Freehold's analytics are built on public records, market transaction data, and validated benchmarks — not guesswork. Here's exactly what powers the platform.
The Data Stack
Each layer builds on the one below. The bottom layers are foundational; the top layers are predictive.
Predictive indicators that a property may become available — tax delinquency, code violations, probate filings, estate activity, equity position changes, and ownership duration.
Signal monitoring expanding to 50+ metros
Supply/demand dynamics, vacancy trends, permit activity, population migration patterns, employment data, and neighborhood trajectory indicators.
Market-level data for all US metros, updated monthly to quarterly depending on source
Expense ratios, management costs, maintenance benchmarks, insurance rates, and tax burden estimates calibrated by market, property type, and vintage.
Benchmarks segmented by metro area, unit count, and building age
Recent sales prices, cap rate transaction data, and comparable sale analytics for properties in the same area and asset class.
Sales comp data refreshed continuously from public records
Current asking rents, lease transaction data, and historical rent trends for comparable units in the same submarket.
Rent comps available for 200+ US metro areas
The foundation. Every analysis starts with verified property-level data — ownership, tax assessment, parcel information, building characteristics, and transaction history.
Available for all US residential and multifamily properties
Scoring
The Freehold Score synthesizes data across these layers into a single 0–100 rating. Every score is transparent: you can see exactly which pillars (Cash Flow Strength, Market Positioning, Risk Profile, Growth Potential) drive the number, and what data inputs feed each pillar.
Full methodologyData Quality
We prioritize verified public data sources — county records, tax rolls, deed filings — over scraped estimates or user-submitted data. When we use modeled estimates, we label them clearly.
Not all data is equally reliable. When Freehold uses an estimate instead of observed data (e.g., projected rent vs. actual lease rate), the interface flags it as an estimate with a confidence indicator.
Every Freehold Score includes a breakdown. Every rent comp shows its source. Every benchmark cites its basis. You should never have to take a number on faith.
Our benchmarks are validated against institutional operating data and updated as new public records become available. Stale data is labeled with its vintage so you know how fresh it is.
Limitations
Freehold is not a substitute for due diligence. We provide data-driven analysis to help you make better decisions faster — but we don't guarantee outcomes, predict market direction, or replace a professional appraisal. Our models estimate; you decide.
We don't provide investment advice or recommendations to buy or sell specific properties
We don't guarantee property values, rent estimates, or return projections
We don't access private MLS data that requires a real estate license (we use publicly available records and licensed data partnerships)
We clearly label estimates vs. observed data throughout the platform
Coverage
Freehold's data coverage is expanding continuously. Our current coverage:
We're actively building data partnerships to deepen coverage and improve accuracy. If you're a data provider interested in working with Freehold, contact data@getfreehold.com.
FAQ
Rent estimates are derived from current asking rents for comparable units in the same submarket, adjusted for unit size, property type, and building vintage. We cross-reference rental listing data with Census ACS Fair Market Rent data to validate our estimates. When sufficient local data isn't available, we use metro-level medians and flag the estimate as 'metro-level' rather than 'submarket-level.'
The Freehold Score is a structured analytical framework, not a prediction. It evaluates a property or portfolio across four dimensions using observed and modeled data. The score is as accurate as its inputs — which is why we recommend using it alongside your own due diligence, not as a replacement for it. We publish our methodology transparently so you can evaluate whether the scoring framework aligns with your investment criteria.
We use AI to process, structure, and analyze data — not to fabricate it. Our underlying data comes from public records, market transactions, and institutional benchmarks. AI helps us extract insights from that data faster (e.g., identifying owner distress signals across thousands of records), but we don't use AI to generate property values, rent estimates, or market projections from thin air.
It depends on the source. Property records update as counties publish them (typically monthly to quarterly). Rental market data updates continuously from listing aggregators. Operating benchmarks are validated annually against industry surveys. Market intelligence metrics (employment, permits, migration) update monthly to quarterly depending on the source agency.
Property records and basic analytics are available nationwide. Rent comps with submarket granularity are available for 200+ US metro areas. Operating benchmarks are most granular for the top 50 markets. Freehold Find's owner/distress signal monitoring is currently available in the top 50 metros and expanding. If you own properties in a market where coverage is limited, the platform will tell you.
Every number in Freehold traces back to real data. Try it yourself.