https://github.com/harshbrar03/retail-store-sales-analysis

A beginner-friendly data analytics project demonstrating the complete ETL workflow using a retail dataset. It involves data cleaning with Python, storing in PostgreSQL, querying with advanced SQL, and visualizing insights through Python.
https://github.com/harshbrar03/retail-store-sales-analysis

matplotlib pandas postgresql-database python sql sqlalchemy

Last synced: 3 months ago
JSON representation

A beginner-friendly data analytics project demonstrating the complete ETL workflow using a retail dataset. It involves data cleaning with Python, storing in PostgreSQL, querying with advanced SQL, and visualizing insights through Python.

• Host: GitHub
• URL: https://github.com/harshbrar03/retail-store-sales-analysis
• Owner: harshbrar03
• Created: 2025-06-21T09:07:17.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2025-06-21T09:46:35.000Z (about 1 year ago)
• Last Synced: 2025-06-21T10:23:53.616Z (about 1 year ago)
• Topics: matplotlib, pandas, postgresql-database, python, sql, sqlalchemy
• Language: Jupyter Notebook
• Homepage:
• Size: 706 KB
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: readME.md

Awesome Lists containing this project

README

          
# 🛒 Retail Store Sales Analysis (SQL + Python + PostgreSQL)

This project demonstrates a complete data workflow — from data extraction and transformation to loading it into a PostgreSQL database and performing business analysis using SQL and Python. It follows a simplified ETL (Extract, Transform, Load) process, similar to those used in real-world data workflows.

---

## 📦 Dataset

- **Source**: [Kaggle - Retail Orders Dataset](https://www.kaggle.com/datasets/ankitbansal06/retail-orders)

- **Data Size**: 9,994 records

- **Columns**: Order ID, Order Date, Product ID, Region, Category, Sub-Category, Segment, Cost Price, Discount, Profit, etc.

---

## 🧰 Tools & Technologies Used

- **Python** (Pandas, Matplotlib)

- **SQL** (PostgreSQL)

- **SQLAlchemy** (for DB connection)

- **Kaggle API** (for dataset download)

- **Jupyter Notebook**

---

## 📌 Folder Structure

Retail-Store-Analysis/


├── retail_analysis.ipynb     # Main Jupyter Notebook  

├── orders.csv               # Raw dataset (extracted)  

├── orders_cleaned.csv       # Cleaned dataset after transformation  

├── retail-orders.zip        # Downloaded zip from Kaggle  

├── queries.sql              # SQL queries for BI tool integration  

├── README.md                # Project overview and insights  


---

## 🧭 Project Workflow Diagram

![Workflow Diagram](workflow.jpg)

## ⚙️ Project Workflow

1. **Download** dataset from Kaggle using the API

2. **Extract** and read the file into a Pandas DataFrame

3. **Clean & transform** the data

    - Renamed columns

    - Handled missing values

    - Added new columns like `discount`, `sale_price`, and `profit`

4. **Push data** to a PostgreSQL database using SQLAlchemy

5. **Write SQL queries** to answer business questions

6. **Visualize** results using Python plots (Matplotlib)

---

## ❓ Business Questions Answered

1. What is the total sales and profit over time (monthly/yearly)?

2. Which regions are generating the highest profit and sales?

3. Most profitable categories and sub-categories?

4. Which ship modes are most used and profitable?

5. Top-performing cities/states by sales and profit

6. Impact of discount on profit

7. Category-wise Sales and Profit Overview

8. Which customer segments bring the most sales and profit?

9. Does Profit Margin Vary Across Sub-Categories?

---

## 📊 Key Insights

- Sales and profit vary month to month. Some months perform better than others.

- Western and Eastern regions have higher sales and profit.

- Copiers, Phones, and Accessories are among the most profitable sub-categories.

- Standard Class is the most used ship mode, but others also contribute well to profits.

- States like California lead in total sales.

- Higher discounts generally reduce profit, can even lead to losses.

- Technology is the most profitable and high-selling category overall.

- Consumers contribute the most to sales, but all segments bring value.

- Sub-categories like Copiers have better profit margins compared to others.

---

## 🧠 SQL Techniques Used

This project goes beyond basic querying and applies a variety of SQL techniques to extract meaningful insights from the data. Below are some of the key concepts used:

- **Aggregations & Grouping**: Used `SUM()`, `AVG()`, `COUNT()` with `GROUP BY` to calculate sales, profit, and orders.

- **Filtering & Sorting**: Applied `WHERE`, `CASE`, and `ORDER BY` to categorize data (e.g., by discount ranges or top performers).

- **Common Table Expressions (CTEs)**: Used `WITH` clauses to break complex queries into readable steps (e.g., monthly growth, top selling products per region).

- **Window Functions**: Applied `ROW_NUMBER() OVER(PARTITION BY ...)` to rank products by sales within each region or category.

- **Date Functions**: Used `EXTRACT()`, `TO_CHAR()` to analyze trends by month and year.

- **Profit Margin Calculation**: Computed custom metrics like profit margins using arithmetic in `SELECT` statements.

📄 The full set of queries can be found in the `queries.sql` file.

---

## 🚀 How to Run This Project

1. Clone the repository

2. Install dependencies: `pip install pandas sqlalchemy psycopg2-binary matplotlib`

3. Set up PostgreSQL and create a database named `retail_store`

4. Run the notebook `retail_analysis.ipynb` step by step