# Credit Decisioning Engine

*/Opportunities/Credit_Decisioning_Engine*

## Opportunity Overview

**Wedge**: Start with auto and personal loans at mid-tier credit unions, where underwriting volume is high but individual loan sizes keep the automated risk tolerable. Win this niche because credit unions urgently need to approve younger members but currently reject them due to thin credit histories. Expand outward by moving into higher-complexity asset classes like commercial SMB lending and eventually residential mortgages.
**Timing**: LLMs with large context windows now parse multi-page, unstructured financial documents with high accuracy and trace reasoning back to specific line items. This capability satisfies stringent regulatory requirements for adverse action notices, which black-box machine learning models previously failed to clear.
**Why This I C P**: Alternative lenders and credit unions face immense pressure to grow loan portfolios but lack the in-house data science teams of Tier 1 banks. They are highly motivated to adopt off-the-shelf automated decisioning to capture younger, thin-file demographics without increasing human underwriting headcount.
**Size Of Prize**: There are roughly 11,000 mid-market lending institutions in the US, including credit unions, community banks, and alternative non-bank lenders. At an average annual spend of $80,000 per institution for underwriting software and manual review labor, the addressable prize is roughly $880M annually.
**Gap Narrative**: Mid-market lenders and credit unions rely on rigid rule-based underwriting systems that reject thin-file applicants and require manual review for edge cases. They need an engine that ingests unstructured alternative data, such as bank statements and gig income receipts, alongside traditional credit files to render instant, explainable credit decisions.
**Defensibility**: The system builds a proprietary feedback loop of alternative data features correlated with actual loan performance across a fragmented market. As the engine processes more edge-case loans, its default prediction models become more accurate than any single lender's internal models, creating a strong data network effect. Furthermore, switching costs become insurmountable once the engine is hardwired into the lender's core origination APIs.
**Why This Thesis**: A Service-as-Software approach directly replaces the manual underwriter's workflow of cross-referencing documents. By delivering a fully automated decision rather than just a software dashboard, the product captures the full value of the labor replacement while integrating directly into the lender's existing loan origination system.

## Opportunity Linked Thesis

**Thesis**: [Software](/Theses/Software)

## Opportunity Linked I C P

**Icp**: [Alternative Lender](/CompanyTypes/Alternative_Lender)

## Opportunity Market Sizing

_Illustrative — target and order-of-magnitude estimate figures, not an achieved track record (this Thing is concept-stage)._

**S A M**: ~$600M-900M US and EU mid-tier consumer and SME alternative lenders
**S O M**: ~$15M-35M
**T A M**: ~30,000 global alternative lending firms × ~$60,000-80,000/yr average decisioning software spend ≈ ~$1.8B-2.4B
**Growth Rate**: ~14-19%/yr, driven by the expansion of embedded finance and rising default rates requiring tighter underwriting controls
**Paid Comparable Spend**: ~$100,000-300,000/yr on legacy rules engines, fragmented alternative data API subscriptions, and manual underwriting labor

## Opportunity Incumbents

- [Experian PowerCurve](/Products/Experian_PowerCurve) — Tool
- [FICO Origination Manager](/Products/FICO_Origination_Manager) — Tool
- [Provenir Decisioning](/Products/Provenir_Decisioning) — Tool
- [Zoot Enterprises](/Products/Zoot_Enterprises) — Service
- [In-House Rules Engine](/Products/In-House_Rules_Engine) — DIY
- [Internal Excel Models](/Products/Internal_Excel_Models) — Spreadsheet

## Opportunity Win Conditions

**Kill Thresholds**:
- Integration time exceeds 60 days for mid-tier lenders
- Human escalation rate remains above 75% after 30 days in production
- Sales cycle exceeds 120 days to secure a $60,000 ACV contract
- Less than 20% of pipeline prospects agree to run historical data backtests
**Leading Metrics**:
- Time from pilot kickoff to first production decision
- Percentage of automated loan approvals versus manual escalations
- API response latency per credit decision
- Alternative data source mapping completion rate
- Compliance audit export frequency
**What Proves Right**: Mid-tier alternative lenders deploy the engine into production alongside legacy systems within 30 days. Pilot users migrate at least 40% of their origination volume to the platform and achieve a 25% reduction in manual underwriter escalations. Early adopters convert from pilots to paid contracts starting at $60,000 annually.
**What Proves Wrong**: Risk officers refuse to trust the outputs and mandate manual review for 100% of processed loan applications. Integration cycles stretch beyond 90 days due to incompatible core banking systems or bespoke alternative data feeds. Prospects churn back to internal Excel models because the engine fails to provide sufficient audit trails for compliance teams.

