Optimizing Banker Sales Activities

A leading regional bank’s commercial sector. The bank operates in the United States and has a large commercial banking division. This division is responsible for managing high-margin deals and client interactions, which are crucial for the bank's revenue growth. The bank employs thousands of bankers who engage in various activities such as client calls, visits, and lead management. The bank's CRM system logs millions of client interaction entries per year, providing a wealth of data that can be analyzed to optimize banker activities and improve revenue. The bank is focused on leveraging data analytics to enhance its commercial banking operations and drive incremental revenue growth.

The bank had been tracking banker behavior using a CRM software that logged millions of client interaction entries per year across thousands of bankers. This CRM system housed a large amount of valuable data, including banker activities (e.g., client calls, visits, etc.) and details about prospect leads that the bankers were pursuing. The bank was unsure about which outreach tactics and leads were most beneficial for each banker, and since these bankers were pursuing high margin deals, a slight reallocation of their limited time could translate to millions in profit improvement. Previous attempts to analyze the data had resulted in inconclusive results. Moreover, the results were largely skewed by naturally successful bankers; the bank couldn’t measure how much of their success was attributable to their choice in activities and how much was not controllable. The bank needed to understand the true incremental impact of a banker changing activities and reprioritizing leads to identify which bankers should receive specific recommended outreach tactics.

Using APT’s Test & Learn® software to analyze the CRM data, the bank was able to identify naturally occurring changes in banker activity to measure the test vs. control impact of making such changes. “Test” bankers were matched to similar “control” bankers using APT’s patented control matching algorithms, ensuring a clear, reliable read. The software was able to correct for the naturally occurring bias whereby, on average, more successful bankers engaged in certain activities more frequently. By matching test employees only to control employees who had similar historical performance, the bank could measure the true impact of increasing any one activity. The software measured a statistically significant 2.4% increase in revenue from bankers who increased visit frequency. However, the bank found that increasing other activities (for example, calls, appointments) did not drive a significant test vs. control lift in revenue. In addition, bankers who received leads in specific client segments drove an even greater lift of 3.1%. The bank was then able to segment overall results to discover that certain customers should be targeted for additional outreach; specifically, customers in two major client segments, as well as customers with a higher average balance. Additionally, the bank was able to explore which bankers performed better when they increased visits. Management was surprised to find that the results did not differ widely between banker roles. However, bankers with a higher percent of sales accounts in late stages benefitted much more from increased visits. Additionally, lower tenure bankers and bankers located in areas with a high concentration of businesses benefitted more. The bank was able to combine these attributes into a predictive model to determine which employees should be recommended for outreach and lead changes.

• The software measured a statistically significant 2.4% increase in revenue from bankers who increased visit frequency.
• The bank found that increasing other activities (for example, calls, appointments) did not drive a significant test vs. control lift in revenue.
• Bankers who received leads in specific client segments drove an even greater lift of 3.1%.
• The bank was able to segment overall results to discover that certain customers should be targeted for additional outreach; specifically, customers in two major client segments, as well as customers with a higher average balance.
• The bank was able to combine these attributes into a predictive model to determine which employees should be recommended for outreach and lead changes.

• The bank increased annual revenue by an incremental $5MM.
• APT’s Test & Learn software showed that increasing visits to commercial customers drove a 2.4% lift in revenue on average.
• Prioritizing which customers were visited (e.g., leads) led to an even greater revenue increase of 3.1%.