Getting ahead of attrition risk.

• Author: Clapp, Bruce
• Position: Customer Retention

Practices such as customer satisfaction indexing, mystery shopping, transactional surveys, etc. have been the mainstay of attrition management for years. However, these only provide a snapshot of time and in most instances, it is of a time and activity that has passed! The reality is that most customers behave in predictable patterns before closing accounts and before closing relationships. The key is gaining the crucial customer insight to allow you to understand those behaviors, model the behaviors, and apply predictive analysis to your remaining customer portfolio. The key is getting ahead of the attrition curve!

You can invest in a CRM system, MCIF system, employ regression analysis, and do many other actions to help predict behavior. All of these are excellent and I fully support their review for appropriateness at your institution. However, in the interim, you can begin on a smaller scale and employ more simplistic measurements. For instance, I believe the true "loyalty indicator" for a financial institution is the ownership of the following five account categories: (1) checking; (2) savings; (3) loan; (4) access (debit card, ATM card, online banking, etc.); (5) investment (CD, IRA or traditional investment services).

To me, the presence of these account categories translates into an active household that is not only using your services to a high extent, but is loyal. Measure who and how many have these relationships. Then measure those that do not have the individual categories. You then simply start marketing the "missing" categories to those customers that do not have all five categories. Basic. Simple. Effective. If nothing else, a place to start!

To get more sophisticated, you need to employ an analytical process to "score" your accounts and households for attrition risk. Many firms can assist you with a model that fits your bank, your customer base...