Customer Segmentation Using RFM

Overview

This repository demonstrates the implementation of Customer Segmentation using the RFM (Recency, Frequency, Monetary) model and K-Means clustering. The analysis focuses on segmenting customers of an online retail store based on their purchasing behaviour, providing actionable insights to enhance customer engagement and marketing strategies.

---

Table of Contents

• Overview
• About RFM Analysis
• Features of the Project
• Dataset
• Project Workflow
• Insights and Findings

---

About RFM Analysis

RFM is a technique used in marketing to evaluate and segment customers based on their purchasing behavior:

• Recency (R): How recently a customer made a purchase.
• Frequency (F): How often a customer makes purchases.
• Monetary (M): How much money a customer spends.

By assigning scores to each customer for R, F, and M metrics, we can create meaningful customer segments and design targeted strategies for each group.

---

Features of the Project

1. Data Cleaning:

   • Removed duplicates and irrelevant data.
   • Handled missing values.
   • Filtered valid transactions (excluding cancellations and refunds).

2. RFM Feature Engineering:

   • Computed Recency, Frequency, and Monetary values for each customer.
   • Scored customers based on their RFM metrics.

3. K-Means Clustering:

   • Scaled data using StandardScaler for normalization.
   • Used the Elbow Method to determine the optimal number of clusters.
   • Applied K-Means clustering to segment customers.

4. Data Visualization:

   • Visualized RFM metrics and clusters using scatter plots, heatmaps, and bar charts.

5. Insights Generation:

   • Identified key customer segments such as "High-Value Customers" and "At-Risk Customers."

---

Dataset

The dataset is from an online retail store, containing transactional data, including:

• Customer ID
• Invoice Date
• Invoice Number
• Transaction Amount

(Note: Actual data files are available in the data/ directory. Ensure the dataset is clean before proceeding.)

---

Project Workflow

1. Data Cleaning:

   • Handle missing values and filter relevant data.

2. RFM Analysis:

   • Compute Recency, Frequency, and Monetary values.
   • Score customers and create RFM segments.

3. Scaling and Clustering:

   • Scale the RFM data using StandardScaler.
   • Determine the optimal number of clusters using the Elbow Method.
   • Perform K-Means clustering.

4. Visualization and Insights:

   • Visualize clusters and generate actionable insights.

---

3. Follow the Jupyter notebooks in the notebooks/ directory to reproduce the analysis:
   • 01_data_cleaning.ipynb
   • 02_rfm_analysis.ipynb
   • 03_clustering_kmeans.ipynb
   • 04_visualization.ipynb

---

Insights and Findings

Through the analysis, the following customer segments were identified:

1. High-Value Customers: Recently purchased, frequent buyers, and high spenders.
2. Loyal Customers: Frequent buyers with moderate spending.
3. At-Risk Customers: Customers who haven't purchased recently and have low spending.
4. Potential Loyalists: Customers who recently purchased and spent a significant amount but have low frequency.

These insights can be used to:

• Target marketing campaigns.
• Design loyalty programs for high-value customers.
• Re-engage at-risk customers.

---