End-to-End-Funnel-Drop-Off-Return-and-Retention-Analysis

Analysed 400K orders to find a 45% cart drop-off, 14% COD cancellation rate & 46% 30-day retention | Python · Pandas · Jupyter

Comment

🛒 E-Commerce User Behavior & Retention Analysis

Live Demo - https://end-to-end-funnel-drop-off-return-and-retention-analysis-wkpam.streamlit.app/

📌 Project Overview

This project analyzes user behavior in an e-commerce setting using real-world datasets. The goal is to understand how users interact with the platform, what drives conversions, and how operational factors like delivery time impact customer satisfaction.

The analysis combines:

  • User session behavior (RetailRocket dataset)
  • Order, delivery, and review data (Olist dataset)

🎯 Objectives

  • Analyze user journeys from browsing to purchase (Funnel Analysis)
  • Measure the impact of delivery time on return rates
  • Explore customer behavior through sessions and interactions
  • Generate actionable business insights

📂 Datasets Used

1. RetailRocket Dataset

  • User interaction events (views, cart, purchases)
  • Session-level behavior tracking

2. Olist E-Commerce Dataset

  • Orders and order items
  • Delivery timestamps
  • Customer reviews (used as proxy for returns)

⚙️ Key Features & Analysis

🔹 1. Sessionization

  • Built user sessions using a 30-minute inactivity threshold
  • Identified user browsing patterns and engagement levels

🔹 2. Funnel Analysis (User Journey)

Tracked user flow across key stages:

  • View → Add to Cart → Purchase

📊 Metrics:

  • Conversion rates between each stage
  • Drop-off points in the funnel

💡 Insight:

  • Significant drop-off between browsing and purchase stages indicates potential UX or pricing friction

🔹 3. Delivery Time Impact Analysis

📦 Created a delivery_days feature:

delivery_days = delivery_date - order_date

Analyzed:

  • Return rate vs delivery time
  • Delivery delays and customer dissatisfaction

📊 Key Finding:

  • Longer delivery times correlate with higher return rates
  • Fast delivery significantly improves customer experience

🔹 4. Return Behavior Analysis

  • Used review scores as a proxy for returns
  • Flagged low ratings (≤2) as likely returns

💡 Insight:

  • Poor customer experience is strongly linked to returns

📊 Visualizations

The project includes:

  • Funnel bar charts (conversion stages)
  • Delivery time vs return rate charts
  • Distribution plots for delivery time

All visualizations are included in the notebook.

🧠 Key Insights

  • 🚚 Delivery speed is critical Orders delivered late show significantly higher return rates

  • 🛍️ Major drop-off in funnel Many users browse but don’t convert → optimization opportunity

  • ⭐ Customer experience drives retention Poor reviews strongly correlate with negative outcomes

🛠️ Tech Stack

  • Python
  • Pandas
  • NumPy
  • Matplotlib
  • Seaborn
  • Jupyter Notebook

▶️ How to Run

  1. Clone the repository:
git clone https://github.com/your-username/ecommerce-analysis.git
  1. Open the notebook:
jupyter notebook
  1. Run all cells to reproduce the analysis

📈 Future Improvements

  • Add cohort-based retention analysis (with correct user mapping)
  • Build predictive models for returns and churn
  • Create an interactive dashboard (Streamlit / Power BI)

Built With

Share this project:

Updates