📊 Project Executive Summary: Amazon Product Sales & Insights

Data Scientist: Moniza Kidwai Domain: E-Commerce & Retail Analytics

---

🎯 Project Overview

The main objective of this project was to analyze Amazon product sales data to optimize business growth, pricing strategies, and customer satisfaction. Utilizing Python (Pandas, NumPy) alongside data visualization libraries (Matplotlib, Seaborn), a complete exploratory data analysis (EDA) pipeline was executed to extract actionable business intelligence.

---

🔑 Key Business Insights & Strategic Recommendations

📈 1. Category Performance & Inventory Optimization

• The Reality: The Electronics category (15.7 Million ratings) and Computers & Accessories (7.7 Million ratings) serve as the absolute primary growth drivers for the platform.
• The Problem Area: Niches like Toys & Games, Health & Personal Care, and Car & Motorbike suffer from a severe lack of assortment, with only 1–2 products listed.
• Business Action:
  • Maintain high safety stock thresholds (>20%) in warehouses for top-performing Electronics and Computer items to prevent stockouts.
  • Actively onboard new vendors in low-performing categories to aggressively expand the product catalog and assortment.
  • Launch targeted Retargeting Ads to bring back previously bounced customers by showcasing the newly expanded catalog.

💰 2. Price Elasticity & Consumer Behavior

• The Reality: There is a weak negative correlation (-0.0271) between a product's price and its sales volume. The lower price tier (₹0 - ₹5,000) represents a massive volume driver where consumer traction is dense.
• The Exception: High-value premium products (e.g., smartphones and premium laptops) bypass this trend, generating substantial revenue despite lower absolute unit volumes due to high brand loyalty.
• Business Action:
  • Prioritize budget-friendly items in promotional banners and algorithmic recommendations to boost click-through rates (CTR) and impulse purchases.
  • Introduce financial flexibility options (e.g., No-Cost EMI, banking credit card discounts) specifically for premium tiers to lower the barrier to entry.

📉 3. Discount Psychology vs. Customer Retention

• The Reality: A consistent downward trend in product ratings is observed as discount percentages reach extreme levels:
  • Low Discounts (0-20%): 4.15 Average Rating
  • Extreme Discounts (80-100%): 3.97 Average Rating
• The Catch: The negative correlation (-0.1553) validates that extreme discounts either trigger skepticism regarding product authenticity or indicate that vendors are liquidation-dumping subpar stock.
• Business Action:
  • Focus marketing campaigns around the "Sweet Spot" range of 40% - 60% discount, where both sales volume and customer satisfaction ($4.08+$ rating) remain optimally balanced.
  • Deploy the Quality Control (QC) team to run an automated audit on the 59 products within the 80-100% discount tier to weed out defective items and protect marketplace integrity.

⭐ 4. Amazon's Superstar Products

• The Reality: The top 10 superstar products (ranked by highest review velocity with a rating $\ge 4.0$) are entirely dominated by the Electronics vertical—specifically Amazon Basics HDMI Cables, boAt Bassheads Earphones, and Redmi budget smartphones.
• Business Action:
  • Cross-Selling Strategy: Bundle these high-velocity, low-cost "superstars" as automated recommendations ("Frequently Bought Together") on the checkout pages of high-value electronic items like TVs and Laptops.

---

🛠️ Tech Stack Used

• Language: Python
• Libraries: Pandas, NumPy, Matplotlib, Seaborn
• Environment: Jupyter Notebook