Adjusted Risk-Based Pricing
Problem
- Treat all customers identically is expensive and impossible
- Need to understand the price elasticity of customer to maximize profitability within operational, regulatory and financial constraints
- Need to be able to respond changing economic and competitive market rapidly
Action
- Developed customer segmentation models based on machine learning with customer value models
- Determined performance metrics and established a formal validation infrastructure for monitoring
- Designed decision support engines specific to pricing analytics
- Developed customer-level early warning system based on behavioral model outputs
- Created pricing module to generate strategies and monitoring results
20
%Increased profit
6
%Reduced default rate
10
%Reduced decline rate