Performance Metrics and Insights – A Data-Driven Approach to Supply Chain Optimization of the Look eCommerce
Key Findings
📈 Sales Growth: Jeans sales surged from $4.7K (2019) to $130.8K (2024); Outerwear led in gross profit (↑ from $2.78M to $71.84M).
🛒 Conversion Uplift: Website conversion rates rose from 0.09% (2019) to 8.25% (2023), reflecting improved funnel efficiency.
⚠️ Cart Abandonment: 26.7% of users abandon carts—checkout friction is a key opportunity.
📦 Operational Stability: Service levels improved (0.755 → 0.768) with greater consistency (std dev ↓ from 0.05 to 0.01).
🧩 No Demographic Impact: Age, gender, and country show no significant effect on cancellations (p > 0.05).
📅 Seasonal Peaks: Strong December spikes and summer demand (June–August) highlight timing for targeted campaigns.
Project Overview
This project delivers a comprehensive supply chain and operational analysis of The Look eCommerce, a synthetic dataset by Google’s Looker team. Using Python (Google Colab) and SQL (BigQuery), the analysis spans data exploration, cleaning, transformation, and KPI modeling across core tables: products, orders, users, events, and distribution_centers.
Key metrics evaluated:
- Sales & Profitability: Category performance, margin, trends
- Customer Experience: Delivery time, OTD, return rate
- Inventory Management: Turnover, fill rates, service level
- Marketing Effectiveness: Conversion funnels, session trends, Markov chains Advanced techniques include K-means, DBSCAN, and hierarchical clustering for customer segmentation, and Markov chain modeling to map user journeys. Navigation flows (home → department → product → cart) show high efficiency (100% transition), but cart-to-purchase conversion remains suboptimal.
Interactive Tableau Dashboards
Complementing the Colab analysis, a suite of four interactive Tableau dashboards visualizes key insights:
- Sales Metrics: Trends, category performance, profitability
- Customer Metrics: Delivery time, return rate, OTIF
- Marketing Metrics: Sessions, conversion funnel, traffic sources
- Inventory Management: Turnover, fill rates, processing time
These dashboards enable real-time exploration and support data-driven decision-making.
Actionable Recommendations
Optimize checkout UX to reduce cart abandonment (e.g., guest checkout, progress indicators) Leverage seasonal demand with targeted promotions in Q4 and summer Focus inventory planning on high-growth categories (Jeans, Outerwear) Improve fulfillment reliability in underperforming distribution centers Use behavioral (not demographic) data to predict cancellations This end-to-end project demonstrates how data analytics can drive operational excellence in eCommerce, from backend SQL/Python analysis to front-end Tableau visualization.