RFM Customer Segmentation

            Key Achievement: Leveraged K-Means clustering to identify high-value "Champion" segments, providing a roadmap for 20% more efficient ad spend.
        

Python Scikit-Learn K-Means Clustering

Strategy & Analysis

This project involved segmenting 10,000+ customers based on Recency, Frequency, and Monetary (RFM) scores to optimize marketing ROI.

The "What": Technical Methodology

The technical core of this project utilized **K-Means Clustering**, an unsupervised machine learning algorithm. After calculating individual RFM scores from raw transaction logs, I standardized the data to ensure equal weighting. I then used the **Elbow Method** and **Silhouette Analysis** to determine the optimal number of clusters. This resulted in five distinct personas—ranging from "Loyal Champions" to "Hibernating Customers"—allowing for precision targeting in promotional campaigns.

The "Why": Data Science Impact

This project bridges the gap between raw behavioral data and business strategy. In Data Science, our goal is to drive actionable insights. By moving beyond aggregate averages and looking at specific user clusters, I empower a business to personalize their engagement. This "Zero-Discrepancy" approach to customer behavior ensures that marketing resources are allocated to the segments with the highest potential Lifetime Value (LTV), directly improving the bottom line.

← Back to Portfolio Dashboard