Decomposing the sales promotion bump accounting for cross-category effects

Extant research on the decomposition of unit sales bumps due to price promotions considers these effects only within a single product category. This article introduces a framework that accommodates specific cross-category effects. Empirical results based on daily data measured at the item/SKU level show that the effects of promotions on sales in other categories are modest. Between-category complementary effects (20%) are, on average, substantially larger than between-category substitution effects (11%). Hence, a promotion of an item has an average net spin-off effect of (20 - 11 =) 9% of its own effect. The number of significant cross-category effects is low, which means that we expect that, most of the time, it is sufficient to look at within-category effects only. We also find within-category complementary effects, which implies that competitive items within the category may benefit from a promotion. We find small stockpiling effects (6%), modest cross-item effects (22%), and substantial category-expansion effects (72%). The cross-item effects are the result of cross-item substitution effects within the category (26%) and within-category complementary effects (4%). Approximately 15% (= 11% / 72%) of the category-expansion effect is due to between-category substitution effects of dependent categories.

A comparison of sales response predictions from demand models applied to store-level versus panel data

In order to generate sales promotion response predictions, marketing analysts estimate demand models using either disaggregated (consumer-level) or aggregated (store-level) scanner data. Comparison of predictions from these demand models is complicated by the fact that models may accommodate different forms of consumer heterogeneity depending on the level of data aggregation. This study shows via simulation that demand models with various heterogeneity specifications do not produce more accurate sales response predictions than a homogeneous demand model applied to store-level data, with one major exception: a random coefficients model designed to capture within-store heterogeneity using store-level data produced significantly more accurate sales response predictions (as well as better fit) compared to other model specifications. An empirical application to the paper towel product category adds additional insights. This article has supplementary material online.