# Pre-Harvest Underwriting Service

*/Opportunities/Pre-Harvest_Underwriting_Service*

## Opportunity Overview

**Wedge**: Begin with specialty crop lenders in California underwriting high-value assets like almonds and wine grapes. This niche faces acute intra-season weather volatility and holds the highest per-acre financial risk, driving immediate willingness to pay for precise predictions. Expand sequentially into broad-acre row crops in the Midwest, and finally cross-sell the proven models to federal crop insurance providers.
**Timing**: High-frequency multi-spectral satellite imagery is now commercially available at sub-meter resolution for pennies per acre. Simultaneously, foundational vision models accurately translate this raw biomass data into quantitative yield predictions without requiring bespoke localized training sets.
**Why This I C P**: Regional agricultural banks and crop insurance adjusters hold direct, unhedged financial risk tied to intra-season yield variation. They possess dedicated budgets for loan loss provisioning and manual field inspections, making the purchasing decision a straight substitution of existing spend.
**Size Of Prize**: Approximately 4,500 agricultural lending institutions and specialized insurance desks in the US spend around $60,000 annually on manual field appraisals and macro-level climate data subscriptions, creating an addressable market of roughly $270M for automated mid-season underwriting.
**Gap Narrative**: Agricultural lenders and crop insurers lack granular, field-level yield predictions during the growing season. Existing underwriting models rely on historical county averages, ignoring localized weather events and real-time crop health, which forces lenders to misprice risk until post-harvest audits occur.
**Defensibility**: Accuracy compounds systematically as the service ingests continuous ground-truth post-harvest yield data from early customers to calibrate its mid-season predictions. Once the underwriting API embeds into a bank's loan origination system, switching costs become prohibitive due to the rigorous compliance and back-testing requirements necessary to approve a new risk model.
**Why This Thesis**: Financial institutions require a conclusive risk score or predicted yield number to attach to a policy or loan, not a complex geospatial dashboard. A Service-as-Software approach delivers the finalized underwriting decision directly into their origination systems, bypassing the need for them to hire in-house agronomists or data scientists.

## Opportunity Linked Thesis

**Thesis**: [Service-as-Software](/Theses/Service-as-Software)

## Opportunity Linked I C P

**Icp**: [Agricultural Lender](/CompanyTypes/Agricultural_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**: ~$400-600M North American agricultural lenders
**S O M**: ~$15-30M
**T A M**: ~15,000 global agricultural lending institutions × ~$100k/yr ≈ ~$1.5B
**Growth Rate**: ~10-15%/yr, driven by increasing climate volatility and tightening regulatory scrutiny on agricultural credit portfolios
**Paid Comparable Spend**: ~$50k-200k/yr per institution on contracted field agronomist site visits, legacy weather data subscriptions, and manual loan appraisal labor

## Opportunity Incumbents

- [Farm Credit Services](/Products/Farm_Credit_Services) — Service
- [Conservis Ag](/Products/Conservis_Ag) — Tool
- [Excel Yield Projections](/Products/Excel_Yield_Projections) — Spreadsheet
- [Paper Ledger History](/Products/Paper_Ledger_History) — DIY
- [Traction Ag](/Products/Traction_Ag) — Tool
- [Local Agronomist Appraisals](/Products/Local_Agronomist_Appraisals) — Service
- [NAU Country Insurance](/Products/NAU_Country_Insurance) — Service

## Opportunity Win Conditions

**Kill Thresholds**:
- Human override rate exceeds 40 percent after 30 days of usage
- Integration and data mapping takes longer than 14 days per institution
- Prediction variance exceeds 10 percent against local agronomist appraisals
- Customer acquisition cost exceeds 15,000 dollars per deployed institution
**Leading Metrics**:
- Percentage of pre-harvest loans approved without a physical site visit
- Average hours from data ingestion to final loan decision
- Manual override rate by human loan officers
- Yield prediction variance percentage versus historical baseline
**What Proves Right**: Agricultural lenders route at least 20 percent of their pre-harvest loan applications through the service without requiring a physical agronomist site visit. Institutions issue loan approvals in under 48 hours, replacing the baseline process of three weeks. Pilot users sign 50,000 dollar annual contracts within 90 days after validating the prediction accuracy against their historical ledger data.
**What Proves Wrong**: Loan officers manually override the system outputs and dispatch field agronomists to verify crop conditions, destroying the margin advantage. Data ingestion from legacy platforms requires excessive manual mapping, pushing integration time beyond 30 days. Lenders abandon the service because the yield prediction variance exceeds their strict 5 percent risk tolerance threshold.

