Retail-Customer-Segmentation

Analysis of UK Online Retail data using RFM Segmentation and Market Basket Analysis to improve retention.

🛒 E-Commerce Customer Segmentation & Market Basket Analysis

📌 Project Overview

Analyzed a dataset of 500,000+ transactions from a UK-based online retailer. The goal was to transform raw sales data into actionable business intelligence using RFM Analysis and Association Rule Mining.

🔧 Techniques Used

• Data Cleaning: Handled missing values, removed duplicate entries, and filtered cancelled transactions.
• Feature Engineering: Created TotalSales and RFM metrics (Recency, Frequency, Monetary).
• Customer Segmentation: Segmented users into "Champions", "Loyal", and "At Risk" groups using Quantile Scoring.
• Market Basket Analysis: Used the Apriori Algorithm to find products frequently bought together.

📊 Key Results & Insights

1. Customer Segmentation (RFM)

The analysis revealed that the customer base follows the Pareto Principle (80/20 rule).

• Champions (11%): High spenders who buy recently.
• At Risk (44%): Large portion of customers who haven't purchased in >150 days.
• Action: Recommended a "Win-Back" email campaign targeting the "At Risk" group.

2. Market Basket Analysis

We identified strong product associations (Lift > 10) to drive cross-selling strategies.

• Key Discovery: Customers purchasing specific color variants (e.g., Green Alarm Clock) are highly likely to purchase the alternative color (Red Alarm Clock).
• Action: Implement "Bundle & Save" offers on product pages.

🚀 How to Run the Code

1. Clone the repo:

   git clone [https://github.com/Razorface1919/Retail-Analytics.git](https://github.com/Razorface1919/Retail-Analytics.git)