Customer Behaviour Analysis – Data Analytics Project

📌 Project Overview

This repository contains an end-to-end data analytics project focused on customer shopping behavior.
The project uses Python for data cleaning and preprocessing, SQL for exploratory analysis, and Power BI for interactive dashboards to uncover actionable insights.

Objective:
Analyze a dataset of 3,900 customer transactions to understand purchasing patterns, segment valuable customers, and provide recommendations for marketing, sales, and inventory management.

---

📂 Dataset

The dataset contains detailed information about customer transactions, including demographics, purchase details, subscription status, and shipping preferences.

Sample Columns

• CustomerID – Unique identifier for each customer
• Age – Customer age
• Gender – Customer gender
• Item Purchased – Name of purchased item
• Category – Product category (e.g., Clothing, Accessories)
• Purchase Amount (USD) – Transaction value
• Location – Customer location
• Size – Product size
• Color – Product color
• Season – Season of purchase
• Review Rating – Customer review rating
• Subscription Status – Whether the customer is a subscriber
• Shipping Type – Type of shipping chosen
• Discount Applied – Whether discount was applied
• Promo Code Used – Promo code usage
• Previous Purchases – Count of previous purchases
• Payment Method – Payment mode
• Frequency of Purchases – Purchase frequency

(Dataset file: customer_shopping_behavior.csv)

---

🛠 Tools & Technologies

• Python: Pandas, NumPy, Matplotlib, Seaborn
• SQL Databases: PostgreSQL / MySQL / SQL Server
• SQLAlchemy: Database connectivity
• Power BI: Interactive dashboard
• Jupyter Notebook / VS Code: Development environment
• Git & GitHub: Version control

---

📘 Project Workflow

1. Data Loading & Preprocessing (Python)

• Load the dataset using Pandas
• Inspect data types, missing values, and duplicates
• Generate summary statistics
• Feature engineering (e.g., Age Groups, Revenue Metrics)
• Cleaned data is loaded into PostgreSQL via SQLAlchemy

2. Exploratory Data Analysis (EDA)

• Distribution analysis of purchase amounts
• Category-level revenue analysis
• Customer segmentation by age, gender, subscription status
• Visualizations using Matplotlib and Seaborn

3. SQL Analysis

The cleaned dataset is analyzed using SQL queries (customer_behaviour_analysis.sql) to uncover business insights:

• Top revenue-generating products and categories
• Customer segmentation: New, Returning, Loyal
• Impact of discounts and subscription status on revenue
• Shipping preferences and satisfaction analysis
• Age group and demographic behavior analysis

4. Power BI Dashboard

An interactive Power BI dashboard was created to present insights:

Key KPIs & Insights:

• Total Customers: 3,900
• Total Revenue: $233K
• Average Purchase Amount: $59.76
• Average Review Rating: 3.75
• Top Product Categories: Clothing and Accessories
• Subscriber Analysis: 27% of customers, higher average spend
• Top Customer Segment: Young Adults
• Shipping Preferences: Free Shipping and 2-Day Shipping are most popular

Dashboard Features:

• Interactive filters: Gender, Age Group, Category, Subscription, Shipping Type
• Revenue & sales analysis by category, age group, and subscription status

---

🚀 How to Use This Project

1. Setup Database

1. Create a PostgreSQL database (e.g., customer_behavior)
2. Update database connection in Customer_Shopping_Behaviour_Analysis.ipynb

2. Run Python Notebook

• Execute cells in Customer_Shopping_Behaviour_Analysis.ipynb
• Load, clean, and preprocess data
• Push cleaned data to PostgreSQL

3. Run SQL Queries

• Connect using PgAdmin, DBeaver, or any SQL client
• Execute queries in customer_behaviour_analysis.sql

4. Power BI Dashboard

• Open customer_behaviour_analysis.pbix in Power BI Desktop
• Connect to your PostgreSQL database and update credentials
• Explore interactive visualizations and KPIs

---

📈 Key Insights

• Top Categories: Clothing and Accessories dominate revenue
• High-Value Segment: Young Adults contribute most to purchases
• Subscribers vs Non-Subscribers: Subscribers spend more and are more loyal
• Shipping Preferences: Free Shipping & 2-Day Shipping lead to higher satisfaction
• Purchase Behavior: Average purchase ~$59–$60, Average review rating ~3.75

---

📁 Repository Structure

├── .gitignore ├── customer_shopping_behavior.csv # Raw dataset ├── Customer_Shopping_Behaviour_Analysis.ipynb # Python notebook for ETL ├── customer_behaviour_analysis.sql # SQL exploratory queries ├── customer_behaviour_analysis.pbix # Power BI project file ├── customer_behaviour_analysis_dashboard.png # Exported dashboard image └── README.md # Project documentation

---

📌 Conclusion

This project demonstrates an end-to-end data analytics workflow, integrating Python, SQL, and Power BI to analyze customer purchasing behavior. The insights derived can help businesses improve marketing strategies, inventory planning, and customer retention.

---

📫 Connect / Contact

• GitHub: https://github.com/msahid-cse
• LinkedIn: https://www.linkedin.com/in/msahid-cse/
• Website: https://msahid-cse.github.io/portfolio/