The agent tool landscape

What the public record gets wrong

6 min read

A tool is only as accurate as the record it reads, and the record is wrong more often than it looks.

The short version

Automated valuation models build their estimates from public records: square footage, bedroom and bathroom counts, lot size, year built, and prior sales. Each of those fields is entered, transcribed, and updated by people and systems that make errors and fall behind. Recorded square footage can differ from a measured figure by hundreds of feet. A renovation may never appear because the permit was not pulled or not posted. A prior sale may be a transfer between family, not a market price. The model does not flag any of this. It reports a number that looks exactly as confident as one built on clean data, which is why the published off-market median error runs near 7.5%, roughly $45,000 on a $600,000 home.

The record is entered by hand, and lags

Public property data is assembled from deeds, permits, and assessor files, updated on the assessor's schedule, not the market's. Fields get transcribed wrong, additions are missed, and updates arrive months late. The gaps are not exotic, they are routine, and they sit underneath every automated estimate that reads the file.

Record field	A common error	Effect on the estimate
Living area	Tax figure differs from measured area	Scales the entire estimate up or down
Bed and bath count	Stale after a conversion or addition	Miscounts a core value driver
Renovations	Work never permitted or never posted	Prices an unrenovated home
Prior sale	A transfer or refinance, not a market sale	Plants a false comparable
Lot size	A mapping or rounding error	Mis-sizes the land value

Small input errors, large price swings

Because square footage scales an estimate, a modest error moves real money. A record that understates a home by 200 square feet, at a neighborhood rate around $300 a foot, shifts the implied value by about $60,000 before anything else is considered. A missed renovation prices a finished home as a dated one. A miscounted bathroom drops a value driver. These are not rounding errors in the output, they are structural errors in the input.

The prior-sale trap

Models lean hard on a home's own sale history, and not every recorded transfer is a market sale. A deed moving a home between family members, a refinance recorded as a transaction, or a foreclosure can all look like a sale in the data and anchor the estimate to a price no open-market buyer ever paid. In non-disclosure states the opposite failure appears, the sale price is withheld entirely, the gap covered in non-disclosure states and the data gap.

What the record cannot hold at all

Beyond the errors sit the omissions. Condition, light, road noise, the school line, and the character of the block never enter the record in any form, which is the separate limit described in what a Zestimate cannot see. Errors are wrong entries. Omissions are missing fields. A confident automated number carries both, and announces neither, which is why the model machinery can be precise and wrong at the same time.

What this means for an agent

Before you price, read the record against the house. Confirm the square footage, the bed and bath count, the permitted work, and whether the last recorded sale was an arm's-length deal. Correcting a record error is often the single largest adjustment in a valuation, and it is one an agent can make and an algorithm cannot. Showing a seller exactly where the online estimate read a wrong field is the basis of our home-value accuracy page, and the honest-range consumer estimate is the home-value page.

Frequently asked questions

Is public property data accurate?

Often not. Public records are entered by hand and updated on the assessor's schedule, so square footage, bed and bath counts, renovations, and prior sales frequently carry errors or lag reality. Automated tools that read those fields inherit the mistakes.

Why is the square footage on my property record wrong?

Because it was transcribed from an old deed or plan, or measured under a different standard than a professional would use. Tax records, builder plans, and measured floor areas can differ by hundreds of square feet on the same home.

Do home renovations show up in public records?

Not reliably. A renovation appears only if a permit was pulled and the update was posted, and often neither happens on schedule. As a result, records routinely describe a finished home as if the work was never done.

Why does my home's record show the wrong number of bedrooms?

Usually because a conversion or addition was never reflected in the file, or a room was classified differently than you would classify it. Bed and bath counts are core value drivers, so a stale count can move an automated estimate.

How do I correct wrong information in my property record?

Contact your county assessor's office with documentation, a survey, permits, or an appraisal, showing the correct figures. Correcting a record error is often the largest single adjustment in a home's valuation.

Sources: County assessor and recorder practice on how property records are compiled and updated, appraisal guidance on verifying record data against the subject property, and published automated valuation methodology on the role of public-record inputs. The 7.5% off-market median error reflects Zillow's own published Zestimate accuracy figures; dollar impacts are illustrative.

This article is general information and analysis, not financial, lending, or appraisal advice. Verify any home value with a licensed professional before acting.