Case Study
Client
Major Bankcard Issuer
Project
Reduce Attrition
Background
DataCo has built many anti-attrition models for the credit card industry. Major bankcard issuers have used these models as a basis for retention marketing programs that have reduced attrition by as much as 40%.
Objective
Build a statistical model that proactively detects characteristics of people who are most likely to close his/her account within 90 days. Validate that the model rank-orders the customer file based on actual attrition rates.
Execution
DataCo scored the issuer's file as of March to determine each customer's likelihood to close the account within 90 days. Attrition rates were projected at each 10% increment of the file (i.e. at each decile). Those scoring in decile 1 were most likely to close the account. After 90 days (June), the issuer provided a then-current customer file, from which actual attrition rates were calculated for each decile.
Results
When the actual attrition rates were calculated, 95% of the customers who had closed the accountd, scored in deciles 1 or 2. That means that anti-attrition efforts could be focused on 20% of the customer base, to almost entirely address the attrition risk.
|
