Project Overview

Project Title: Retail Sales Analysis
Level: Beginner
Database: sales

Project Context

Retail businesses generate thousands of transactions daily, but raw transaction data is not directly useful unless it is properly structured and cleaned. In this project, I worked with retail sales transaction data that needed to be organized into a database and validated for missing/invalid records so that accurate analysis could be performed. The goal was to understand category performance, customer behaviour, and time-based sales trends using SQL.

Primary Goal

The primary goal of this project is to analyze retail sales data and uncover actionable insights that can support business decision-making. I focused on identifying sales patterns, revenue drivers, customer segments, and operational time windows (shifts) that can be used for better planning, marketing, and performance tracking.

Data Foundation (What I Built)

To build a strong base for analysis, I:

• Created a database (sales) and a transaction table (retail_sales) containing customer, product, and financial attributes (transaction date/time, customer demographics, category, quantity, unit price, COGS, and total sale).
• Validated the structure and performed initial checks (sample records, row counts, basic exploration).
• Performed data cleaning (Null Handling) by identifying and removing records with missing values in key business fields to ensure reliable analysis and trustworthy results.

Exploratory Data Analysis (EDA)

After cleaning, I performed EDA to understand the dataset at a high level, including:

• Total number of sales transactions available for analysis
• Number of unique customers in the dataset
• Product categories present (to understand product diversity and category-level behaviour)

Business Questions & Analysis Approach

Once the data foundation was ready, I answered a set of real business questions using SQL. The queries in this project cover:

• Date-based transaction retrieval (to track specific-day performance)
• Category-based filtering and monthly conditions (e.g., Clothing bulk purchases in Nov-2022)
• Category-level performance metrics (total sales and order volume)
• Customer demographic insights (e.g., average age of Beauty customers)
• Identification of high-value transactions (total_sale > 1000)
• Gender × category transaction patterns (to support targeted marketing)
• Monthly trend analysis and identifying the best-performing month each year (using ranking logic)
• Top customer identification (top 5 customers by total spending)
• Unique customers per category (to understand category reach)
• Shift-based order distribution (Morning/Afternoon/Evening) based on sale time (to support staffing and operational planning)

Key Summary (What the Analysis Shows)

This project converts raw retail transaction records into meaningful insights such as:

• Overall business scale (transactions, customers, revenue, categories)
• Which categories drive the most revenue vs. the most orders
• Which customer segment appears more active in certain categories
• When demand peaks during the day (shift analysis) and what that means operationally

Overall, this is a complete beginner-friendly SQL case study that demonstrates an end-to-end workflow: database setup → cleaning → EDA → business question solving → insight generation.