Valuation
Accurate valuation is the foundation of everything downstream of ownership: pricing a sale, deciding how much to lend against a parcel, and giving owners a credible sense of what they hold. Fabrica treats valuation as a first-class part of the protocol, surfaced on every property page, written into token metadata, and published as a signed onchain reference price.
Why land valuation is hard
Most automated valuation models are built for houses, where sales are frequent and comparable. Raw land is the opposite case, and it is where general-purpose models are weakest:
- Sparse comparables. Many parcels rarely trade, and nearby sales can be scarce or stale.
- Land is not improved property. The value sits in location, size, access, topography, and zoning, not in a building that can be measured room by room.
- Noisy source data. A large share of recorded transactions are not arms-length (family transfers, bulk deals, gifts), and listing or asking prices are not the same as closed sale prices. Treating them as equal is a common cause of overvaluation.
These are real, well-studied problems, and Fabrica's approach is built to work with them rather than assume them away.
A multi-signal approach
Fabrica does not rely on a single black-box number. Several independent valuation signals are produced and shown for each property, each with its own context, so that buyers, sellers, lenders, and researchers can see where they agree and where they diverge:
| Signal | What it is |
|---|---|
| FabricaAVM | Fabrica's in-house automated valuation model, purpose-built for raw land, with a confidence level per estimate. |
| Third-party signal | A land valuation signal from a leading third-party land valuation provider, published alongside Fabrica's own so users can compare independent estimates. |
| Assessed value | The local tax assessor's value, useful for tax context (usually well below market). |
| Last recorded sale | The most recent county-recorded sale price, for historical context. |
| Lending pool valuation | For pool-based lending, the pool applies its own valuation and risk filters on top of the signals above (see Pool-Based Lending). |
Fabrica remains open to adding valuation systems over time. No single model is treated as ground truth. What each figure on a property page means, and which one leads, is explained in Property Valuation.
How lending uses valuations
Lending does not act on a single estimate. An underwriting view triangulates the available signals and sets conservative loan-to-value accordingly:
- When independent signals agree, confidence is higher and terms can be set with a normal margin.
- When signals diverge, or when confidence is low (for example a remote parcel with few comparables), the system stays conservative, widening the margin or declining.
This is intentional. Several imperfect, independent estimates that can disagree are safer to underwrite against than one estimate trusted blindly. It is also why Fabrica keeps a third-party signal in view next to its own.
The lending venues consume these signals differently:
- Pool-Based Lending (the primary live integration) uses estimated value as a starting point and applies additional risk filters to determine automated loan terms. The precise pricing mechanics are on Price Oracle.
- Peer-to-peer lenders evaluate properties individually and set their own terms; the signals inform but do not determine their offers.
Where valuations appear
- Property pages: On the Fabrica marketplace.
- Token metadata: Estimated value and confidence travel with the token to external marketplaces such as OpenSea and any application that reads the metadata.
- A signed onchain reference price: A price derived from the same valuation data is published as an EIP-712 signed quote (a signature format smart contracts can verify) through a public endpoint, consumable by lending pools, vault curators, and analytics tools. See Price Oracle for the signing scheme and integration details.
For capital providers and researchers
Because the signals are published on property pages, written into token metadata, and available as a signed onchain reference price, capital providers can build their own underwriting policies on top of them, and researchers can study land-valuation behavior across a real, growing set of tokenized parcels. Live protocol and lending statistics are available on the Dune dashboard.
Estimates are a starting point
Every estimated value here is a starting point, not an appraisal.
Land valuation is inherently imprecise, estimates vary between providers, and remote or unique parcels are harder to value accurately. Buyers and lenders should review comparable sales, consider a property's specific characteristics, and obtain a formal appraisal when the stakes warrant it. The published confidence level is there precisely so that low-confidence estimates are treated with appropriate caution.
Updated 12 days ago