Project Overview: This project analyzes an A/B test dataset to determine whether a new landing page performs better than the old one. We derive insights for product and business decisions using statistical tests, user segmentation, and time-based trend analysis.
Key Objectives:
- A/B Testing & Conversion Rate Analysis
- Conducted an A/B test to compare conversion rates between control and treatment groups.
- Result: No statistically significant difference found (p-value ≈ 0.063).
- Time-based Trend Analysis
- Analyzed conversion trends over time to detect seasonal patterns or anomalies.
- Result: No major spikes/trends, suggesting uniform user behavior over time.
- User Segmentation
- Merged A/B test data with user country data to analyze conversion by region.
- Result: Identified regional differences in conversion rates, highlighting potential localization opportunities.
- Product Analysis & Business Insights
- Assessed whether user behavior varies based on country or landing page.
- Recommendation: Further testing is needed, especially region-specific optimizations.
Key Findings & Actionable Insights:
- The new landing page did not significantly outperform the old one, suggesting no need for immediate rollout.
- Some countries had higher conversion rates, indicating potential for regional personalization.
- No major time-based trends, meaning no seasonality effects on user engagement.
Next Steps:
- Conduct further A/B tests with localized page variations.
- Experiment with different CTA placements, messaging, and layouts for regional segments.
- Analyze additional factors (device type, user journey) to refine the conversion strategy.
(Kaggle link:https://www.kaggle.com/datasets/putdejudomthai/ecommerce-ab-testing-2022-dataset1/data)