This project demonstrates an end to end data analytics workflow focused on evaluating vendor performance using large scale transactional data. The goal was to transform raw purchase, sales, pricing, and freight data into meaningful business insights that support decision making around vendor selection, pricing strategies, and inventory optimization.
The analysis begins with structured data ingestion into a relational database, followed by advanced SQL querying using joins and common table expressions to build vendor level summaries. These aggregated datasets are then analyzed using Python to uncover patterns in sales performance, profit margins, freight costs, and inventory behavior.
Through exploratory data analysis and visualization, the project identifies top performing vendors, underperforming vendors, pricing inefficiencies, and high inventory risk areas. Statistical techniques such as correlation analysis, distribution analysis, and confidence interval comparison were applied to validate trends and support data driven conclusions.
The final outcome is a comprehensive analytical report with clear visual storytelling, actionable insights, and performance comparisons that can be used by business stakeholders to improve vendor strategy and operational efficiency.
Key Objectives
Analyze vendor level sales, purchases, profit margins, and freight costs
Identify top revenue generating and high profit vendors
Detect underperforming vendors and low margin products
Evaluate inventory turnover and unsold inventory risk
Support pricing and promotional decision making using data
Tools and Technologies
Python (Pandas, NumPy)
SQL (CTEs, joins, aggregations)
Matplotlib and Seaborn for data visualization
SQLite for data storage and querying
Jupyter Notebook for analysis and reporting
Analysis Highlights
Built KPI level summaries for sales, purchases, profit margin, and freight cost
Visualized vendor and brand performance using bar charts, scatter plots, and heatmaps
Performed correlation analysis to understand relationships between pricing, sales, and profitability
Compared top and low performing vendors using confidence interval analysis
Identified brands suitable for pricing adjustments or promotional strategies
Business Insights
A small group of vendors contributes disproportionately to total sales revenue
Profit margin does not always increase with higher sales volume
Freight cost has a measurable impact on overall vendor profitability
Certain brands show high inventory levels with low sales velocity, indicating optimization opportunities
Outcome
This project showcases my ability to handle the complete analytics lifecycle from raw data ingestion and SQL modeling to Python based analysis, visualization, and business insight generation. It reflects real world analytical thinking and demonstrates how data can be translated into actionable decisions.