Welcome to the Customer Segmentation Using RFM Analysis Project! 🎉 This project focuses on analyzing customer purchasing behavior through the RFM (Recency, Frequency, Monetary) model, uncovering patterns that help businesses understand their customers better. 🛍️💡 By segmenting customers based on their buying habits, we aim to empower decision-makers with actionable insights for personalized marketing strategies and customer retention.✨
Retail and e-commerce businesses generate massive amounts of transaction data daily 🏬. Analyzing this data with RFM provides powerful insights into who your best customers are, who needs attention, and who may be slipping away. In this project, we focused on:
- ✨ Measuring Recency, Frequency, and Monetary value for each customer
- ✨ Assigning RFM scores to categorize customers 📊
- ✨ Creating customer segments (Champions, Loyal, At Risk, Hibernating, etc.)
- ✨ Building multiple visualizations (heatmaps, scatter plots, bar charts, etc.)
- ✨ Suggesting targeted marketing strategies for each segment 🎯 This project transforms raw sales transactions into a customer-centric view of business, enabling smarter decisions and long-term growth. 🌱
- 🔹 Clean and preprocess the sales dataset for accurate analysis
- 🔹 Calculate RFM metrics for each customer 👥
- 🔹 Assign RFM scores and create meaningful segments 📊
- 🔹 Build visualizations to showcase patterns & clusters 🎨
- 🔹 Highlight customer behavior insights (best vs. risk customers)
- 🔹 Recommend marketing actions tailored to each segment 💬
- Language: Python 🐍
- Libraries: Pandas, NumPy, Seaborn, Matplotlib, Datetime
- Analysis Methods: Descriptive Analysis | RFM Segmentation | Customer Analytics
- Visualizations: Heatmaps 🔥 | Line Charts 📈 | Bar Charts 📊 | Scatter Plots 🎯 | Boxplots 📦 | Histograms 📉 | KPI-style summaries
The dataset contains transaction-level records of an online retail store, with the following key fields:
- 📅 Invoice Date – Date of purchase
- 🧾 Invoice Number – Transaction ID
- 👤 Customer ID – Unique customer identifier
- 📦 Quantity & Unit Price – Purchase details
- 💲 Total Sales – Derived from Quantity × Price
- Imported the dataset into Python (Pandas)
- Removed missing values & cancelled invoices
- Created new fields (e.g., TotalPrice)
- Recency → Days since last purchase
- Frequency → Number of purchases made
- Monetary → Total spending of each customer
Assigned R, F, M scores (1–5) using quantiles Created a combined RFM Score (e.g., 555 = best customer) Classified customers into segments such as:
- ⭐ Champions
- 💎 Loyal Customers
- 🎯 Potential Loyalists
⚠️ At Risk- 💤 Hibernating
- Created 10+ different visualizations, including:
- Recency, Frequency, Monetary Distributions 📉
- RFM Correlation Heatmap 🔥
- Segment-Wise Customer Count 📊
- Avg. Monetary Value per Segment 💲
- Scatter Plots (Recency vs Frequency, Monetary vs Frequency) 🎯
- Boxplots of Monetary by Segment 📦
- Line Chart of Avg. RFM values per Segment 📈
Some key findings include:
- ✔️ Champions are the most valuable customers → Reward with loyalty programs & VIP offers 🎁
- ✔️ Loyal Customers shop frequently → Upsell & cross-sell relevant products 🔄
- ✔️ Potential Loyalists need nurturing → Send personalized promotions 💬
- ✔️ At Risk Customers are slipping → Win-back campaigns with discounts
⚠️ - ✔️ Hibernating Customers are inactive → Re-engage via awareness campaigns 📢
- 📌 Cleaned Dataset → RFM_Cleaned.csv
- 📌 Python Notebook → RFM_Analysis.ipynb
- 📌 Segmentation Report → Customer_Segmentation_Report.docx / PDF
- 📌 Visualizations → Heatmaps, Charts, and Segment Analysis
This project demonstrates how RFM Analysis transforms raw retail data into customer-focused insights. By applying segmentation techniques and visualizing patterns, we can identify top customers, prevent churn, and design smarter marketing strategies. 🌟 The result is a data-driven customer segmentation framework that helps businesses build stronger relationships, improve engagement, and maximize profitability. 💡
💼 Portfolio: https://my-dashboard-canvas.lovable.app/
📧 Email: umerabdullah048@gmail.com
