Retail Sales Analysis & Visualization using Power BI

Project Overview

This project analyzes Walmart weekly sales data to understand sales trends, store performance, and the impact of holidays and economic factors. The analysis includes data cleaning, exploration, visualization in Python, and an interactive Power BI dashboard for business insights.

Dataset: Walmart Sales Forecasting (Kaggle) Tools Used: Python (pandas, matplotlib, seaborn), Power BI Desktop

Problem Statement

Walmart generates huge amounts of sales data across multiple stores. However, raw data alone does not provide actionable insights. The challenge is to identify sales trends, top-performing stores, and the effect of holidays and external factors on sales.

The goal is to clean the dataset, explore patterns, visualize key insights, and build a dashboard to support data-driven decision making.

Workflow

1️ Data Cleaning

• Fixed missing values in numerical columns (Temperature, Fuel Price, CPI, Unemployment).
• Converted Date column to proper datetime format.
• Created new time-based columns: Year, Month.
• Removed duplicates and handled outliers where needed.

2️ Data Exploration (EDA)

• Analyzed total weekly sales trend over time.
• Identified Top 10 stores with highest average weekly sales.
• Compared holiday vs non-holiday sales.
• Studied top months and seasonal sales trends.
• Checked correlation between economic factors (Fuel Price, CPI, Unemployment) and sales.

3️ Visualization (Python)

• Line Chart: Sales trend over time.
• Bar Charts: Top stores and top months.
• Boxplot: Holiday vs Non-Holiday sales.
• Heatmap: Correlation between sales and economic factors.

4️ Power BI Dashboard

• KPI Cards: Total Sales, Average Weekly Sales, Total Stores, Total Months.
• Line Chart: Weekly Sales over time.
• Bar Chart: Top 10 Stores by Sales.
• Column Chart: Top 10 Months by Sales.
• Donut Chart: Holiday vs Non-Holiday sales.
• Scatter Plot: Relationship between Fuel Price and Sales.
• Filters: Year, Store

Key Insights (Example)

• Sales peak during holiday weeks (Thanksgiving, Christmas).
• Some stores consistently outperform others.
• Certain months show strong seasonal trends in sales.
• Economic factors (Fuel Price, CPI, Unemployment) show weak-to-moderate correlation with sales.

Folder Structure

Retail-Sales-Analysis-and-Visualization-using-Power-BI/
│─ Datasets/
│   ├─ walmart.csv                  # Original dataset from Kaggle
│   └─ walmart_cleaned.csv            # Cleaned dataset for Power BI
│─ notebooks/
│   └─ Retail_Sales_Analysis.ipynb  # Python notebook for cleaning & visualization
│─ walmart_sales_dashboard.pbit     # Power BI dashboard template
│─ README.md                        # Project overview

How to Run

1. Place walmart.csv in the Datasets/ folder.
2. Open Retail_Sales_Analysis.ipynb in Jupyter Notebook and run all cells to perform cleaning and visualizations.
3. The cleaned dataset will be saved as Datasets/walmart_cleaned.csv.
4. Open Power BI Desktop → Get Data → Import walmart_cleaned.csv.
5. Use the walmart_sales_dashboard.pbit template to build and explore the dashboard.

Skills & Tools Demonstrated

• Data Cleaning & Preprocessing (Python, pandas)
• Exploratory Data Analysis (Python, matplotlib, seaborn)
• Data Visualization & Storytelling
• Interactive Dashboard Creation (Power BI)
• Basic Time-Series & Seasonal Analysis