E-Commerce Customer Segmentation & Recommendation System

E-Commerce Customer Segmentation & Recommendation System

E-commerce businesses often face challenges in understanding their customer base, tailoring marketing strategies, and optimizing product offerings. By the Implementation of an ML clustering model for customer segmentation using Recency, Frequency, and Monetary (RFM) analysis. Enhanced by a product recommendation system (RecSys) using Market Basket Analysis (MBA). This project aims at providing valuable insights that can enhance marketing effectiveness, increase customer retention, and boost overall revenue through tailored and targeted (per Segment) Upselling and Cross-selling of listed products.

The project identified significant cost-saving opportunities by strategically limiting investments in the lower-value segments. This resulted in a total cost savings of R$1,289,345.17, driven mainly by reduced marketing spend on the Price Sensitive (R$327,109.53) and Hibernating (R$614,030.65) segments. The dual approach of increasing revenue while cutting costs resulted in an overall financial impact of R$4,417,666.02, underscoring the effectiveness of the ML models in enhancing business profitability.