A Customer Segmentation Framework for Targeted Category Coupons in Retail

It is standard procedure to use short-term measures for designing customized marketing campaigns and evaluating their success. We consider the integration of customer lifetime value (CLV) thinking into the category selection problem for targeted coupons. This problem involves two primary considerations: should the coupon be a category-specific one? – and, if so, should short-term or rather long-term objectives govern the choice of category? Our paper is the first to address these questions by developing a data-driven customer segmentation framework based on the dimensions of churn, frequency, and loyalty. We base this framework solely on the transactional data available to every retailer, and we combine statistical tests and $k$-centroid clustering so that each customer is targeted with either a category-specific or a non-category-specific marketing strategy. This personalized targeting strategy enables multi-category retailers to exploit the full potential of customized marketing by optimizing the trade-off between short-term and CLV considerations. Our analysis reveals that this strategy – as compared with the one most commonly used in retail practice – can yield a significantly higher CLV even as it reduces wasteful marketing costs