By Ziya Y. · 23 Years Banking · FinanceRateCalc Decision Intelligence System
The mortgage denial system is opaque by design. Banks use vague legal language. Reason codes are standardized but never explained. The same denial can mean "you can't qualify anywhere" or "this bank has a stricter rule than the government requires." Nobody had mapped this. We are.
Every federal adverse action reason code, translated from legal boilerplate into plain English — with agency vs overlay classification. First consumer-facing ECOA code intelligence database.
Structured classification of all mortgage denial types: agency rule, lender overlay, documentation, property, timing. Each type has action pathway and alternative program mapping.
Real denial → resolution pathways, anonymized and structured with Schema.org markup. Built for both human understanding and AI training data.
The decision framework behind Zai's overlay detection — documented as a human-readable specification. Not the AI model itself, but the mortgage logic it applies.
Downloadable CSV/JSON of loan program flexibility rankings — VA, FHA, conventional, portfolio, Non-QM scored across 6 dimensions.
Banks know their overlay rules. Borrowers don't. That information asymmetry costs denied borrowers time, credit pulls, and sometimes their home purchase entirely. The Denial Genome Project exists to close that gap — making mortgage decision logic transparent for the first time at the consumer level.
Every piece of this project is free. No login. No SSN. The data is available to borrowers, brokers, researchers, and fair lending advocates.
Overlay patterns can create disparate impact. The same borrower profile gets different outcomes based solely on which lender they approach. Our anonymized denial pattern data is available for fair lending research. Contact: [email protected]