Start hereZillow publishes how often its Zestimate misses: on 1 October 2026, the median miss for listed homes was 1.79%. Yet in 2021 the same company wound down its home-buying business after its price forecasts missed. So what can these tools tell you?
The verdicts
What an automated valuation is
Here, an “AI property tool” means software that does some mix of three jobs: estimating what a home is worth, finding listings, and working out rent, costs and loan cover. The first job rests on an automated valuation model.
A 2024 study published by the US Department of Housing and Urban Development (HUD) gives a definition. An automated valuation model, or AVM, is a computer-driven formula that estimates a property’s value from its characteristics, local market information and price trends.
Zillow’s Zestimate is one example, and Zillow publishes how it works and how often it misses. Zillow says the Zestimate uses a neural-network-based model that draws on county and tax assessor records and direct feeds from multiple listing services (MLS), along with user-submitted information. Zestimates cover residential properties only: large multi-family buildings, vacant land and commercial properties are not eligible.
How accurate the estimates are
When we read Zillow’s page on 1 October 2026, it put the nationwide median error of the Zestimate at 1.79% for homes on the market and 7.20% for off-market homes. For listed homes, the estimate was within 1.79% of the final sale price half of the time. The other half of the time, it was further off.
On 13 October 2026, the same page gave 7.20% for off-market homes but “NA” for homes on the market, in a table last refreshed on 4 October 2026.
The company gives a reason for the gap. Listed homes have more up-to-date information, including listing details and recent market activity.
Two things follow from Zillow’s own figures. A home that is not for sale falls in the off-market group. In the 1 October figures, its median miss was about four times larger.
A percentage also turns into money quickly: on a $300,000 home, a 7.20% miss is $21,600. Neither figure tells you how far off the estimate is for any one home.
Zillow explains how it measures them. For listed homes it uses the last Zestimate before the home goes under contract, which already draws on the listing. For off-market homes it counts only homes that eventually sell.
Zillow also leaves foreclosure sales out of the Zestimate, because its analysis found they generally sell at substantial discounts. The Zestimate is meant to estimate what a home would fetch if sold for its full value. That is a different question from what a foreclosure will sell for.
↑ All verdictsAn estimate is not an appraisal
Zillow is direct about this: “It is not an appraisal and can’t be used in place of an appraisal.”
Asked whether a Zestimate can be used to get a loan, its answer is no: most lenders use only professional appraisals for loan decisions. It suggests visiting the home, a professional appraisal, or a comparative market analysis from a real estate agent.
The US rule described next draws the same line. It does not apply to the use of AVMs when a certified or licensed appraiser develops an appraisal.
↑ All verdictsThe US rule for lenders’ automated valuations
Six US agencies (the OCC, the Federal Reserve, the FDIC, the NCUA, the CFPB and the FHFA) published a final rule on 7 August 2024 setting quality-control standards for AVMs. It took effect on 1 October 2025, and the federal eCFR listed its sections as in force when we checked on 1 October 2026.
Separately, US bank and credit-union regulators have given the institutions they supervise guidance on these models since 2010.
Lenders and secondary-market issuers that use an AVM to value a home securing a mortgage need policies and controls. The rule lists five standards those controls are designed to meet:
- High confidence in the estimates.
- Protection against manipulation of the data.
- Avoiding conflicts of interest.
- Random sample testing and reviews.
- Compliance with nondiscrimination laws.
The rule does not prescribe how. The agencies point to the flexibility it gives institutions to design their own policies and controls.
Its scope is narrow. It covers mortgages secured by a consumer’s principal dwelling. A consumer can have only one principal dwelling at a time, so a vacation or other second home is not one.
It is a rule about the models lenders use. By its own definitions, it does not set a standard for a value that a website or an investment tool shows you.
↑ All verdictsBias: what one study found
In the rule’s background section, the agencies wrote of growing concern that AVMs could produce estimates reflecting discriminatory bias, for example by “replicating systemic inaccuracies and historical patterns of discrimination”.
A study in HUD’s journal Cityscape in 2024 tested this in counties of the Atlanta and Memphis areas. Its authors were Linna Zhu and Michael Neal of the Urban Institute and Caitlin Young of Yale Law School. The authors note the hope that AVMs would reduce appraisal bias by taking the appraiser out of the process.
They added data on each property’s condition and used machine-learning methods, and still found “evidence that AVMs yield larger valuation errors in majority-Black neighborhoods”. Condition mattered too: in one of the study’s regressions, a home rated poor rather than good had an error larger by 4.35 percentage points.