## Opportunity Build Profile

**Hardest Part**: Normalizing heterogeneous financial data from raw bank feeds and messy accounting ledgers into a uniform schema for underwriting. A single miscategorized transaction alters the risk profile and causes immediate capital misallocation.
**Min Viable Scope**: A pure rules-based engine automating initial pass/fail filters for B2B revenue-based financing using Plaid and Codat integrations. Deliberately exclude machine learning risk scoring, complex multi-tier human approval workflows, and consumer credit bureau connections.
**Cold Start Problem**: The engine lacks historical default data to calibrate risk thresholds on day one. Break this by running backtests on a single mid-market lender's historical loan book to validate the rule logic against known outcomes.
**Time To First Value**: 2-4 weeks of onboarding, gated by mapping the lender's proprietary credit policy into the engine's logic layer.
**Data Moat Available**: true
**Technical Difficulty**: High

## Neighborhood

### Where the gap lives

- [Furniture, Home Furnishings, Electronics, and Appliance Retailers](/Industries/Furniture,_Home_Furnishings,_Electronics,_and_Appliance_Retailers) — latent gap · Industries
- [Other Miscellaneous Retailers](/Industries/Other_Miscellaneous_Retailers) — latent gap · Industries
- [Inbound Payment Cycle Time](/Metrics/Inbound_Payment_Cycle_Time) — latent gap · Metrics

### Incumbent in

- [Provenir Decision Engine](/Products/Provenir_Decision_Engine) — incumbent in · Products
- [In-House Excel Models](/Products/In-House_Excel_Models) — incumbent in · Products
- [FICO Origination Manager](/Products/FICO_Origination_Manager) — incumbent in · Products
- [In-House Rules Engine](/Products/In-House_Rules_Engine) — incumbent in · Products
- [Zoot Enterprises](/Products/Zoot_Enterprises) — incumbent in · Products
- [Experian PowerCurve](/Products/Experian_PowerCurve) — incumbent in · Products

### Applies thesis

- [Alternative Lender](/CompanyTypes/Alternative_Lender) — applies thesis · CompanyTypes

### Embodies

- [Software](/Theses/Software) — embodies · Theses

### Similar Opportunities

- [Decisioning as a Service](/Opportunities/Decisioning_as_a_Service) — similar · Opportunities
- [Trade Credit Underwriter](/Opportunities/Trade_Credit_Underwriter) — similar · Opportunities
- [Underwriting as a Service](/Opportunities/Underwriting_as_a_Service) — similar · Opportunities
- [Credit Node](/Opportunities/Credit_Node) — similar · Opportunities
- [Automated Credit Underwriting](/CompanyTypes/Direct-to-Consumer_Mortgage_Lender/Opportunities/Automated_Credit_Underwriting) — similar · Opportunities
- [AI Underwriting for Private Credit](/Opportunities/AI_Underwriting_for_Private_Credit) — similar · Opportunities
- [AI Debt Underwriting for Private Credit](/Opportunities/AI_Debt_Underwriting_for_Private_Credit) — similar · Opportunities
- [Automated Credit Risk Modeling](/CompanyTypes/Commercial_Bank/Opportunities/Automated_Credit_Risk_Modeling) — similar · Opportunities
- [Farm Credit Underwriting](/Opportunities/Farm_Credit_Underwriting) — similar · Opportunities
- [Automated Model Validator](/Opportunities/Automated_Model_Validator) — similar · Opportunities
- [Trade Credit Underwriter](/Industries/Wholesale_Trade/Opportunities/Trade_Credit_Underwriter) — similar · Opportunities
- [Pre-Harvest Underwriting Service](/Opportunities/Pre-Harvest_Underwriting_Service) — similar · Opportunities
- [Real-Time Credit for Apparel Showrooms](/Opportunities/Real-Time_Credit_for_Apparel_Showrooms) — similar · Opportunities
- [Asset Quality Forecasting](/Opportunities/Asset_Quality_Forecasting) — similar · Opportunities
- [Automated Tenant Underwriting](/Opportunities/Automated_Tenant_Underwriting) — similar · Opportunities
- [Agronomic Credit Engine](/Industries/Agriculture,_Forestry,_Fishing_and_Hunting/Opportunities/Agronomic_Credit_Engine) — similar · Opportunities
- [Autonomous Credit Factor](/Opportunities/Autonomous_Credit_Factor) — similar · Opportunities
- [Progress Billing Verification for Banks](/Opportunities/Progress_Billing_Verification_for_Banks) — similar · Opportunities
- [Litigation Underwriting Engine](/Occupations/Legal_Occupations/Opportunities/Litigation_Underwriting_Engine) — similar · Opportunities
- [Illiquid Asset Valuation](/Opportunities/Illiquid_Asset_Valuation) — similar · Opportunities