📉 Customer Retention & Churn Analytics — Power BI + SQL Project

An end-to-end Customer Analytics project using SQL, Power BI, Power Query, and DAX to measure retention, churn, repeat customers, customer lifetime value patterns, and overall customer behavior.

This project demonstrates complete workflow skills:
data cleaning → SQL analysis → data modeling → DAX → dashboard design → churn insights.

---

🔧 Tech Stack

• SQL
• Power BI
• Power Query
• DAX
• Excel
• Data Modeling
• GitHub Documentation

---

🚀 Project Overview

Customer retention is critical for growth.
This project analyzes:

• Which customers are active vs churned
• Repeat purchase behavior
• Average spend patterns
• Customer lifetime behavior
• Regional churn distribution
• High-risk customer segments

The dashboard highlights churn drivers and retention opportunities.

---

🧩 Business Problem

Companies lack clear visibility into:

• Why customers leave (churn)
• Spend pattern differences between retained vs churned users
• Repeat order behavior
• Region-wise churn
• Customer lifetime value patterns

This dashboard solves those gaps with interactive insights.

---

🎯 Project Goals

• Identify active vs churned customers
• Calculate repeat rate & retention rate
• Measure average spend & average orders
• Analyze churn by region, segment, and order frequency
• Create a Customer Status Model using SQL & DAX

---

🗂 Folder Structure

Customer_Retention_Churn_Analytics/
│
├── Dataset/
│     └── customer_retention_dataset.csv
│
├── SQL/
│     └── customer_retention_queries.sql
│
├── PowerBI_Files/
│     └── Customer_Retention_Churn_Analytics.pbix
│
├── Images/
│     ├── dashboard_page_1.png
│     └── dashboard_page_2.png
│
└── README.md

---

🛠 SQL Queries Used

-- 1. Active customers (purchased in last 90 days)
SELECT customer_id, last_purchase_date
FROM customer_data
WHERE DATEDIFF(day, last_purchase_date, '2024-01-01') <= 90;

-- 2. Churned customers (inactive > 90 days)
SELECT customer_id, last_purchase_date
FROM customer_data
WHERE DATEDIFF(day, last_purchase_date, '2024-01-01') > 90;

-- 3. Repeat customers
SELECT customer_id
FROM customer_data
WHERE total_orders > 1;

-- 4. Average orders per customer
SELECT AVG(total_orders) AS avg_orders
FROM customer_data;

-- 5. Average spend per customer
SELECT AVG(total_spent) AS avg_spend
FROM customer_data;

-- 6. Region-level churn distribution
SELECT region,
       SUM(CASE WHEN churn_flag = 1 THEN 1 END) AS churned,
       SUM(CASE WHEN churn_flag = 0 THEN 1 END) AS active
FROM customer_data
GROUP BY region;

---

🧮 DAX Measures

Active Customers =
CALCULATE(COUNT(customer_data[customer_id]), customer_data[churn_flag] = 0)

Churned Customers =
CALCULATE(COUNT(customer_data[customer_id]), customer_data[churn_flag] = 1)

Repeat Customers =
CALCULATE(COUNT(customer_data[customer_id]), customer_data[total_orders] > 1)

Repeat Rate % =
DIVIDE([Repeat Customers], [Active Customers], 0)

Avg Orders =
AVERAGE(customer_data[total_orders])

Avg Spend =
AVERAGE(customer_data[total_spent])

---

📊 Dashboard Pages

📌 Page 1 — Customer Status Overview

• Total Customers
• Active Customers
• Churned Customers
• Repeat Customers
• Repeat Rate %
• Avg Orders
• Avg Spend
• Churn by Region (Bar Chart)
• Churn Trend (Line Chart)

📌 Page 2 — Customer Details & Behavior

• Customer Segmentation Matrix
• Customer-wise Spend Table
• Region Map
• Order Frequency Breakdown
• High-risk churn group analysis

---

📸 Dashboard Preview

Page 1 – Customer Status

Page 2 – Customer Behavior

---

📈 Key Insights

• 30% customers are churned — needs strong retention strategy
• Repeat customers have 2.5× higher lifetime value
• West region shows lowest churn rate
• Low-frequency buyers churn the most
• High-spend customers show highest retention
• Churn risk increases after 120+ days inactivity

---

▶ How to Use the Dashboard

1. Download the .pbix file
2. Open with Power BI Desktop
3. Explore slicers, filters, and pages
4. Review customer-level details under Page 2

---

👤 Author

G. Shyam Venkat

Data Analyst | SQL | Power BI | DAX | Excel | Data Modeling

🔗 GitHub: https://github.com/shyamcodes-ai
🔗 LinkedIn: https://www.linkedin.com/in/g-shyam-venkat-304ab536b

---

⭐ If you found this project useful, consider giving the repository a star!