Knowledge that Transforms

This article describes the implementation of a promotion-event forecasting system, PromoCast™, and its performance in several pilot applications and validity studies. Pilot studies involved retail grocery chains with 95 to 185 stores per trading area. The goal was to provide short-term, tactical forecasts useful for planning promotions from a retailer's perspective. Thus, the forecast system must be able to handle any of the over 150,000 UPCs in each store's item master file, and must be scalable to produce approximately 800,000,000 forecasts per year across all the retailers served by efficient marketing services, inc. (ems, inc.). This is a much different task than one that confronts a manufacturer, even one with a broad product line. Manufacturers can benefit from custom modeling in a product line or category. Retailers need a production system that generates forecasts that help promotion planning. Marketing scientists have typically approached promotion analysis from the manufacturer's perspective. One objective of this article is to encourage marketing scientists to rethink promotion analysis from a different perspective. From the retailer's point of view the “planning unit” is the promotion event. Neither weekly store-tracking data nor shopping-trip data from consumer panels are easily aggregated to reflect total sales during a promotion event. We describe the promotion-event databases and the statistical model developed using these databases. The data are the strategic asset. Our goal is to help retailers use their data to increase the profitability of promotions. We have data on the performance of each UPC in each store under a variety of promotion conditions, on each store's adeptness at executing various styles of promotions, as well as on chain-wide historical performance for each UPC. We use many historical averages from these databases to build a 67-variable, regression-style model. The forecast incorporates a simple bias correction needed when using a log-transformed dependent variable (the natural log of total unit sales). We argue that the historical averages matching the planned ad and display conditions provide a benchmark superior to the widely used “base-times-lift” method. When aggregated into case units (the natural unit for product ordering), 69% of the forecasts in our first validation study were within ± one case compared to 39% within ± one case using the appropriate historical averages. We report the results of two over-time validity studies that reflect the value of our model for retailers. The limitations and implications of this planning tool for managerial decision making concerning stocking levels are discussed. Whenever historical data are the strategic asset we face inherent limitations. Our model does not forecast new products. The forecast error increases when an existing product is promoted in a new way. Over 99.5% of the time, we have full data from which to create a forecast. However, with a database for a typical chain market containing over 20 million promotion events in the 30-month time frame we use, 100,000 events have less than ideal data. The breadth of the database (typically 150,000 UPS) makes it impractical to incorporate data on competitive offerings. We find that regression-style modeling is not adept at incorporating information on the 1,200 subcommodities managed in our pilot stores or the 1,000 manufacturers who supply those stores. Despite these limitations we show the value of using promotion-event data, how tactical forecasts based on these data can directly impact the bottom line of grocery retailers, and how store-by-store forecasts can help retailers with problems of running out of stock or overstocking.

Price promotions are used extensively in marketing for one simple reason—consumers respond. The sales increase for a brand on promotion could be due to consumers accelerating their purchases (i.e., buying earlier than usual and/or buying more than usual) and/or consumers switching their choice from other brands. Purchase acceleration and brand switching relate to the primary demand and secondary demand effects of a promotion. Gupta (1988) captures these effects in a single model and decomposes a brand's total price elasticity into these components. He reports, for the coffee product category, that the main impact of a price promotion is on brand choice (84%), and that there is a smaller impact on purchase incidence (14%) and stockpiling (2%). In other words, the majority of the effect of a promotion is at the secondary level (84%) and there is a relatively small primary demand effect (16%). This paper reports the decomposition of total price elasticity for 173 brands across 13 different product categories. On average, we find that 25% of the elasticity is due to primary demand expansion (i.e., purchase acceleration) and 75% to secondary demand effects or brand switching. Thus, while Gupta's finding that the majority of promotional response stems from brand switching is supported, the average magnitude of the effect appears to be smaller than first thought. More important, there is ample evidence that promotions have a significant primary demand effect. The relative emphasis on purchase acceleration and brand switching varies systematically across categories, and the second goal of the paper is to explain this variation as a function of exogeneous covariates. In doing this, we recognize that promotional response is the consumer's reaction to a price promotion, and therefore develop a framework for understanding variability in promotional response that is based on the consumer's perspective of the benefits from a price promotion. These benefits are posited to be a function of: (i) category-specific factors, (ii) brand-specific factors, and (iii) consumer characteristics. The framework is formalized as a generalized least squares meta-analysis in which the brand's price elasticity is the dependent variable. Several interesting results emerge from this analysis. • Category-specific factors, brand-specific factors, and consumer demographics explain a significant amount of the variance in promotional response for a brand at both the primary and secondary demand levels. • Category-specific factors have greater influence on variability in promotional response and its decomposition than do brand-specific factors. • There are several instances where exogenous variables do not affect total elasticities yet significantly affect individual components of total elasticity. In fact, the lack of a significant relationship between the variables and total elasticity is often due to offsetting effects within two or more of the three behavioral components of elasticity. This is particularly true for brand-specific factors, which typically have no effect on total elasticity, yet have important effects on the individual behaviors. • There is some evidence to suggest that not all promotion-related increases in primary demand are due to forward-buying—in some cases promotions appear to increase consumption. We use these results to illustrate how category- and brand-specific factors work to drive primary and secondary demand elasticities in different directions. In short, this paper offers an empirical generalization of a key finding on promotional response—how elasticities decompose across brand choice, purchase incidence, and stockpiling—and new insights into factors that explain variance in promotional response. These findings are likely to be of interest to researchers who are concerned with theory development and the generalizability of marketing phenomena, and to managers who plan promotion campaigns.