Skip to content

VaibhaviDavee/Funnel_Drop-Off_Analysis

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Funnel Drop-Off Analysis - E-commerce Conversion Optimization

Python | Pandas | Matplotlib | Seaborn | SciPy


📌 Overview

This project analyzes user behavior across an e-commerce funnel to identify where users drop off and how it impacts overall conversion.

Out of 5,000 users, only 1,010 completed a purchase, resulting in a 20.2% conversion rate.
The goal is to pinpoint the exact stage where users drop and understand the underlying reason using data and statistical testing.


🚨 Key Insight

The most critical drop (~60%) occurs at:
👉 Product Page → Cart

Users are actively spending time on product pages but are not adding items to the cart.
This indicates a conversion problem at the product page, not an issue with traffic or checkout.


📊 Analysis Performed

  • Funnel stage drop-off analysis
  • Conversion rate calculation
  • Device-wise comparison
  • Referral source analysis
  • Time-based behavior trends
  • Product page engagement analysis
  • Statistical testing using Chi-square

🧪 Key Findings

  • Device type → ❌ No significant impact (p = 0.98)
  • Referral source → ❌ No significant difference (p = 0.47)
  • Time of visit → ❌ No clear pattern
  • Engagement time → ⚠️ Same for buyers & non-buyers

👉 Conclusion:
The issue is not who is coming or when they visit —
it is how effectively the product page converts interest into action.


💡 Business Impact

  • ~2,388 users drop at the product page stage
  • Recovering just 10% of these users
    → can generate ~$11,900 additional revenue per 5,000 sessions

This makes the Product Page → Cart transition the highest impact optimization point.


⚠️ Challenges Faced

  • Tracking user journey across funnel steps accurately
  • Ensuring data consistency and correct session mapping
  • Applying and interpreting statistical tests (Chi-square) correctly
  • Avoiding misleading conclusions from visual patterns
  • Identifying limitations due to synthetic dataset (uniform behavior across segments)

🛠️ Tech Stack

  • Python
  • Pandas, NumPy
  • Matplotlib, Seaborn
  • SciPy (Statistical Testing)
  • Jupyter Notebook

▶️ How to Run

git clone https://github.com/VaibhaviDavee/funnel-drop-analysis.git
cd funnel-drop-analysis
pip install -r requirements.txt
jupyter notebook

💡 Conclusion

This analysis highlights that conversion is not limited by traffic volume, device type, or checkout efficiency.

The real bottleneck lies at a single, high-impact stage — the Product Page → Cart transition, where user intent fails to convert into action.

Despite users actively engaging with product pages, the lack of persuasive design, clarity, or trust signals leads to significant drop-offs.

👉 This makes the product page the most critical optimization point in the entire funnel.

Even small improvements at this stage can drive disproportionately high revenue gains, making it the most valuable area for business focus.

In short, the problem is not attracting users —
it is converting interested users into buyers.

About

Finding where users drop — and what it costs the business.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors