Retail banks operate in a competitive environment where survival depends on increasingly sophisticated ways to predict consumer behavior and enhance customer experience while maintaining tight cost controls. The majority of our client’s retail customers were in the Middle East but a sizeable minority were spread across several European countries. Overall, the bank had low levels of e-banking usage among its retail base, but due to cultural differences, there were considerable disparities in rates of digital banking (online and mobile) adoption among different regions. However, those who used digital banking reported higher levels of customer satisfaction than those who didn’t. Simultaneously, digital channel delivery provided significant cost savings to the bank.
The bank embarked on a set of major business process re-engineering initiatives with a key goal being increased retail e-banking adoption. They wanted quantifiable insights to guide their marketing strategy and forecast the likely uplift in digital adoption.
Using our client’s rich RFM and demographic data for nearly 200,000 of their retail customers across more than 15 countries, we designed a comprehensive data mining, segmentation, and behavior scoring model. Additional data were captured through surveys and direct interviews. Data were analyzed using SOM-network visualizations and clustering models with customized distance metrics. We identified branch-specific factors negatively impacting adoption and built target group profiles using behavior-predictive and customer lifetime value (CLV) attributes. One key finding was that the segments with the lowest CLVs had the lowest e-banking adoption rates. Because the average cost to the bank for an in-branch transaction was more that 400% greater than an e-banking transaction, these customers were disproportionately responsible for adverse cost impact.