🛍️ Online Retail Analysis

This project explores online retail transaction data using a mix of SQL for data cleaning and exploration and Power BI for creating an interactive, insightful dashboard.

---

📂 Project Structure

Here's how everything is organized:

Online_Retail_Analysis/

│

├── README.md

├── DATA/

│ ├── online_retail

│ └── online_retail_clean.csv # cleaned data for analysis

│

├── Power BI/

│ └── Dashboard_OnlineRetail.pbix

│ └── Power BI report preview.pdf

│

├── SQL/

│ └── create_table.sql

│ └── data_cleaning.sql

│ └── data_querying.sql

|

---

📌 About the Dataset

The dataset contains ~540K transaction records from a UK-based online retailer (2010–2011). It includes:

• Customer info (CustomerID, Country)
• Product info (StockCode, Description)
• Transaction info (InvoiceNo, Quantity, UnitPrice, InvoiceDate)

---

🧹 Data Cleaning (SQL in pgAdmin)

Before jumping into Power BI, I used SQL to clean and prep the data:

• Replaced missing CustomerID with '0' (represents guest customers)
• Removed invalid entries (Quantity <= 0, UnitPrice <= 0)
• De-duplicated records using GROUP BY
• Trimmed product descriptions and made them lowercase

---

🔍 SQL Data Exploration

I also ran several queries to get a sense of what was going on in the data:

• Total revenue per customer
• Customers with only one order
• Top countries by number of customers
• Most purchased products and combinations
• Monthly revenue trends

These insights helped shape the Power BI report structure.

---

📊 Power BI Dashboard Highlights

💰 Executive Summary Page

• Monthly revenue trend line chart
• Monthly transactions trend line chart
• KPI : Total Revenue, Total Orders
• Country-wise revenue

🗺️ Map View Page

• Country-Wise Revenue Map showing where your money comes from

🧍 Customer Report Page

• Top 10 spenders & top order-makers
• Total spend and order count per customer
• Guest vs Registered customer counts
• Repeat Customer Rate (%)
• Slicer: Country filter

📦 Product Report Page

• Top-selling and highest revenue products
• Slicer: Country filter and date filter

---

🧮 DAX Measures Used

Some custom measures built using DAX:

• Total Revenue
• Total Orders per Customer
• Repeat Customer Rate (%)
• Guest Customer Count
• Registered Unique Customers
• Monthly Revenue
• Top Products by Revenue / Quantity

---

🧰 Tools & Tech

• PostgreSQL (via pgAdmin) – Data cleaning & SQL queries
• Power BI Desktop – Dashboard design & visuals
• DAX – Custom calculations
• VS Code – Script management

---

🚀 Possible Improvements

• Add RFM-based customer segmentation
• Build predictive revenue trends
• Tag or cluster products (if more metadata available)

---

Thanks for reading! ⭐️