E-Commerce Operational Efficiency Analysis

This project analyzes e-commerce operational performance across buyer behavior, sales activity, referral efficiency, and vendor data quality. The goal is to uncover patterns that improve operational decision-making and highlight the business drivers behind revenue, profit, and customer experience.

Overview

The analysis combines:

• buyer profile data
• sales transaction data
• vendor product metadata
• operational metrics such as order timing, referral impact, and seasonal uplift

Through a structured data cleaning and enrichment pipeline, the project builds a master dataset that supports operational and strategic questions across customer segments, product performance, and channel behavior.

Business Questions Explored

• When do orders peak throughout the day?
• How do customer segments perform by quarter in terms of net profit?
• What is the financial impact of the referral program?
• How does Black Friday affect sales volume and revenue?
• Which customer groups deliver the strongest operational value?

Project Structure

.
├── data/
│   ├── inputs/
│   │   ├── buyer.csv
│   │   ├── sales.csv
│   │   └── Vendor Datasets/
│   ├── output_vendor/
│   └── output_vendor/   # cleaned vendor outputs
├── ouptput/
│   └── output_plots/
├── src/
│   ├── buyer_dq_check.py
│   ├── buyer_dq_fix.py
│   ├── sales_dq_check.py
│   ├── sales_dq_fix.py
│   ├── vendor_dq_check.py
│   ├── vendor_dq_fix.py
│   ├── merge_analyze.py
│   ├── question_answers.py
│   ├── visualizations.py
│   └── DataChallenge_C1.ipynb
├── DataChallenge_Slides.pptx
├── README.md
└── .gitignore

Key Outputs

The project generates business-ready insights and charts in the analysis pipeline, including:

• hourly order distribution
• profit heatmap by segment and quarter
• referral cost analysis
• Black Friday uplift comparison
• best customer base and KPI summary

Visual Highlights

Order Volume by Hour

Profit by Segment and Quarter

Referral Program Impact

Black Friday Impact

Best Customer Base

Presentation

A slide deck summarizing the project is included here:

• DataChallenge_Slides.pptx

Tools and Workflow

• Python
• pandas
• matplotlib
• Jupyter Notebook
• data validation and cleaning pipelines

How to Run

1. Open the notebook in src/DataChallenge_C1.ipynb
2. Run the cells in order from top to bottom
3. Review the cleaned outputs and generated plots
4. Use the analysis results for further exploration or reporting

Notes

This repository is intended to demonstrate end-to-end data analysis, data cleaning, and presentation-ready storytelling for an e-commerce operations challenge.