Information quality and customer acquisition: A risk management approach for record matching

Consumer credit markets constitute a critical underpinning for the U.S. economy, and outstanding consumer credit accounts for over two-thirds of the U.S. gross domestic product. A viable approach to improve the customer acquisition process and lower the acquisition risks is to implement matching logic software to find entries in a database that belong to the same entity as in the application. In this paper we first employ the use of record matching software and design optimal configuration to minimize financial costs. This analysis shows that while record matching is indeed a useful approach, it fails to capture and correct all errors. Next we explore the option of procuring additional information to help improve accuracy of record matches. We develop models to decide conditions under which it is optimal to exercise the option of information procurement. Our extensive computational analysis explores statistical properties when practical criteria restrict the usage of models only to a sample of data, and explores the calibration of record matching, and when additional information procurement is financially beneficial to credit granting businesses.