📊 Vendor Performance Analysis

Overview

This project combines Python, SQL, and Power BI to deliver a comprehensive Vendor Performance Analysis for the retail and wholesale sector. It demonstrates an end-to-end workflow — from data ingestion and transformation to visual analytics and KPI dashboards — to identify top-performing vendors, underperforming brands, and opportunities to optimize procurement and pricing decisions.

Business Problem

In retail and wholesale, profitability can be eroded by poor pricing, inefficient vendor management, and slow inventory turnover.
This project addresses those challenges by:
🏷️ Identifying underperforming brands needing promotional or pricing adjustments.
💰 Highlighting top vendors driving sales and gross profit.
📦 Analyzing how bulk purchasing impacts unit costs.
🔄 Evaluating inventory turnover to reduce holding costs.
📈 Assessing profitability variance between high- and low-performing vendors.\

Project Structure

Vendor-Performance-Analysis/
│
├── ingestion_db.py # Ingests CSVs into SQLite database
├── get_vendor_summary.py # Script to get cleaned vendor summary table and ingest into database
├── Exploratory Data Analysis.ipynb # Python-based analysis and insights
├── Vendor Performance Analysis # Pyhon-based Vendor performance specific analysis
├── Dashboard.pbix # Power BI file ├── logs/ # Ingestion logs
├── data/ # Raw CSV data (Sample)
└── README.md # Project documentation\

Tech Stack

Category	Tools Used
Programming & Analysis	Python (Pandas, NumPy, Seaborn, Matplotlib)
Database	SQLite via SQLAlchemy
Visualization	Power BI
Scripting	Python-based data ingestion
Statistical Testing	Confidence intervals, t-tests

Analytical Workflow

Python Phase

1. Data Ingestion: Using SQLAlchemy and pandas to store CSVs in SQLite.
2. Data Cleaning: Filtering invalid rows (zero sales, negative profits).
3. EDA: Summary stats, correlations, and outlier detection.
4. Analysis:
   • Profitability comparison across vendors.
   • Statistical tests (t-test, confidence intervals).
   • Bulk purchase cost savings.
   • Unsold inventory valuation.

Power BI Phase

1. Python Data Connection: Used Python script to import the cleaned dataset from SQLite.
2. Transformations: Applied Power Query filters to refine vendor summary data.
3. DAX Calculations: Created calculated tables, columns, and measures for KPIs.
4. Dashboard Design
   Each visual ties back to analytical findings derived in Python, providing a real-time, interactive representation of vendor performance.