This project focuses on customer segmentation based on credit card behavior using the K-Means algorithm. By analyzing the spending habits, cash advances, and installment purchases of around 9000 active credit card holders, the goal was to define customer groups for targeted marketing strategies. After preprocessing the data by handling missing values, outliers, and standardizing it, the optimal number of clusters was determined to be six using the Elbow method. These clusters were then interpreted and visualized using PCA to assist in decision-making, such as offering special installment options to customers in Cluster 3 or low-commission cash withdrawals for Cluster 2.