Customer Shopping Behavior — End-to-End Data Analytics Project 🔍 Overview
This project analyzes customer shopping behavior using transactional data from 3,900 purchases across multiple product categories.
The objective is to understand:
spending patterns
product preferences
subscription behavior
customer segments
and translate these insights into clear business recommendations.
The project follows a full analytics pipeline — from raw data to dashboard and presentation.
📁 Dataset
Rows: 3,900 Columns: 18
Key features include:
Demographics: Age, Gender, Location, Subscription Status
Purchases: Item, Category, Amount, Season, Size, Color
Behavior: Discount Applied, Previous Purchases, Frequency, Review Rating, Shipping Type
Missing data: 37 values in Review Rating (handled during cleaning)
🛠️ Tools & Technologies
Python (Pandas, NumPy, Matplotlib/Seaborn) – EDA, cleaning, feature engineering
MySQL – structured business queries
Power BI – interactive dashboard
Gamma – presentation & storytelling
🧭 Project Steps 1️⃣ Load & Explore Data (Python)
Imported data using pandas
Reviewed structure with info() and describe()
Checked duplicates and missing values
2️⃣ Data Cleaning & Preparation
Imputed missing review ratings (median per category)
Standardized column names (snake_case)
Dropped redundant fields (promo_code_used)
Ensured datatype consistency
3️⃣ Feature Engineering
Created age_group (binned age ranges)
Created purchase_frequency_days
Prepared a clean dataset for SQL analysis
4️⃣ SQL Analysis (MySQL)
Loaded the dataset into the database and answered key business questions such as:
Revenue by gender
High-spending discount users
Top products by rating
Standard vs express shipping spend
Subscriber vs non-subscriber value
Discount-dependent products
Customer segmentation (New, Returning, Loyal)
Top products per category
Subscription likelihood among repeat buyers
Revenue by age group
5️⃣ Dashboard (Power BI)
Designed an interactive dashboard to visualize:
KPIs and revenue trends
Customer segments
Product performance
Subscription & discount behavior
6️⃣ Reporting & Presentation
Wrote a concise insights report
Built a Gamma presentation summarizing findings and recommendations
📈 Key Results & Insights
Discounts boost sales — but high spenders still purchase without deep discounts.
Loyal and returning customers generate significantly higher lifetime value.
Certain product categories rely heavily on discounts.
Express shipping customers tend to spend more.
Specific age groups contribute the majority of revenue.
🎯 Business Recommendations
Promote subscription benefits to increase recurring revenue
Introduce loyalty rewards for repeat buyers
Optimize discount strategy to protect margins
Highlight top-rated and best-selling products in campaigns
Target high-value segments with focused marketing
Python 3.x
MySQL
Power BI Desktop
Gamma account (for presentation)
Steps
Clone this repository
Open the Python notebook and run the EDA/cleaning workflow
Load the cleaned dataset into MySQL
Run the SQL queries provided
Open the Power BI file and refresh the data
View the final report and presentation
⭐ What This Project Demonstrates
✔ End-to-end analytics workflow ✔ Strong EDA and data cleaning ✔ SQL for business-driven questions ✔ Dashboard storytelling ✔ Clear communication of insights and recommendations