Comprehensive analysis of the Olist Brazilian E-Commerce dataset (100,000+ orders from 2016-2018). The goal was to transform raw transactional data into actionable business insights regarding revenue drivers, customer segmentation, and logistical performance.
View Interactive Tableau Dashboard
- Data Processing: Python (Pandas, SQLAlchemy)
- Database: SQLite, Google BigQuery
- Cloud: Google Cloud Platform
- Visualization: Tableau Public
- Environment: Jupyter Notebook
- Revenue Concentration: Credit cards are the primary revenue driver, capturing 78% of total R$16M revenue
- Geographic Dominance: São Paulo (SP) has 3x higher customer density than any other state
- Logistics Friction: AL and MA show disproportionately high late delivery rates relative to their customer base size
- Product Performance: Bed, Bath & Table and Health & Beauty are the core pillars of GMV
- Data Extraction: Queried and joined 8 relational tables using SQL
- Analysis: Wrote 10+ business-driven SQL queries covering revenue, logistics, customer behavior and seller performance
- Documentation: Documented findings with business recommendations in Jupyter Notebook
- Visualization: Built interactive dashboard in Tableau Public
Olist Brazilian E-Commerce Dataset via Kaggle
Yugal Jagtap