Analytics Case Study

Customer Retention, Segmentation, and Churn Prediction

A complete online retail case study spanning retention analysis, RFM segmentation, and churn prediction from data cleaning through business recommendations.

Dataset

Online Retail II

Output

Retention + Churn

Narrative

Hiring-ready

Problem Statement

E-commerce teams need a clearer way to identify high-value customers, churn-risk groups, and the retention actions worth prioritizing.

Key Finding

Purchase frequency and recency sharply separate high-value customers from high-risk churn segments, which makes retention prioritization more actionable.

Project Framing

This project grows out of an existing online retail analysis script. The core questions are which customers are most valuable, which are most likely to churn, and which actions deserve business focus first.

Instead of showing only a notebook, the portfolio version turns the problem definition, analysis chain, and business actions into a coherent story.

Method

The workflow covers cleaning, cohort retention analysis, RFM segmentation, churn labeling, and a logistic-regression baseline.

This structure demonstrates both core analytical thinking and the ability to move from descriptive analysis into simple modeling.

Why It Matters

Hiring teams are rarely only looking for code. They want to see whether you can break down a business problem, make judgments, and recommend actions.

That is the purpose of this case-study page: to make the project read more like decision support than a classroom exercise.