01 / 04
The visible economy
Markets transmit prices continuously. Logistics systems locate goods. Sensors watch factories. Land, by contrast, is often assessed with intermittent samples and incomplete records.
The intelligence
beneath our feet.
What if the next trillion-dollar economic opportunity begins with understanding the living world under our feet?
SCROLL TO EXPLORE ↓ACT I / THE BLIND SPOT
We mapped the markets. We barely mapped the soil.
We know the price of almost everything above ground. Beneath our feet, the biological machinery of productivity remains hard to observe.
02 / 04
A living system, not an inert asset
Soil is a dynamic interaction among minerals, water, air, roots, microbes, organic matter and management. No single measurement captures its condition.
03 / 04
The wrong abstraction
A field is often treated as acreage plus yield history. But two fields with similar yields may have very different resilience, water behavior and future costs.
04 / 04
The inversion
What if the most valuable land data were not its location or last harvest, but how it responds to intervention?
INTERACTIVE / SIX DIMENSIONS
Soil is not a single score.
LAYER 1 / 6
Biology
Microbial communities, roots, decomposition and nutrient cycling. Genetic potential is not the same as measured activity.
Soil intelligence must preserve context across all six dimensions.GO DEEPER / FOLLOW THE EVIDENCE
The soil is an ecosystem, not an ingredient list.
Texture, pores, organisms, roots, moisture and chemistry interact; an isolated indicator cannot describe the system’s behavior.
THE CAUSAL CHAIN
- 01
Management changes the physical environment and the availability of resources.
- 02
Water, oxygen and roots constrain microbial processes and nutrient transport.
- 03
Organisms transform organic matter and nutrients at different rates.
- 04
Plants respond through growth, stress and yield under local weather conditions.
What would count as evidence?
- Paired chemical, physical and biological assays with documented methods
- Repeated samples across depth, season and management history
- Yield and economic outcomes tied to the same geospatial units
What could break the thesis?
- Measurements are too variable or costly to change practice
- Conventional agronomy explains the outcome just as well
QUESTIONS TO TAKE INTO THE FIELD
Follow the next connection →Which soil functions are both measurable and actionable?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
What sampling resolution changes a real decision?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
Which properties remain stable across seasons?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
THE REFRAME / 01
The asset is not the dirt. It is the capacity of a living system to respond.
ACT II / THE CONVERGENCE
A new instrument for the invisible.
Remote sensing, spectroscopy, field assays, computational biology and AI are beginning to connect previously isolated observations.
01 / 04
See from above
Satellite and aerial imagery can reveal vegetation and moisture-related patterns, but rarely measure subsurface biology directly.
02 / 04
Sample below
Chemical tests, physical measurements, DNA assays and microbial functional indicators provide richer evidence—at a cost.
03 / 04
Model between observations
Multimodal models may infer changes between sparse measurements, but require local calibration and honest uncertainty.
04 / 04
From a snapshot to a hypothesis
The leap is not a prettier dashboard. It is a testable prediction: what happens if management changes here, this season?
GO DEEPER / FOLLOW THE EVIDENCE
The measurement problem comes before the model.
A sophisticated prediction cannot repair unknown sampling bias, incompatible laboratory protocols or missing provenance.
THE CAUSAL CHAIN
- 01
Define a soil unit, depth, timestamp and collection protocol.
- 02
Collect independent observations and record uncertainty.
- 03
Normalize across instruments and laboratory methods without erasing differences.
- 04
Test repeatability before using observations as model labels.
What would count as evidence?
- Replicate and blinded sampling
- Inter-lab comparison and calibration records
- Explicit missing-data and measurement-error models
What could break the thesis?
- Measurement noise overwhelms treatment effect
- Costs exceed incremental decision value
QUESTIONS TO TAKE INTO THE FIELD
Follow the next connection →What is the minimum trustworthy observation package?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
Who certifies the method and its uncertainty?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
Can low-cost proxies replace expensive assays?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
THE REFRAME / 02
More measurements are not the breakthrough. Better decisions are.
