An end-to-end Data Analytics project that analyzes customer shopping patterns using Python, PostgreSQL, and Power BI to generate actionable business insights.
This project analyzes transactional retail data to understand customer behavior, spending patterns, product performance, and subscription impact.
The goal is to demonstrate a complete analytics workflow — from raw data to dashboard, report, and presentation.
- ~3,900 customer purchase records
- 18 features including demographics, purchase details, and shopping behavior
Key attributes
- Age, Gender, Location, Subscription Status
- Item Purchased, Category, Purchase Amount
- Discount Applied, Review Rating, Shipping Type
Missing values were identified and handled during preprocessing.
- Python: pandas, numpy, matplotlib, seaborn
- Database: PostgreSQL
- Visualization: Power BI
- Reporting: Analytical report (PDF)
- Presentation: Gamma (AI-powered PPT)
- Data Loading & EDA (Python)
- Data inspection, summary statistics, visualization
- Data Cleaning & Feature Engineering
- Missing value treatment
- Column standardization
- Customer segmentation & age grouping
- SQL Analysis (PostgreSQL)
- Revenue and spending analysis
- Product and customer behavior queries
- Dashboard Development (Power BI)
- Interactive KPIs and filters
- Reporting & Presentation
- Business insights report
- Executive presentation using Gamma
- Revenue by gender
- High-spending discount users
- Top-rated and top-selling products
- Subscribers vs non-subscribers comparison
- Revenue by age group
- Shipping type impact on spending
The interactive dashboard includes:
- Customer count & average purchase KPIs
- Revenue by category and age group
- Subscription distribution
- Sales and behavior trends
Designed for quick insight discovery and business decision support.
- Loyal customers form the largest segment
- Certain products are highly discount-dependent
- Subscribers and non-subscribers show similar average spend but different volumes
- Young and middle-aged customers contribute the highest revenue
- Python 3.x
- PostgreSQL
- Power BI Desktop
# Install dependencies
pip install pandas numpy matplotlib seaborn psycopg2- Run Python scripts for EDA and data cleaning
- Load cleaned data into PostgreSQL
- Execute SQL queries for analysis
- Open Power BI file to explore the dashboard
- This project showcases:
- Strong data cleaning & EDA skills
- Practical SQL for business analytics
- Dashboard storytelling with Power BI
- End-to-end data-to-decision workflow
Tarun Kumar S A Data Analytics | Python | SQL | Power BI