Customer-Behavior-Analysis

Project Overview

This project analyzes the shopping habits of 3,900 customers to uncover actionable business insights. It combines Python, SQL Server, and Power BI to perform end-to-end data analysis, including data cleaning, feature engineering, exploratory analysis, and visualization.

Key Highlights:

• Average purchase amount: $59.76
• Average review rating: 3.75
• Subscription penetration: 27% subscribed, 73% unsubscribed
• Top-performing category: Clothing

---

Table of Contents

• Dataset
• Tools & Technologies
• Workflow
• Key Insights
• Recommendations
• Project Files
• How to Run

---

Dataset

The dataset customer_shopping_behavior.csv contains transactional and demographic information for 3.9K customers, including:

• Customer ID, Age, Gender
• Product Category, Purchase Amount
• Review Rating, Purchase Frequency
• Subscription Status, Discount Usage

---

Tools & Technologies

• Python (Pandas, NumPy) – Data cleaning, feature engineering
• SQL Server – Exploratory Data Analysis
• Power BI – Dashboard creation and visualization
• Git & GitHub – Version control

---

Workflow

1. Data Preparation

• Cleaned and standardized column names
• Handled missing review ratings using median by category
• Created age groups for segmentation
• Mapped purchase frequency to numeric days
• Dropped redundant columns

2. Database Integration

• Uploaded the cleaned dataset to SQL Server in a table named customer_behavior_table.

3. Exploratory Data Analysis

• Revenue by demographics (gender, age group)
• Discount and spending analysis
• Top product performance
• Customer segmentation (new, returning, loyal)
• Subscription behavior

4. Visualization

• Interactive Power BI dashboard with filters for subscription, gender, category, and shipping type
• KPI cards: total customers, average purchase, average review rating
• Category and demographic performance charts

---

Key Insights

1. Clothing is the top revenue driver
2. Young adult and middle-aged customers are the most valuable demographic groups
3. Subscription potential is high – 73% of customers are unsubscribed

---

Recommendations

• Launch targeted subscription campaigns for loyal customers (>10 purchases)
• Focus marketing and inventory on Clothing category for high-converting age groups
• Implement strategic discount bundles to encourage higher spending

---

Project Files

• customer_shopping_behavior.csv – Raw dataset
• project_report.pdf – Final project report
• Jupyter Notebooks – Python scripts for cleaning and feature engineering
• Power BI Dashboard files

---

How to Run

1. Clone the repository:
   git clone https://github.com/Manish-2310/Customer-Behavior-Analysis.git
2. Open the Jupyter Notebook to perform data cleaning and feature engineering.
3. Run SQL scripts on SQL Server for exploratory analysis.
4. Open the Power BI dashboard to visualize insights.

---

Author

Manish Sharma – Data Analyst / Data Science Enthusiast

---

License

This project is for educational purposes.