Case Study Overview
Segmenting a retail client's active customer base to target specific cohorts and design optimized, personalized promotional campaigns.
Problem Statement
Group customers using transaction behavior to increase promotional email response rates and customer lifetime value (LTV).
Analytical Focus
Developing RFM (Recency, Frequency, Monetary) analytics models in Python and publishing cluster dashboards in Tableau.
The Analytical Approach
I pulled active client lists, computed RFM scores, and ran K-Means clustering algorithms in a Jupyter notebook to define distinct shopper cohorts (e.g. VIP Shoppers, Churn-Risk Bargain Buyers, Steady Shoppers).
The data clusters were mapped into Tableau to show cluster trends, purchasing patterns, and preferred product categories, allowing marketing to directly retrieve contact lists.
Results & Business Value
Key Metric Improvement
18% Lift in Campaign Response Rates
- Enabled marketing teams to send custom product offerings tailored to specific clusters.
- Highlighted an active cohort representing 12% of shoppers that generates 45% of monetary sales.
- Automated pipeline to refresh customer clusters monthly inside client databases.