The gap is in percentage terms. In the study’s figures for 2000 to 2018, errors measured in dollars were mostly smaller in majority-Black neighbourhoods, where average home values were lower.
This is one study, whose main analysis uses 2018 sales in two metro areas. It shows that a gap in percentage errors remained after the authors accounted for property condition. It cannot say how large the gap is elsewhere, today, or for any particular tool.
↑ All verdictsValuing today is not forecasting tomorrow
On 2 November 2021, Zillow’s board decided to wind down Zillow Offers, the company’s business buying and selling homes. Its filing gave the reasons as home pricing unpredictability, capacity constraints and other operational challenges.
The filing recorded a $304.4 million write-down on homes bought at prices above Zillow’s current estimates of their future selling prices after selling costs. It expected the wind-down to cut its workforce by about 25%.
Zillow’s shareholder letter that day said the business depended on forecasting home prices accurately three to six months ahead. A few sentences later: “We have been unable to accurately forecast future home prices at different times in both directions by much more than we modeled as possible.”
Its chief executive said the unpredictability in forecasting home prices far exceeded what the company had anticipated.
The same letter names a pandemic and an unprecedented rise in prices as part of the story, so this is not proof that prices can never be forecast. It is a company with its own valuation model explaining why its forecasts were not good enough to run that business. A valuation estimates today’s price; a deal often depends on tomorrow’s.
↑ All verdictsWhat deal-finding tools do, and what is known about them
Of the three jobs, the third is mostly arithmetic. One calculation such a tool may run is the debt service coverage ratio, or DSCR.
The OCC’s handbook for bank examiners defines the DSCR as net operating income divided by the year’s debt service requirements. Net operating income is annual gross income less operating expenses.
The formula is simple, and so are its limits. The ratio is only as good as the rent, the expenses and the loan terms put into it. If the rent is itself an estimate, any error in it carries straight into the result.
Do the AI parts find better deals than a person would? We looked for an independent test, a study comparing what such tools picked with how those deals turned out, and found none.
What we found were vendors’ own case studies and review lists. That does not show the tools fail. It means the claim has not been tested in public in anything we could find.
↑ All verdictsWhat US regulators say about money-making pitches
In October 2021 the US Federal Trade Commission sent a notice to more than 1,100 businesses that pitch money-making ventures. It summarised earlier cases finding it unlawful to make false or misleading claims about earnings. Examples include claiming that participants will make a profit, or that the profits shown are typical.
It also listed falsely telling consumers they need no experience to earn income, or that they must act immediately to take part. Penalties at the time were up to $43,792 per violation; after yearly inflation adjustments, the maximum in force when we checked in October 2026 was $53,088.
The FTC’s consumer advice puts it plainly: “If it promises guaranteed income, large returns, or a ‘proven system,’ it’s likely a scam.”
On real estate seminar and investment coaching scams, it says such schemes routinely use made-up stories and statistics, claim a guaranteed return, or stress how cutting-edge their offer is. It also warns that success stories and testimonials might not be true or typical.
On 25 September 2024 the FTC announced five law enforcement actions under the name Operation AI Comply. They were about a tool for writing fake reviews, an AI lawyer service, and companies claiming AI could help consumers make money from online stores. None was about property.
In one, the FTC alleged that a scheme “falsely claimed its ‘cutting edge’ AI-powered tools would help consumers quickly earn thousands of dollars a month in passive income”. These are the FTC’s allegations; the case it filed in court is for the court to decide.
The FTC’s chair at the time said: “there is no AI exemption from the laws on the books.”
↑ All verdictsQuestions to ask of any property tool
These are questions, not advice. The answers are yours to weigh, with a licensed professional where the money is large.
- Where does the data come from, and how fresh is it? One large property site, for example, names its inputs and says it refreshes its estimates multiple times per week.
- Is the number an estimate or an appraisal? The answer decides whether a lender will rely on it.
- Does it publish its error rate, for listed and unlisted homes separately? At least one large site has published both, and the gap between the two is the useful part.
- Are the income figures typical, and what are they based on? The FTC suggests asking how the business would make money, what your own expenses would be, whether you can afford it, and when you would expect a profit.
- If it offers or arranges loans, who is licensed? In the US, companies and individuals that originate mortgages must be licensed or registered through the Nationwide Multistate Licensing System, and anyone can check status on NMLS Consumer Access. For investment professionals, FINRA’s BrokerCheck links to the SEC’s adviser database. These are US registers; other countries keep their own.
