Developing an EMI Financing ML model
For a Large Asian Bank
CLIENT & PROBLEM STATEMENT
- The client is a large Asian bank with diversified asset and liability products portfolio.
- The bank wanted to enhance its offerings by introducing an EMI product to its qualified card customers. The challenge lay in identifying which of these customers were most likely to avail themselves of the EMI product, thereby optimizing marketing efforts and increasing product uptake
APPROACH
- To achieve the client’s objective of predicting customer propensity for the EMI financing product, we employed binary classification modeling.
- This involved leveraging advanced algorithms such as Random Forest and XGBoost to analyze customer data and predict the likelihood of EMI product adoption.
- We segmented customers based on their revolving behavior and decile ranking, enabling targeted marketing strategies tailored to each segment’s preferences and propensity for the EMI product.
SOLUTION & OUTPUT
- We accurately predicted customer behavior, enabling the bank to streamline marketing efforts and maximize product uptake.
- Our model was able to accurately predict the customer who eventually took the EMI financing.
