The Estimation of Pre- and Postpromotion Dips with Store-Level Scanner Data

One of the mysteries of store-level scanner data modeling is the lack of a dip in sales in the week(s) following a promotion. Researchers expect to find a postpromotion dip because analyses of household scanner panel data indicate that consumers tend to accelerate their purchases in response to a promotion – that is, they buy earlier and/or purchase larger quantities than they would in the absence of a promotion. Thus, one should also find a pronounced dip in store-level sales in the week(s) following a promotion. However, researchers find such dips usually neither at the category nor at the brand level. Several arguments have been proposed for the lack of a postpromotion dip in store-level sales data. These arguments explain why dips may be hidden. Given that dips are difficult to detect by traditional models (and by a visual inspection of the data), we propose models that can account for a multitude of factors which together cause complex preand postpromotion dips. We use three alternative distributed leadand lag structures: an Almon model, an Unrestricted dynamic effects model, and an Exponential decay model. In each model, we include four types of price discounts: without any support, with display-only support, with feature-only support, and with feature and display support. The models are calibrated on store-level scanner data for two product categories: tuna and toilet tissue. We estimate the dip to be between 4 and 25 percent of the current sales effect, which is consistent with household-level studies.