Geodemographic drivers of store-level demand and marketing mix sensitivities

Abstract

We study changes in marketing mix sensitivities in the United States following the Great Recession using weekly scanner data from 2010 representing 16,294 stores (9,900 food stores, 3,185 mass merchandisers, and 3,209 drug stores). We estimate baseline category sales and the dependence of category demand on marketing activities (price, display and feature). We then relate the store-level demand to geodemographic characteristics of stores’ zip codes, accounting for local competitive intensity faced by stores. We find that as the percentage of college educated consumers in a store’s zip code increases, category sales increase and price sensitivity of the store decreases (regardless of store format). Median income of the store’s zip code explains most of the observed variation in estimated price sensitivities across stores. We find that a majority of stores are underpricing cereal, coffee and cola, i.e., pricing lower than what the optimal mark-ups implied by their price elasticities would suggest.