🛒 Zepto Products Analysis using Python

This project performs an inventory data analysis on product listings from Zepto using Python. The analysis includes data cleaning, exploratory data analysis (EDA), and visualizations to uncover insights related to product pricing, discounts, and stock availability.

---

📁 Dataset Overview

The dataset contains detailed product-level data, including:

• Category
• Product name
• MRP and Discounted Price (in paise)
• Discount Percent
• Available Quantity
• Weight (in grams)
• Stock Status (In Stock or Out of Stock)
• Quantity per pack

---

🧹 Data Cleaning (using pandas)

Key steps in data cleaning:

• Checked shape, info, and null values using .shape, .info(), .isnull().sum()
• Removed rows with MRP = 0 and weight = 0
• Converted columns from paise to rupees
  • mrp, discountedSellingPrice = divided by 100
• Identified and removed duplicate categories like:
  • Personal Care = Paan Corner
  • Cooking Essentials = Munchies
  • Ice Cream & Desserts = Chocolates & Candies
  • Dairy, Bread & Butter = Beverages
• Dropped exact duplicates using .drop_duplicates()

---

📊 Exploratory Data Analysis (using matplotlib)

1️⃣ Top 10 products with highest discounts

• Bar chart showing product names vs. discount %

2️⃣ Stock status analysis

• Pie chart showing count of products that are in stock vs out of stock

3️⃣ Products with high MRP but low weight

• Bar chart identifying luxury/small items (like cosmetics)

4️⃣ Cheapest products after applying discount

• Bar chart of final price (in ₹)

5️⃣ Relationship between Discount % and Final Price

• Line chart showing trend between discount % and average final price

---

🗝️ Key Insights

This analysis revealed that while a majority of products are well-stocked, heavy discounts are mostly offered on low-cost everyday ready-to-eat items like wafers and liquid masalas to attract more customers. On the other hand, luxury or premium products like saffron and skincare items have a significantly higher price per gram, indicating niche value. The relationship between discount percentage and final price is not linear—high discounts do not always mean high-value savings, as they are often applied to lower-priced products.

---

📌 Tools Used

• Python
• Pandas – data cleaning and transformation
• Matplotlib – data visualization
• Jupyter Notebook – code and analysis

---

Visualization📈

✅ Final Output

This project helped explore:

• Discount patterns
• Inventory stock status
• Product pricing behavior
• Data cleaning on real-world messy data

---

💡 How to Run

1. Clone this repo
2. Open the Jupyter Notebook
3. Install required packages:

   pip install pandas matplotlib

4. Run all cells to see the full analysis