promotional margin is not sufficiently offset by incremental units. Second, there is substantial variation in net profit impact across categories. Our field test shows that eliminating promotions chainwide in 15 of the worst performing categories will decrease sales by about $7.8 million but will improve profit by approximately $52.6 million. This is very impressive given that CVS front store sales in 2003 were approximately $9 billion while the net profit impact of promotions was -$25.3 million.
Key words: promotion profitability; retail promotions; retail promotion impact History: This paper was received August 18, 2005, and was with the authors 3 months for 1 revision; processed by Gary Lilien.
*********
1. The Managerial Problem
1.1. Company Background CVS is a U.S. drugstore chain with over 4,000 retail stores in 35 states. It sells over 200 categories of health, beauty, edible grocery, and general merchandise products. "Front store" revenues (i.e., excluding prescriptions) were over $9 billion in 2003. In recent years, competition has intensified significantly in this industry, especially from mass merchants. It is telling, for instance, that in Texas $0.72 out of every consumer dollar spent on health and beauty products goes to mass merchants. These products are the mainstay of a drug chain.
CVS is a HILO (high-low) retailer, i.e., it offers price promotions on several products each week. Approximately 30% of its sales are made on promotion. The company's research showed that direct competition from every day low price mass retailers like Wal-Mart is lowering consumers' reference prices and hurting the price perception of CVS. Although the company's promotional prices are low, regular price points are in many cases so much higher than their every day low price competition as to not be sustainable. Further, effectiveness seems to vary widely among the tens of millions of promotions that CVS runs each year.
The company has no intention of abandoning its HILO positioning but it wants to ensure that its promotional pricing decisions are effective in competing with other retailers and in their net sales and profit impact for the company. To do so, it needs to (1) determine which promotions are effective, which ones are not, and why; (2) eliminate or modify ineffective promotions; and (3) reinvest the savings in more competitive prices and better merchandising.
1.2. Research Objective
The overarching purpose of this project was to quantify and improve the net impact of CVS's promotions. We compiled data for each of the over 30 million promotions offered in any of its approximately 4,000 stores during 2003 and (a) estimated the immediate or gross lift of the promotion; (b) decomposed the gross lift into three components, i.e., current period switching from other brands in the store, stockpiling from future period category sales in the store, and incremental lift for the store; (c) estimated the "halo" effect of the promotion, i.e., the extent to which it increases sales of other product categories in the store; (d) accounted for differences in promotional and regular margins and the funding provided by the manufacturer; and (e) calculated the net unit and profit impact of the promotion. We then conducted a meta-analysis of how this net impact varies with characteristics of the promotion, brand, category, and market, and identified the most unprofitable promotions as candidates for elimination. This analysis not only helped the company determine which promotions were particularly ineffective and which ones were more effective but also provided a deeper understanding of why such variation exists, and how to implement more effective promotions.
Promotion impact is certainly not a new subject for academics or practitioners, with the studies based on scanner data dating back to the eighties (e.g., McAlister 1986, Gupta 1988). Packaged goods manufacturers spend over 50% of their total marketing budgets on promotions (Cannondale Associates 2000) and market research companies generate significant revenue from the reports they provide to manufacturers quantifying the gross lift that manufacturers' brands get from promotion. Further, academics have spent considerable effort in understanding why the gross lift varies across categories and brands as well as in decomposing that lift into brand switching, consumer stockpiling, and primary demand components, at least for a few categories.
Why then did we need to conduct our own study of promotion effectiveness? The reason is that most work on promotion impact takes the perspective of manufacturers (see Cooper et al. 1999, Srinivasan et al. 2004 for two exceptions). Further, lack of publicly available cost data has prevented analyses of the profit impact of promotion. Retailers make decisions about promotions that are offered to consumers, they experience the direct financial impact of those decisions, and effectiveness for a retailer is quite different from effectiveness for a manufacturer. Any promotion-induced increase in category consumption benefits both manufacturers and retailers, but aside from that manufacturers benefit by switching consumers from competing brands whereas retailers benefit by switching consumers from competing stores.
Figure 1 depicts all the components of the gross lift for a promoted brand in a given store in a given period and highlights the components that comprise incremental lift for the retailer. Along with this incremental lift within the category, the retailer must consider any "halo" effect that the promotion may have on sales of other categories in the store to determine the net unit impact of the promotion in the store. Further, the retailer must account for regular and promotional margins, including manufacturer funding, to determine the net profit impact of the promotion in the store. Even after accounting for manufacturer funding, CVS' promotional margin is often lower than regular margin.
[FIGURE 1 OMITTED]
Although there is a significant gross lift when a promotion is offered, substantial variation seems to exist in the net impact of different promotions. How much of the gross lift is incremental and how much is simply switched from other brands in the store or pulled forward from consumers'...
|