ACT III / THE DECISION
The soil report is not the product.
Information becomes valuable only when it changes a decision and produces an outcome someone can verify.
01 / 04
Different actors, different questions
A farmer asks about profit and resilience. A lender asks about downside risk. A buyer asks about supply reliability. A scientist asks what is causal.
02 / 04
The intervention
Cover cropping, irrigation, tillage, fertilization and biological amendments are candidate decisions—not universal prescriptions.
03 / 04
The evidence loop
Observe → model → recommend → intervene → measure → learn. Each turn should improve both the model and the decision.
04 / 04
The economic test
Does the decision improve outcomes beyond current agronomic advice, after accounting for measurement, operational costs and uncertainty?
INTERACTIVE / CHANGE PERSPECTIVE
Same soil. Different decisions.
THE DECISION-MAKER / FARMER
“Which management change is most likely to improve my margin?”
Evidence needed: Net return after input, labor, sampling and transition costs.
GO DEEPER / FOLLOW THE EVIDENCE
A soil foundation model must earn its generality.
A model trained in one region may fail when climate, soil texture, crops or management regimes change.
THE CAUSAL CHAIN
- 01
Fuse imagery, weather, terrain, management and laboratory evidence.
- 02
Pretrain representations while preserving location and time context.
- 03
Test against mechanistic and conventional agronomic baselines.
- 04
Measure uncertainty and out-of-distribution failure on withheld geographies.
What would count as evidence?
- Geographically and temporally held-out evaluation
- Ablations for every expensive data modality
- Calibration and error analyses for minority soil types
What could break the thesis?
- Local models outperform the foundation model at lower cost
- Uncertainty is too high for high-stakes recommendations
QUESTIONS TO TAKE INTO THE FIELD
Follow the next connection →Which tasks genuinely benefit from shared representations?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
Where must mechanistic science constrain machine learning?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
How will the model admit ignorance?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
THE REFRAME / 03
A soil model matters only if it changes an outcome.
ACT IV / VALUE MIGRATION
Who captures the value of healthier land?
Soil intelligence could affect multiple markets, but ecosystem value is not the same as software revenue.
01 / 04
On-farm economics
Reduced inputs, more stable yields and better water management are possible sources of value—but must be demonstrated field by field.
02 / 04
Risk and finance
Verified soil condition might improve underwriting and land diligence if it adds predictive power beyond existing risk models.
03 / 04
Food and sourcing
Buyers may reward reliable supply and verifiable practices. Nutritional and health claims demand a separate, higher standard of evidence.
04 / 04
The platform question
Can trusted, permissioned observations be reused across buyers, financiers and researchers without taking control away from growers?
INTERACTIVE / VALUE CAPTURE
How big could the economics become?
HYPOTHETICAL ANNUAL ECONOMICS
Value unlocked
$4.00BPlatform revenue
$400.0MScenario only, not a market forecast. Assumes incremental realized value; excludes deployment and measurement costs, churn, and overlapping claims.GO DEEPER / FOLLOW THE EVIDENCE
Where does value move when land becomes legible?
Better information only creates a market when it changes costly choices and someone can capture a fair share of the improvement.
THE CAUSAL CHAIN
- 01
Identify a decision with substantial economic downside.
- 02
Estimate the value of improved information relative to today’s baseline.
- 03
Deliver a recommendation inside the decision workflow.
- 04
Verify realized gains net of measurement, adoption and operating costs.
What would count as evidence?
- Prospective comparisons against business-as-usual
- Net farm-margin impact and sensitivity analysis
- Documented willingness to pay by decision owner
What could break the thesis?
- Benefits accrue to actors unwilling to pay
- Deployment costs erase the theoretical improvement
QUESTIONS TO TAKE INTO THE FIELD
Follow the next connection →Who experiences the benefit and who signs the contract?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
What is the smallest profitable decision loop?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
Does the business improve farmer bargaining power?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
THE REFRAME / 04
Economic value is not the same as capturable platform revenue.
