Mortgage lenders use the public property record behind each loan: parcel identifiers and characteristics, owner of record, assessed and estimated market value, sales and transfer history, and the mortgages and liens already recorded against the property. Almost all of it originates with county assessor, recorder, and GIS offices. Lenders either pull it county by county or license it from a provider that normalizes roughly 3,000 county formats into one schema and delivers it by API or bulk transfer.
The property record is the one input every lending workflow shares. Origination teams use it to pre-fill applications and confirm the collateral matches what the borrower described. Collateral review uses assessed values, an automated valuation, and comparable sales to prepare for and check the appraisal. Servicing and capital markets teams watch recorded liens, estimated balances and equity, and ownership transfers across a portfolio. The data is public record, but it is published by roughly 3,000 counties in different layouts and on different schedules, so the practical question is not whether it exists but how to get it in one consistent shape. That is what a normalized property data provider does: it keys everything to the parcel, standardizes the fields, and delivers the record by API for per-loan lookups or by bulk file for portfolio-wide work.
Four groups cover most lending use cases. Identity and characteristics: parcel ID, address, lot size, building area, year built, and zoning, used to confirm the collateral is what the application says it is. Ownership of record and mailing address, used to pre-fill and to route servicing mail. Value: assessed value, tax amount, an automated valuation, and comparable sales, used for collateral review and appraisal preparation. Encumbrances and history: recorded mortgages and liens with lender, estimated balance, estimated equity and combined loan-to-value, plus the transfer history with dates, prices, and document types.
The same record serves every stage. Origination reads identity and value, underwriting reads value and encumbrances, servicing reads encumbrances and transfers, and analytics reads all of it aggregated by geography.
The appraisal is what establishes the collateral's value for a mortgage, and it usually arrives weeks into the process. Until then the file has no independent read on the property. An automated valuation model (AVM) fills that interval with a reference estimate built from public records and recent sales, available from the moment the address first enters the pipeline.
A useful AVM for collateral review does two things. It gives the team an early reference value, so appraisal preparation and collateral review can start while the appraisal is in progress. And it shows its work, returning the comparable sales behind the estimate so a reviewer can compare the appraisal against independent evidence when it arrives. Refreshed on the same rolling, per-county cycle, the estimate gives a servicing portfolio a current reference value between formal valuations. It is a reference point for collateral review, not a substitute for the appraisal and not an input to any decision about a borrower.
Nearly all of it is public record. County assessors publish characteristics, assessed values, and tax rolls. County recorders file deeds, mortgages, releases, and liens. GIS departments publish parcel boundaries and zoning. The figures that are not public record, such as an automated valuation, a lien balance estimate, or an equity estimate, are modeled from those records and from recorded sales.
The difficulty is scale. Each county chooses its own field names, formats, identifiers, and publication cadence, so a lender operating in more than a few markets ends up maintaining a parser per county. A provider like Realie collects at the source, normalizes the counties into one schema keyed by parcel, refreshes it on a rolling, per-county cycle as counties publish, so vintage varies by county, and delivers it through an API for per-loan lookups or bulk transfers for portfolio-wide work.
Property data describes a parcel, not a person. It is not a credit report, and Realie is not a consumer reporting agency. Realie's Terms of Use prohibit using the data as a factor in an individual's credit eligibility or for any other FCRA-regulated purpose, and Realie does not market owner lookups by name.
Address-first collateral workflows sit inside that boundary because they start from a property the lender already knows: the address on an application, the collateral behind a loan, or the parcels in a serviced portfolio. The data informs what is known about the property, not any decision about the borrower. Keeping the workflow address-first is both the compliant pattern and the one that produces the cleanest data.