A discrete/continuous model of fruit promotion, advertising, and response segmentation

Commodity marketing managers are demanding an increasingly detailed description of their customers in order to make more efficient use of their marketing budgets. This study uses a purchase-occasion, household panel data set to investigate consumers’ discrete choice among varieties of apples and their continuous decision of how much to buy. We segment households based on their elasticities of response to several marketing variables: price, promotion, and advertising using a k-means clustering algorithm. Describing each of these segments according to their demographic attributes provides the type of information required to target particular groups of similar consumers, which in turn allows for a much more effective and efficient use of scarce marketing dollars.