Leveraging loyalty programs using competitor based targeting

Loyalty programs (LPs) are widely used by firms but not well understood. These provide discounts and perks to loyal customers costly administer, produce uncertain changes in spending patterns. We use a large detailed dataset on customer shopping behavior at one of the largest U.S. retailers before after joining loyalty program evaluate how changes. combine this with spatial data store locations, including locations competing firms. find significant associated LP amount heterogeneity across customers. that location relative competitors is factor most strongly increases following LP, suggesting LP’s quantity work primarily through business stealing other demand expansion. next estimate model what variables determine will change LP. high-dimensional relationships between customers, focal firm’s stores, stores as customers’ historical This test whether past sales reflecting customer’s vertical value firm or horizontal vulnerability more important determinants post-LP increases. show LASSO regularization estimated complex effective than models using simpler models. Finally, we can competitor substantially increase performance spatially driven segmentation.