## Opportunity Build Profile

**Hardest Part**: Correlating sparse early-season satellite vegetative indices and localized weather patterns to final harvest yields with sufficient confidence to support direct financial risk.
**Min Viable Scope**: Limit v1 to underwriting a single row crop like corn in a specific, heavily mapped geographic region like the US Midwest. Deliberately exclude specialty crops, livestock, and complex multi-peril derivative products.
**Cold Start Problem**: No ground-truth harvest yield data exists at the field level to train the initial predictive models. Break this by partnering directly with a regional agricultural cooperative to access a decade of historical payout and yield data in exchange for free preliminary risk scoring.
**Time To First Value**: 2-4 days to process submitted field boundary files and deliver a binding underwriting decision
**Data Moat Available**: true
**Technical Difficulty**: High

## Neighborhood

### Where the gap lives

- [Agriculture, Forestry, Fishing and Hunting](/Industries/Agriculture,_Forestry,_Fishing_and_Hunting) — latent gap · Industries

### Incumbent in

- [Traction Ag](/Products/Traction_Ag) — incumbent in · Products
- [NAU Country Insurance](/Products/NAU_Country_Insurance) — incumbent in · Products
- [Paper Ledger History](/Products/Paper_Ledger_History) — incumbent in · Products
- [Conservis Ag](/Products/Conservis_Ag) — incumbent in · Products
- [Excel Yield Projections](/Products/Excel_Yield_Projections) — incumbent in · Products
- [Farm Credit Services](/Products/Farm_Credit_Services) — incumbent in · Products
- [Local Agronomist Appraisals](/Products/Local_Agronomist_Appraisals) — incumbent in · Products

### Applies thesis

- [Agricultural Lender](/CompanyTypes/Agricultural_Lender) — applies thesis · CompanyTypes

### Embodies

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

### Similar Opportunities

- [Agronomic Credit Engine](/Industries/Agriculture,_Forestry,_Fishing_and_Hunting/Opportunities/Agronomic_Credit_Engine) — similar · Opportunities
- [Yield Prediction Models](/Opportunities/Yield_Prediction_Models) — similar · Opportunities
- [Yield Prediction Models](/CompanyTypes/Identity_Preserved_Soybean_Producers/Opportunities/Yield_Prediction_Models) — similar · Opportunities
- [Decisioning as a Service](/Opportunities/Decisioning_as_a_Service) — similar · Opportunities
- [Farm Credit Underwriting](/Opportunities/Farm_Credit_Underwriting) — similar · Opportunities
- [Credit Decisioning Engine](/Opportunities/Credit_Decisioning_Engine) — similar · Opportunities
- [Block Yield Forecasting](/Opportunities/Block_Yield_Forecasting) — similar · Opportunities
- [AI Wildfire Underwriting for Commercial Insurers](/Opportunities/AI_Wildfire_Underwriting_for_Commercial_Insurers) — similar · Opportunities
- [Underwriting as a Service](/Opportunities/Underwriting_as_a_Service) — similar · Opportunities
- [Predictive Aquifer Yield Modeling](/Opportunities/Predictive_Aquifer_Yield_Modeling) — similar · Opportunities
- [AI Field Scouting for Farms](/Opportunities/AI_Field_Scouting_for_Farms) — similar · Opportunities
- [Automated Credit Underwriting](/CompanyTypes/Direct-to-Consumer_Mortgage_Lender/Opportunities/Automated_Credit_Underwriting) — similar · Opportunities
- [Parametric Claims Underwriter](/Occupations/Farming,_Fishing,_and_Forestry_Occupations/Opportunities/Parametric_Claims_Underwriter) — similar · Opportunities
- [Flood Risk Engine](/Opportunities/Flood_Risk_Engine) — similar · Opportunities
- [YieldGuard Vision](/Opportunities/YieldGuard_Vision) — similar · Opportunities
- [Trade Credit Underwriter](/Opportunities/Trade_Credit_Underwriter) — similar · Opportunities
- [Automated Subsidence Modeling](/Opportunities/Automated_Subsidence_Modeling) — similar · Opportunities
- [Vegetation Encroachment Monitoring](/Opportunities/Vegetation_Encroachment_Monitoring) — similar · Opportunities
- [Automated Canal Leak Detection](/Opportunities/Automated_Canal_Leak_Detection) — similar · Opportunities
- [Climate Risk Underwriter](/Opportunities/Climate_Risk_Underwriter) — similar · Opportunities