Customer Segmentation using RFM Analysis

Overview

This project applies RFM (Recency, Frequency, Monetary) Analysis to segment customers based on their purchase behavior. It's a smart way to identify high-value customers, inactive ones, and plan targeted marketing strategies.

---

Dataset

• Source: Simulated customer data (1,000 records)
• Fields Included:
  • Customer ID, Name, Email, Subscription Date
  • Simulated:
    • LastPurchaseDate (randomized from Jan 2023 – Apr 2024)
    • TotalPurchases (Frequency)
    • TotalSpend (Monetary)

---

Technologies Used

• Python
• Pandas, NumPy
• (Optional) Matplotlib, Seaborn
• Jupyter Notebook / Kaggle

---

Key Concepts

• Recency: How recently a customer purchased
• Frequency: How often they purchase
• Monetary: How much they spend

---

Steps Performed

1. Data Loading

   • Loaded and explored the dataset using Pandas

2. Feature Simulation

   • Generated realistic values for LastPurchaseDate, TotalPurchases, and TotalSpend using NumPy

3. RFM Metric Calculation

   • Calculated Recency as the days since the last purchase
   • Used Frequency and Monetary values as simulated

4. Scoring

   • Used pd.qcut() to assign scores (1–5) for each R, F, M metric:
     • Lower Recency = higher score
     • Higher Frequency = higher score
     • Higher Monetary = higher score

5. RFM Score Combination

   • Combined R_Score, F_Score, and M_Score to generate a single RFM score like 555, 431, etc.

6. Customer Segmentation

   • Based on RFM_Score, segmented customers as:
     • Champions, Loyal Customers, At Risk, Need Attention, etc.

---

Sample Output

Customer ID	Recency	Frequency	Monetary	R_Score	F_Score	M_Score	RFM_Score	Segment
1001	8	9	4000	5	5	5	555	Champion
1040	88	2	300	1	2	2	122	At Risk

---

Why This Project Matters

• Real-world business application
• Teaches scoring and segmentation logic
• Builds strong data manipulation and analysis skills
• Easily extendable to dashboards (Power BI, Tableau)

---

Future Enhancements

• Create an interactive dashboard in Power BI
• Use real-world transactional data
• Apply clustering or ML for deeper analysis

---