E-Commerce Customer Segmentation: Driving Revenue Growth with RFM & Machine Learning

Description:

Executed an end-to-end customer segmentation initiative for a UK-based online retailer, analyzing 540,000+ transactions to uncover strategic growth opportunities. By combining RFM analysis and K-Means clustering, I transformed raw transaction data into actionable customer segments, directly informing marketing strategy and inventory management.

Key Responsibilities & Impact:

• Identified & Protected the Core Business: Analysis revealed that the top 20% of customers (the "Premium" segment) generated 80% of all revenue. This insight directly shaped a new, high-priority customer retention program designed to protect the company's most valuable asset and reduce churn risk.

• Unlocked a 40% Reactivation Opportunity: Segmentation exposed a large, "Inactive" cluster comprising 40% of the customer base. This quantified a significant dormant market, leading to the development of a targeted win-back campaign to recover lost revenue at a low customer acquisition cost.

• Optimized Inventory & Marketing Spend for a 45% Revenue Event: Pinpointed a predictable 45% revenue surge in November and December. This enabled data-backed inventory planning to prevent stockouts and maximize sales during the critical holiday period, ensuring capital was not tied up in slow-moving products.

• Drove International Expansion Strategy: Discovered that the German market had a 25% higher Average Order Value (AOV) than the UK. This "so what" directly informed the decision to re-allocate international marketing budget towards Germany, focusing on high-value customer acquisition to improve marketing ROI.

How I Delivered Rigor & Insight: *Executed end-to-end analysis of 540,000+ transactions for 37,000 customers using Python (Pandas, Scikit-learn). Combined RFM analysis with a K-Means clustering model (validated with the Elbow Method) to create a robust, actionable 4-tier customer segmentation.*