⚡QCommerce Intelligence Suite — Delivery & Customer Analytics

🚀 Executive Snapshot

• ⚡Swiggy Instamart: Fastest deliveries (11 min avg) → 85% satisfied customers.
• ⚡Blinkit: Market share leader (40% orders, ₹1.74 Cr revenue) but faces evening SLA breaches.
• ⚡JioMart: Highest basket value but 45% customer satisfaction, losing 70%+ to churn.

Quick-commerce is a billion-dollar battle where speed, service quality, and customer trust decide market winners.
This project delivers an end-to-end intelligence system for 100,000 realistic Q-commerce transactions (public dataset) across 3 platforms → covering descriptive, diagnostic, predictive, and prescriptive analytics.

🛠️ Tech Stack & Skills

• Python: Pandas, NumPy, Matplotlib, Seaborn, Plotly, Scikit-learn
• SQL (PostgreSQL): Star Schema design, Fact-Dimension modeling, ELT pipelines
• Forecasting Models: Prophet, ARIMA (time-series forecasting)
• ML Techniques: Logistic Regression (churn), Classification models (delay likelihood)
• Power BI: 3-page executive dashboard (Summary • Deep Dive • Ops)
• Analytics Framework: Descriptive → Diagnostic → Predictive → Prescriptive

📂 Repository Structure

📦 quick-commerce-analysis
 ┣ 📂 dataset
 ┣ 📂 notebooks
 ┃ ┣ 📓 Analysis.ipynb   # Full EDA & insights
 ┃ ┣ 📓 Extract.ipynb    # Data extraction & preprocessing
 ┃ ┗ 📓 Load.ipynb       # Loading into PostgreSQL
 ┣ 📂 sql                # SQL scripts for schema & transformations
 ┣ 📂 reports            # EDA storytelling deck
 ┣ 📂 dashboard          # Power BI dashboard
 ┣ 📄 README.md
 ┗ 📄 requirements.txt

📊 4-Stage Analytics Framework

🔹 Descriptive (What happened?)

• Orders, revenue, SLA performance by platform.
• Daily/hourly demand patterns, top categories.
• Categorizing customers into segments using RFM techniques

🔹 Diagnostic (Why did it happen?)

• Regression analysis (OLS) quantified impact of delivery delays → ratings drop -1.1 per 10 min delay.
• Blinkit breaches SLA 15% during 18:00–21:00.
• JioMart consistently lags with 40–50% delays.

🔹 Predictive (What’s likely?)

• 7-day demand forecast using Prophet/ARIMA.
• Churn prediction via Logistic Regression on customer ratings + refund history.
• Delay likelihood model across platforms.

🔹 Prescriptive (What should they do?)

• SLA optimization during evening peaks → protects ratings.
• Discounts for high-churn categories.
• Retention programs for “price-only” customers.

💡 Business Impact

• 📈 +37% revenue uplift possible via SLA + churn fixes.
• ⏱️ SLA breach reduction → loyalty & higher repeat orders.
• 🎯 Targeted retention campaigns could convert 50% “price-only” buyers → loyalists, lifting Blinkit revenue by 7.3%.

📑 View Full EDA Report PPT

📑 Deliverables

• ✅ SQL schemas + pipelines.
• ✅ Jupyter EDA + ML notebooks.
• ✅ Power BI dashboard.
• ✅ Storytelling PPT deck → EDA_Quick_Commerce.pptx.

🏆 Why This Project Stands Out

• 🔹 End-to-End pipeline: Raw data → SQL → EDA → Forecasting → Dashboard.
• 🔹 Industry KPIs: SLA, P50/P90 delivery, breach rates.
• 🔹 Business-first: Insights linked directly to growth opportunities.

👋 About Me

Hi, I’m Danish Shaikh passionate about turning data into business outcomes.
🔗 LinkedIn | Kaggle | GitHub

📜 License

This project is licensed under the MIT License - see the LICENSE file.