Published Paper Details:

Understanding customer behavior is increasingly important for retaining customers and increasing profitability in the intensely competitive environment of e-commerce. This study aims to design a hybrid analytical model that combines the RFM analysis technique with machine learning algorithms in the processes of customer segmentation and churn analysis. The dataset used in this study was sourced from an e-commerce company in the UK and consists of real transactional data with detailed invoice information, several purchase quantities, and customer IDs. Following the cleaning process of this data, RFM values were calculated for each customer based on their purchasing behavior, while algorithms such as K-Means were applied to segment the customers into meaningful groups. Analyzing these clusters further will identify high-value, loyal, and at-risk customers, offering very useful information to help focus effective marketing strategies. Furthermore, the analytical framework identifies churn-prone customers based on transaction inactivity patterns about recent transaction habits. Thus, this paper shows that the integration of RFM metrics with machine learning algorithm analyses allows for a robust and data-driven analytical approach toward improving customer segmentation, churn analysis, and effective marketing strategy formulation for e-commerce businesses.