🛒 Walmart Sales Data Analysis – SQL Project

📌 Project Overview

This project is an end-to-end data analysis solution designed to extract critical business insights from Walmart sales data. We utilize SQL Server to process raw transactional data, employing advanced techniques like ETL (Extract, Transform, Load), Feature Engineering, and Exploratory Data Analysis (EDA).

The goal is to identify high-performing branches, analyze sales patterns, and decipher customer behavior to support data-driven decision-making.

Key Competencies Demonstrated:

• Advanced SQL: CTEs, Window Functions, Aggregate Functions, Stored Procedures.
• ETL Processes: Data cleaning, validation, and BULK INSERT operations.
• Business Intelligence: Translating data into actionable insights regarding revenue, inventory, and customer segmentation.

---

🛠️ Tech Stack

• Database: Microsoft SQL Server
• Language: T-SQL
• Tools: SQL Server Management Studio (SSMS)
• Techniques: Data Cleaning, Feature Engineering, Window Functions, CTEs

---

📂 Project Structure

SQL-Walmart-Sales-Analysis
│
├── ddl.sql                 # Database schema and table creation
├── load_proc.sql           # Stored procedures for bulk data loading
├── eda_sales_analysis.sql  # Exploratory Data Analysis & Business Questions
├── datasets/               # Raw CSV data files
├── README.md               # Project documentation
└── project_report.pdf      # Detailed project report

---

📊 Business Questions Solved

The analysis is structured around three major pillars: Product, Sales, and Customer.

The analysis queries (eda_sales_analysis.sql) are structured to answer key business questions across multiple dimensions.

---

1️⃣ Generic & Geographic Analysis

• City Coverage:
  How many unique cities and branches does the dataset include?

• Branch Location Mapping:
  In which city is each branch located?

---

2️⃣ Product Analysis

Focused on identifying revenue drivers and underperforming product categories.

• Unique Product Lines:
  How many unique product categories exist in the dataset?

• Top Sellers:
  Which product line has the highest sales quantity?

• Revenue Contribution:
  Which product line generates the highest total revenue?

• Taxation Analysis:
  Which product category incurs the highest VAT (Value Added Tax)?

• Profitability Indicator:
  Which product line has the highest average customer rating?

---

3️⃣ Sales Analysis

• Temporal Trends:
  Which time of day (Morning / Afternoon / Evening) records the highest sales?

• Customer Type Performance:
  Which customer segment (Member vs Normal) generates more revenue?

• City Performance:
  Which city contributes the highest VAT percentage?

• Payment Behavior:
  What is the most frequently used payment method?

---

4️⃣ Customer Analysis

• Demographics:
  What is the gender distribution across different branches?

• Customer Ratings:
  Which time of day receives the highest customer ratings?

• Spending Patterns:
  Which day of the week has the best average customer ratings?

---

⚙️ Engineering Process

1️⃣ Database Design & Schema

A star-schema-inspired design was implemented to support analytical querying and reporting use cases. Data integrity was maintained using Primary Keys and NOT NULL constraints.

Column	Type	Description
invoice_id	VARCHAR(50)	PK – Unique transaction ID
branch	VARCHAR(5)	Store branch (A, B, C)
city	VARCHAR(30)	Store location
customer_type	VARCHAR(30)	Member / Normal
gender	VARCHAR(10)	Customer gender
product_line	VARCHAR(100)	Product category
unit_price	DECIMAL(10,2)	Price per unit
quantity	INT	Number of units sold
tax	DECIMAL(10,2)	5% tax amount
total	DECIMAL(10,2)	Final transaction value
date	DATE	Transaction date
time	TIME	Transaction time
rating	DECIMAL(2,1)	Customer rating (1–10)

---

2️⃣ Feature Engineering

To enable deeper analysis, I engineered the following columns directly in SQL:

• time_of_day → Categorized using CASE statements (Morning, Afternoon, Evening).
• day_name → Derived from date to analyze weekly footfall.
• month_name → Derived from date to track seasonal performance.

These features allow deeper insights into sales trends and customer behavior.

---

3️⃣ Data Wrangling (ETL)

• Cleaned missing and inconsistent values
• Standardized numeric and categorical fields
• Loaded large datasets using BULK INSERT
• Ensured data consistency and integrity

---

💻 Sample Advanced SQL Query

Demonstrating the use of CTEs and Window Functions to identify the top-performing branch for each product line:

WITH BranchSales AS (
    SELECT 
        branch,
        product_line,
        SUM(total) AS total_revenue,
        RANK() OVER (PARTITION BY product_line ORDER BY SUM(total) DESC) AS rank
    FROM sales
    GROUP BY branch, product_line
)
SELECT 
    branch,
    product_line,
    total_revenue
FROM BranchSales
WHERE rank = 1;

🎯 Why This Project Matters

This project demonstrates:

This project simulates how SQL is used in real business environments to support reporting, performance analysis, and decision-making.

• Strong SQL fundamentals
• Business-oriented data analysis
• Real-world ETL concepts
• Analytical thinking
• Clean project structuring

---

🚀 How to Run the Project

1️⃣ Clone the Repository

git clone https://github.com/ManoharManu1999/SQL-Walmart-Analysis.git

2️⃣ Initialize Database

• Open ddl.sql in SQL Server Management Studio (SSMS)
• Execute the script to create the required tables

3️⃣ Load Data

• Run load_proc.sql to import the CSV files into the database

4️⃣ Execute Analysis

• Open and run queries from eda_sales_analysis.sql to perform analysis

👤 About Me

I built this project to strengthen my understanding of SQL, data analysis, and analytical problem-solving.

I enjoy working with data to uncover patterns, solve business problems, and deliver actionable insights.

I’m actively exploring opportunities in:

• Data Analytics
• SQL Development
• Business Intelligence
• Analytics Engineering

📬 Let’s Connect

🔗 LinkedIn 📧 Email

⭐ Final Note

If you found this project useful or interesting, feel free to ⭐ star the repository. Feedback and suggestions are always welcome!