Big Data, Data Mining, and Machine Learning: Value Creation for Business Leaders and Practitioners

CHAPTER 11 Case Study of a Large U.S.-Based Financial Services Company

The U.S. banking industry, because of government regulations and its utilization by almost all citizens, has by necessity lots of data. Some of the data is kept to comply with regulatory requirements, and some is kept to help the institution provide quality service and access to beneficial financial instruments to its customers. To sift efficiently and effectively through all the data being kept and distill useful information from that data poses a challenge for banks.

The bank under consideration serves millions and millions of customers. It frequently engages in marketing campaigns to offer both current and potential customers new relevant products and services. This case study is a marketing campaign to identify likely people to respond to an offer for a new credit card. This scenario is repeated frequently at this and many other banks and also applies to many other companies that offer products to customers. The ability to effectively predict which potential customers will respond to a marketing offer is very valuable for several reasons. The first is that it builds customer confidence by not sending junk mail or spam. The next reason is it allows the bank to reduce its expenses by not printing and mailing unwanted materials or hiring call center staff to make unwanted phone calls. The last reason is that getting customers signed up for the services they want generates revenue for the bank.

Credit cards ...