Overview
This project explores how banks can move beyond generic, one-size-fits-all marketing by genuinely understanding their customers. Using real credit card behavior data, it applies K-Means Clustering, a machine learning technique, to group customers based on how they actually use credit: things like how much of their limit they use, whether they take cash advances, and how often they make purchases.
However, the real value isn't just in the algorithm itself. It's in translating those technical results into something a business can act on. The analysis helps identify customers who might be at risk of default (specifically those maxing out their credit limits), and it creates relatable customer profiles—like the "Cautious Customer" who plays it safe or the "Selective Customer" who's more strategic with spending. From there, the project offers practical recommendations: for example, offering low-interest loans to people who rely heavily on cash advances, or extending higher credit limits to frequent purchasers. The goal is a strategy that benefits both the bank and its customers.