ACT V / THE FRONTIER
A foundation model for living land?
The deep-tech wager is that models can learn transferable representations of soil dynamics, while admitting what they do not know.
01 / 04
Representation
Combine soil properties, management events, climate, crop history, terrain and biological assays in a spatiotemporal model.
02 / 04
Generalization
Can knowledge learned in one soil type, crop or climate transfer to another? Domain shift may defeat naïve foundation-model claims.
03 / 04
Causality
Predictions are not explanations. Intervention studies, matched comparisons and prospective validation are essential.
04 / 04
Food is medicine?
The most ambitious bridge—from soil practices to crop composition to diet to health—must be validated link by link, not assumed.
GO DEEPER / FOLLOW THE EVIDENCE
Food Is Medicine is a chain of separate hypotheses.
Changes in soil do not automatically imply clinically meaningful changes in human health.
THE CAUSAL CHAIN
- 01
Test whether agricultural intervention changes soil function.
- 02
Measure whether that change alters edible crop composition.
- 03
Establish actual consumption, bioavailability and exposure.
- 04
Evaluate clinical outcomes with appropriate study designs and confounder controls.
What would count as evidence?
- Controlled agronomic comparisons
- Independent compositional and dietary assessments
- Human studies designed for meaningful clinical endpoints
What could break the thesis?
- Crop differences disappear after storage and preparation
- Observed nutrient changes do not translate to health outcomes
QUESTIONS TO TAKE INTO THE FIELD
Follow the next connection →Where is the weakest causal link?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
Which outcome is worth testing first?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
How can claims avoid outrunning evidence?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
THE REFRAME / 05
Prediction is not causality. Uncertainty is part of the product.
ACT VI / EARNED IGNITION
Don’t found the company. First, earn the evidence.
The venture thesis should remain in Subzero until the science, customer need and economic mechanism survive deliberate attempts to disprove them.
01 / 04
One crop. One region. One decision.
Start where the cost of a wrong decision is meaningful and reliable outcomes can be measured within a practical timeframe.
02 / 04
Compare against the boring baseline
Conventional soil tests and trusted agronomists are the benchmark. A sophisticated model that cannot beat them has not earned its place.
03 / 04
Prospective proof
Pilot with real growers, track costs, uncertainty and incremental outcomes, and publish what fails as carefully as what succeeds.
04 / 04
The threshold
Form the venture when repeatable economic value, defensible evidence and a credible buyer converge—not because the narrative is exciting.
THE COUNTERARGUMENT
What would prove us wrong?
GO DEEPER / FOLLOW THE EVIDENCE
Subzero: protect the idea from premature certainty.
The objective of early exploration is not to defend a company concept. It is to discover whether the evidence warrants forming one.
THE CAUSAL CHAIN
- 01
Frame a falsifiable decision hypothesis, not a product roadmap.
- 02
Recruit scientists, growers and buyers with conflicting incentives.
- 03
Run a minimal measurement-and-decision experiment.
- 04
Use independent evidence and predetermined kill criteria.
- 05
Only then consider 451° venture formation.
What would count as evidence?
- Named decision owner and paid pilot hypothesis
- Reproducible field protocol and comparative baseline
- Measured net value with uncertainty and documented negative results
What could break the thesis?
- No one owns the decision or will pay for it
- Prediction gains cannot survive field deployment
QUESTIONS TO TAKE INTO THE FIELD
Follow the next connection →What would change our minds in 90 days?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
What should never be centralized?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
Who needs to be in the room before company formation?
Do not assume the answer. Identify the relevant stakeholders, competing explanations, baseline measurement, and a test that could change the decision.
THE REFRAME / 06
Conviction is earned through evidence—not enthusiasm.
Can we build intelligence that helps living systems regenerate?
Explore the science, interrogate the economics, and help determine whether this idea deserves to become a company.