Project Overview
Analyzed three pricing strategies for Domestic & General — ASIS, @22%, and @23% — to evaluate their impact on conversion rates, revenue, and fairness. The goal was to identify the most effective approach for balancing customer acceptance with business performance.
Methodology
Performed EDA to explore patterns, created summary tables of premiums and conversions, and visualized price differences. Measured price elasticity of demand across and within groups. Applied statistical tests (Chi-square, Mann-Whitney, Kruskal-Wallis) to check significance.
Key Findings
- ASIS: highest conversion (23.3%).
- @22%: highest revenue per offer, despite lower conversion.
- Elasticity: ~ -0.43 for @22% and @23%, showing conversion drops with price increases.
- Claims impact: frequent claimants more likely to reject offers.
- Evidence of bias between @22% and @23%.
Recommendations
Adopt a hybrid approach: use @22% where revenue gains outweigh conversion losses, and ASIS/@23% for high-risk groups. Incorporate claims history and product metadata for more personalized pricing.
Skills & Tools
Python (Pandas, NumPy, Seaborn, Matplotlib, Statsmodels, SciPy) · EDA · Statistical Testing · Business Analytics · Price Elasticity Modeling