Acquisition of new customers involves both opportunity and risk, and it is important for firms to predict and manage the risks involved in customer acquisition. Despite its importance, the management of customer acquisition risk has not been the subject of much academic research. This paper develops a framework for firms to manage customer acquisition risk using co-operative databases. We illustrate this framework in the context of the optimal selection of customers for direct mail with a ‘buy now, pay later’ payment option when the acquisition risk manifests as bad debt risk. Using data from a large scale direct marketing campaign, we show that our empirical model that incorporates bad debt risk substantially outperforms suboptimal targeting schemes that overlook bad debt risk. We also demonstrate how alleviating bad debt risk is one beneficial outcome of a fairly recent trend in database marketing, namely the emergence of co-operative databases. © 2014 Direct Marketing Educational Foundation, Inc., dba Marketing EDGE.