- Does the pitch lean on urgency, guarantees or “no experience needed”? Those are the claims the FTC has named.
Quick answers
How accurate is the Zestimate?
Zillow publishes its own median error. On 1 October 2026 it was 1.79% for homes on the market and 7.20% for off-market homes; on 13 October the on-market figure read NA. Half of the estimates miss by more than the median.
Is a Zestimate an appraisal?
No. Zillow says it is not an appraisal and can't be used in place of one, and that most lenders use only professional appraisals for loan decisions.
Are automated valuation models regulated in the US?
For lenders, yes. Since 1 October 2025 a rule from six US agencies has required lenders and secondary-market issuers to have quality controls on the models they use for mortgages on a consumer's principal dwelling. It does not cover values shown on websites or in investment tools.
Can AI find better property deals than people?
We found no independent test of that, only vendors' own case studies and review lists. That does not show the tools fail; the claim has not been tested in public in anything we could find.
What does the FTC say about AI money-making claims?
The FTC has named false claims that profits are typical, that no experience is needed, or that you must act now. Announcing its AI cases in 2024, its chair said there is no AI exemption from the laws on the books.
Use this
Treat any automated value or income claim as a starting estimate, and ask what it rests on.
What is this number based on, and how often is it wrong?
Quick quiz
Nothing is scored or saved. Recalling it once now is what makes it last.
1. On 1 October 2026, what median error did Zillow publish for off-market homes?
Answer: 7.20%.Read it again ↑
2. Who does the 2024 US rule on automated valuations apply to?
Answer: Lenders and secondary-market issuers.Read it again ↑
3. Why did Zillow say it wound down Zillow Offers?
Answer: Home pricing unpredictability and operational challenges.Read it again ↑
How we checked this · Last checked October 2026
- We read Zillow's Zestimate page; the final rule in the Federal Register and its status in the eCFR; the HUD study's abstract and full text; Zillow's 8-K, press release and shareholder letter of 2 November 2021; the OCC handbook's definitions; the four FTC pages and the eCFR penalty table; and the CSBS and FINRA pages.
- Zillow's page changed between our readings. On 1 October 2026 it gave 1.79% for listed homes (table refreshed 16 September); on 13 October it gave NA (table refreshed 4 October), while the off-market figure stayed at 7.20%.
- On 13 October 2026 we also re-read the FTC pages in a browser (they turn away scripts) and the eCFR penalty table, which still gives $53,088 for penalties assessed after 17 January 2025.
- The $21,600 figure is our own arithmetic: 7.20% of a $300,000 home. "Five" law enforcement actions is our count of the cases the FTC's 2024 release describes.
- What we could not find: any independent test of whether AI deal-finding tools pick better deals, or published error rates for those tools' own estimates.
- The rules and registers here are US ones; other countries keep their own.
Sources
- Zillow, “What is a Zestimate?”, read 1 October 2026 (table last refreshed 16 September 2026) and 13 October 2026 (table last refreshed 4 October 2026).Primary
- OCC, Federal Reserve, FDIC, NCUA, CFPB and FHFA, “Quality Control Standards for Automated Valuation Models”, final rule, 89 Fed. Reg. 64538, 7 August 2024; eCFR, 12 CFR part 34, subpart I.Primary
- L. Zhu, M. Neal and C. Young, “Racial Disparities in Automated Valuation Models: New Evidence Using Property Condition and Machine Learning”, Cityscape 26(1), 2024, HUD Office of Policy Development and Research.Scholarly
- Zillow Group, Form 8-K, 2 November 2021, with its press release (Exhibit 99.1) and shareholder letter (Exhibit 99.3).Primary
- OCC, Comptroller’s Handbook: Commercial Real Estate Lending, version 2.0.Primary
- FTC, “FTC Puts Businesses on Notice that False Money-Making Claims Could Lead to Big Penalties”, 26 October 2021.Primary
- FTC, “When a Business Offer or Coaching Program is a Scam”, consumer advice.Primary
- FTC, “FTC Announces Crackdown on Deceptive AI Claims and Schemes”, 25 September 2024.Primary
- Conference of State Bank Supervisors, “Protecting Homebuyers with NMLS and Consumer Access”.Primary
- FINRA, “Check Registration: Sellers and Investments”.Primary
- FTC, “FTC Publishes Inflation-Adjusted Civil Penalty Amounts for 2025”, 11 February 2025; eCFR, 16 CFR 1.98.Primary