Goals
- Analyze pricing and discount patterns across Amazon product categories
- Identify which categories are most discounted, most expensive, and best rated
- Understand the relationship between price, discount, and customer ratings
- Find the top performing products by reviews and ratings
Process
- Cleaned price columns by removing ₹ and comma symbols and converting to float
- Extracted main category from nested category strings using string splitting
- Added two engineered features — savings (actual minus discounted price) and discount tier (binned into 5 ranges)
- Built 19 visualizations across 5 sections — category, price, rating, discount, and product analysis
Insights
- Electronics is the core revenue driver — highest product count (526) and highest average price (₹10,127)
- Office Products has the highest average rating (4.31) with nearly zero discounting — suggesting loyal professional buyers who don't need price incentives
- Discount percentage has near-zero correlation with rating (-0.16) — sellers cannot buy better ratings through higher discounts
- AmazonBasics dominates the most reviewed products with 426,973 reviews — the in-house brand has captured massive customer trust for basic accessories
- The 41–60% discount tier is the sweet spot — 508 products fall here, suggesting Amazon and sellers have converged on this as the optimal pricing strategy
- Cables and accessories are discounted up to 94% — likely used as loss leaders to drive traffic to product pages