An exploratory data analysis project using Python to analyze sales transactions, identify revenue trends, evaluate product performance, and uncover geographic market insights.
This project analyzes a sample sales transactions dataset covering 2003–2005.
The analysis focuses on understanding:
- Revenue performance over time
- Product-line performance
- Geographic market performance
- Data quality and missing values
- Monthly and yearly sales trends
The project demonstrates a complete introductory data analytics workflow, from data inspection and cleaning to analysis, visualization, and business interpretation.
The main objectives of this project were to:
- Understand the structure and quality of the sales dataset
- Identify and handle missing values
- Analyze revenue trends over time
- Determine the best-performing product lines
- Identify the strongest geographic markets
- Create visualizations that communicate key findings clearly
Source: Sample Sales Data — Kaggle
The dataset contains 2,823 sales transactions and 25 variables, including:
- Order information
- Product line
- Customer information
- Sales/revenue
- Geographic information
- Deal size
- Order dates
The dataset contains several data-quality issues, including missing values in fields such as address, state, and territory.
- Python
- Pandas
- Matplotlib
- Google Colab
- Jupyter Notebook
- Imported the dataset using Pandas
- Examined dimensions and data types
- Inspected missing values
- Reviewed categorical and numerical variables
- Identified missing values
- Checked data consistency
- Prepared the dataset for analysis
- Revenue aggregation
- Product-line analysis
- Geographic analysis
- Monthly revenue analysis
- Yearly revenue comparison
Created charts to communicate:
- Revenue by product line
- Monthly revenue trends
Total revenue across the dataset was approximately:
$10.03 million
Classic Cars generated approximately $3.9 million, making it the highest-revenue product line.
The United States generated approximately $3.6 million in revenue, followed by:
- Spain — approximately $1.2M
- France — approximately $1.1M
Annual revenue peaked in 2004 at approximately $4.7 million.
The 2005 data represents only part of the year, so it should not be directly compared with the full-year figures from 2003 and 2004.
The analysis suggests several useful business observations:
- Classic Cars represent a major source of revenue and could receive additional marketing and inventory attention.
- The United States is the strongest geographic market in the dataset.
- Revenue varies considerably across product lines, suggesting opportunities for product-specific strategies.
- Monthly analysis can help identify periods of stronger and weaker sales activity.
- Because 2005 contains partial-year data, conclusions about annual performance should account for the incomplete period.
sales-data-analysis/
│
├── README.md
├── analysis.ipynb
├── sales_data_sample.csv
├── revenue_by_product.png
└── monthly_revenue_trend.png
Through this project, I practiced:
- Data loading and inspection
- Data cleaning
- Missing-value analysis
- Pandas data manipulation
- GroupBy aggregation
- Revenue analysis
- Exploratory data analysis
- Data visualization
- Communicating analytical findings
Possible extensions to this project include:
- Customer segmentation
- Profitability analysis
- Sales forecasting
- Interactive Power BI dashboard
- Advanced time-series analysis
- Automated reporting
Jassica Michael
Statistics Graduate | Data Analyst
GitHub: @jessica795