Predicting Personal Loan Uptake Using Machine Learning

AllLife Bank, a growing U.S. bank, wants to expand its loan customer base by targeting its existing deposit customers. While the bank has a large base of liability customers (those who hold savings or checking accounts), only a small percentage also take personal loans. The business objective is to convert more of these depositors into loan customers—without losing them as depositors.

To support this goal, I built a machine learning model to predict which liability customers are most likely to accept a personal loan offer. The project involved:

• Exploratory data analysis (EDA) to uncover patterns in demographics, income, and product usage
• Modeling using Logistic Regression and Decision Tree classifiers
• Evaluation using classification metrics and ROC AUC score
• Business recommendations to improve marketing efficiency and ROI

This end-to-end project combines data science, business insight, and marketing strategy—providing a practical example of how ML can drive real business impact in the financial services domain.