Real Data Analyst Portfolio project (Not Kaggle) | Python Pipeline + PostgreSQL + PowerBI Dashboard

A complete end-to-end Indian D2C E-Commerce Data Analytics project — from raw data generation to a 6-tab Power BI dashboard. By - Himanshu Gullaiya Not a Kaggle dataset. Not a tutorial follow-along. Built entirely from scratch. 📌 What's covered: → 125K orders · 13K customers · 80 products — generated from scratch, no Kaggle → 9 data quality issues deliberately injected & fixed (Medallion Architecture) → PostgreSQL Gold layer — 6 materialized views powering the dashboard → 6-tab Power BI dashboard with 30+ DAX measures → ₹6.8bn revenue · 46.8% margin · 22.88% return rate — real business insights → Cohort Retention · RFM Segmentation · Return Risk Detection · YOY/MOM Analysis 📊 GitHub (Full Code + SQL + Power BI): https://github.com/himanshugullaiya/D2C-ecommerce_analytics 🔗 Connect on LinkedIn: https://www.linkedin.com/in/himanshugullaiya/ ⏱ Timestamps: 00:00 - About the Project 00:54 - Process Flow & Medallion Architecture 01:24 - Data : Asymmetric Generation 03:57 - Data : Injecting 9 Data Quality Issues 04:52 - Data : Cleaning Methodology 06:06 - Loading Silver Data to PostgreSQL for Views 06:36 - Power BI Measures 06:56 - Tab 1 : Executive 09:41 - Tab 1 : Magic Insights 12:18 - Tab 2 : Products 15:06 - Tab 2 : Granular Comparison 17:06 - Tab 3 : Revenue (MOM & YOY) 21:41 - Tab 4 : Cohort Retention 24:52 - Tab 5 : RFM Segmentation 27:35 - Tab 5 : Risk Insight & Action Plan 28:55 - Tab 6 : Returns Analysis 32:42 - Tab 6 : Return Condition Tracking 36:03 - Conclusion & Thank You #dataanalytics #powerbi #python #sql #postgresql #dataanalyst #dashboarddesign #ecommerce #d2c #cohortanalysis #rfm #medallionarchitecture #dataengineering #pandas #dax #dataanalyticsportfolio #dataanalystportfolio #powerbidesktop #sqltutorial #pythonpandas