Business Problem:
An e-commerce company was offering heavy discounts without a clear understanding of their impact on profits or customer retention.

Approach:
Used a structured three-step process:

1. Excel – Data cleaning & transformation

2. SQL Server – Exploratory data analysis

3. Power BI – Interactive dashboard design

Key Outcomes:

• Flagged 3 discount campaigns causing financial loss

• Identified the 25–44 age group as the most valuable customer segment

• Recommended strategies to stabilize margins during low-sales months (Feb, Jul)

🛠 Tech Stack

🧾 Business Case

The company aimed to improve profitability and retention by understanding:

• Which discounts actually improve margins

• How customer behavior varies across age, gender, location

• Which categories and channels drive most of the revenue

🎯 Project Objectives

• Analyze customer behavior across demographics and regions

• Measure discount effectiveness on both revenue and profit

• Identify high-value segments for personalized marketing

• Deliver actionable insights through a clear and concise dashboard

📦 Dataset Overview

The dataset contains 15,000+ anonymized e-commerce transactions.

🔧 Process Summary

✅ Step 1: Excel – Data Cleaning

• Standardized date, numeric, and categorical formats

• Imputed missing Gender, Location, and Age_Group values

• Created new time features: Purchase_Month, Purchase_Year

✅ Step 2: SQL – Exploratory Data Analysis

Key Findings:

🧮 KPIs

• Transactions: 15,000+

• Customers: ~9,000

• Net Revenue: ₹100M+

📊 Demographics

• 25–44 age group generated the most revenue

• Balanced gender ratio, but higher revenue from females

• Top cities: Delhi, Bangalore, Mumbai

📦 Category Analysis

• Electronics & Clothing were top in revenue

• Home Decor had lower volume but higher profit margins

💸 Discount Effectiveness

• Used in 60%+ transactions

• Not all discounts drove profit – "Flash Sale" was loss-making

• Some discounts exceeded product value, leading to negative profit

📍 Channel Mix

• Online had highest sales volume

• Offline had higher average profit per transaction

📆 Seasonality

• Best months: Nov & Dec

• Weakest: Feb & Jul

• Q4 contributed over 35% of yearly revenue

✅ Step 3: Power BI – Dashboard Creation

The dashboard includes:

🔍 Key Insights

📢 Business Recommendations

• Eliminate or reconfigure 3 unprofitable discount campaigns

• Target 25–44 age group in Tier 1 cities with loyalty offers

• Bundle high-margin products with popular electronics

• Run boosted campaigns in Feb and Jul to smooth seasonal dips

• Double down on offline promotions, leveraging higher profit/order

📈 Business Impact

📁 Folder Structure on GitHub

🔗Repository Link: Github

ecommerce-analysis/
├── Dashboard_preview/   → Dashboard screenshots
├── Excel/               → Cleaned Data 
├── PowerBI/             → .pbix file with full report
├── SQL-Query/           → All SQL code for metrics and analysis

📬 Let’s Connect

I’m actively exploring data analyst opportunities and building real-world, outcome-driven portfolio projects.

• 🔗 GitHub: @Dikshatomarr

• 💬 Open to feedback, collaborations, or mentorship

⭐ If you're a recruiter or hiring manager:

This project demonstrates core skills in data wrangling, analysis, visualization, and business insight generation—exactly what's needed in high-impact analyst roles.