A decision-support tool for recommending promising categories for targeted promotions

Targeting customer segments with tailored promotional activities is widely recognized as a potential successful strategy to gain more of existing customers’ discretionary spending. Most previous research focuses on the selection of promising customers for targeting and typically ignores the decision of which specific product category to select for inclusion in such targeted promotions. We present an innovative two-stage decision-support tool that assists managers in selecting promising categories from large assortments for inclusion in targeted promotions. To empirically test the performance of the derived segment-level category promotions, we implemented and validated our decision-support tool in cooperation with a major home-delivery grocery retailer using a large-scale controlled field experiment. Results point to a very successful campaign: our redemption rates are among the highest reported in both academic and managerial literature. What is more, targeted promotions clearly outperform non-targeted ones in terms of coupon redemption. Yet, not all targeted promotions change purchase behavior successfully. We show that customers who use a coupon where the most influential “draw" category was promoted engage in true cross-category purchases, resulting in an average extra boost in overall shopping basket size of about 10 percentage points (in €-amount and number of SKUs) vis-à-vis a control group. We therefore show that our procedure is an important decision-support tool that can help managers with making smart decisions regarding which category to promote for which customer.status: publishe