The Monsoon Machine Learning Paradigm
Loan
Underwriting
Models
Risk &
Revenue
Models
Utility
Models &
Engines
Lead Prioritization Models
Core Portfolio Management Model
Income Estimation Models
Pre-PD Models
Cross Sell Models
Early Warning Models
Policy Optimization Models
Application Scorecard
↓
↓
Pre-Approval Models
Pricing Model
Credit Limit Models
Collections Scorecard
Product Optimization Models
Loan
Underwriting
Models
Lead Prioritization Models
Pre-PD Models
Application Scorecard
Pricing Model
Risk &
Revenue
Models
Core Portfolio Management Model
Cross Sell Models
Early Warning Models
Credit Limit Models
Collections Scorecard
Utility
Models &
Engines
Income Estimation Models
Policy OptimizationModels
Pre-Approval Models
Product Optimization Models
Our Offerings
Loan
Underwriting
Models
Risk &
Revenue
Models
Utility
Models &
Engines
Delinquencies by 30%
Approvals by 25%
Turn Around Time by 40%
Monsoon’s machine learning powered underwriting models identify persistent patterns that enable us to predict future repayment behavior from borrower data (e.g. credit bureau data, bank statements, demographics, location related data and if available, alternate data). These patterns are unique to every lender based on the way they source borrowers, underwrite to borrowers and collect from borrowers. Our machine learning engine “learns” these patterns from every lender’s historic portfolio and application level data, combined with the repayment data associated with every borrower in the portfolio.
Lead Prioritization Models
Uses Sales/DSA level data such as basic demographic data, income, bureau data to filter good leads & reject low quality ones.
Reduces turn-around-time by upto 40%
Pre-PDModels
Looks at selected leads, buckets them according to risk,
applies appropriate credit policies and recommends either a Personal Discussion, or Straight Through Processing.
Reduces cost by upto 35% and optimizes turn-around-time
Application Scorecard
It looks at Bureau data, banking data and demographics to approve less risky loans while rejecting risky loans quickly.
Reduces delinquencies by upto 30% and improves approval rates by upto 25%.
Pricing Models
The Pricing engine enables you to price loans (interest rates, LTV, tenures on the basis of customer risk)
Optimizes ROI, tenures and other loan parameters on the basis of borrower risk.
Revenue by 35%
Credit Losses by 22%
Value At Risk by 33%
We use the credit underwriting models thus built to generate actionable recommendations and analytics for the client’s loan application portfolio. For ease of interpretation, we generate risk scorecards for each loan application that captures a quantitative risk score, the key aspects of the application from a risk perspective, a recommendation on the application and a set of red flags (if applicable) for that application.
All of this is designed to be available in real-time to sales officers, credit officers as well as risk managers to enable efficient loan processing at scale.
Core Portfolio Management Model
Looks at selected leads, buckets them according to risk, applies appropriate credit policies and recommends a Personal Discussion/Due Diligence or algorithmic processing
Cross Sell Models
Leverages bureau & repayment data to drive upto 35% more revenue via cross sell and retention
Limit Management Model
Collections Scorecard
Runs every month to prioritize collection efforts on stressed loans and optimizes cost & effort while reducing credit losses by upto 22%
Early Warning Models
The application scorecard looks at Bureau data, banking data, demographics & other data to approve less risky loans while rejecting risky loans quickly
Monsoon is happy to work with clients to tailor our offerings or build out new offerings to address their unique needs. Our technology is highly scalable and is built using architectural paradigms that enable easy customization, low latency and easy integration with external systems via APIs and SDKs.
Income Estimation Models
Quickly estimate income to drive better sourcing and underwriting
Policy Optimization Engine
Optimize Credit Policy dynamically to suit changing market conditions
Pre-Approval Models
Significantly raise your book-size with minimal risk by pre-approving upto 35% of your liability base for asset side products
Product Optimization Model
Optimize your asset side and liability products to address untapped market opportunities
How It's Delivered
Built Into Your System
Integrate our models with your existing LOS/LMS systems to achieve a seamless experience.
We can host our models on-premise & hand them over to you along with integration support.
(No data leaves your firewall)
On The Cloud
We expose a range of easy-to-integrate APIs hosted on any of the 3 major Cloud Providers.
We also support both Containerized Architectures and Server-less Architectures.
(No PII Data leaves your firewall)
Bespoke Consulting Solutions
We can make use of varied data sets to build custom models and solutions.
We also provide a consulting layer on top of the technology to make the best sense out of the data and the results.
You can optimize strategies and increase efficiency using our models and insights.