How to Reduce Attrition and Retain Customers in the Age of Digital Banking—Part 2

This is part two of a post looking at the problem of financial institutions losing profitable customers and what can be done about it. Read part 1 here.

By Devon Kinkead, CEO and Founder, Micronotes

Here’s how Micronotes helps financial institutions retain customers at risk of attrition: We consume 6 months of historical data from the banking institution and build a list of customers who were lost during that time period. Those lost customers now become the predicted variable in our propensity scoring procedure. We filter out unprofitable customers and assign a threshold propensity score value, 0.5 for example, then select all customers who are profitable andat high risk of attrition. We take this list of customers who haven’t yet left and push those to the financial institution for outreach via email or integration with the banking provider’s CRM system. 

In one example, the background attrition rate of profitable customers was 0.5 percent per month, whereas the actual attrition rate of profitable customers that Micronotes flagged for outreach was 37.5 percent, so, the models are accurate and actionable. 

An outreach script to a customer who is profitable, at risk of attrition, and doesn’t have a direct deposit relationship with the banking provider might look like this:

“Thank you for banking with us. We’re reaching out in hopes of deepening our relationship with you. We’d like to be your primary financial institution and would like to offer you $400 to make (3) direct deposits of $500+ with us, plus $50 for each additional free eService you adopt and use—such as mobile banking, bill-pay, eStatements, etc. How does that sound?”  

The financial institution won’t be able to claw-back all profitable customers at risk of attrition but, with solid outreach execution and strong incentives to stay, the banking provider will win back some of those profitable customers.

There are at least a dozen factors that determine attrition risk (e.g., deposit balance, ACH deposit amount, creditworthiness, checking balance, checking activity, etc.) which our machine-learning system is superb at identifying using existing banking data. Actionable analytics and excellent personal outreach is the key to retaining profitable customers who are at risk of leaving.