It’s hard work acquiring new customers. The last thing anyone wants is to see those customers leave soon after. Everyone loses customers, but not everyone loses the same percentage of customers. Some Financial Institutions (FIs) are able to retain a very high percentage of their customers year over year, while others struggle with retention. Understanding your attrition numbers, what type of customers you’re keeping and who you’re losing, can make a big difference on the balance sheet. Equally important is developing a formal strategy for retaining more customers.
This starts with understanding which of your customers are most important to the FI’s goals (cross-sell ratio, transactional account users, etc.) and profitability.
Using a model can help to understand which patterns in a customer’s behavior coincide with a higher potential for departure. Through identifying these patterns and the importance of each of these key variables, other customers with similar trends can be identified. Attrition trends can often be seen in the slow closure of accounts, reduction of account balances, reduction in the use of debit cards and online banking services, or in the termination of direct deposit. But it’s not always something detectable by the eye or only seen in these variables. Models are able to assess the importance of all data available.
Not only is it important to pinpoint the people that appear likely to leave, but this information is only useful if it is translated into a plan that nurtures and retains customers that provide long-term profitability and growth.
After identifying these customers with a higher probability of departure and rank-ordering them to direct retention efforts towards the most important to retain, it’s important to validate that your actions had an effect. Comparing the output of the model centering around the potential attrition customers and measuring that against a report of those who truly left provides backend validation of the model and insights into how effective your retention efforts were at mitigating attrition.
As with all things data analytics, understanding which pieces of information (or which metrics) are meaningful from a sea of analysis isn’t always easy. What’s harder is being the person trying to understand or interpret the data when key pieces haven’t been isolated from other less meaningful data. Our computers and models crunch the data and output results, but presenting those results to key decision-makers and those who need to understand the data is critical.
Dashboards have made this review of data far easier in recent years. Using tools that present visuals in a cleaner, easier-to-understand manner is a key element to enabling decision makers to focus their energy on the metrics that matter while cutting out the clutter. Many of these tools also allow for viewing real-time information by being able to connect into database programs, enabling data to be refreshed on the fly.