The Listing an Organisation Does Not Own

The public record of a local business is not published by the business. It is assembled from several inputs, of which the owner's own submission is one, and it is edited continuously by parties with no relationship to the organisation. Google reported in April 2026 that during 2025 it had blocked 79 million inaccurate or unverified edits to Business Profiles. That figure counts the edits Google rejected. It is the clearest published measure of how much external editing pressure sits on records businesses are assumed to control.

Where the record comes from

Google states the sourcing plainly. Its Business Profile documentation lists four inputs: crawled web content, licensed third-party data, users who contribute factual information, such as addresses and phone numbers, including owners who claim their profiles, and Google's own interactions with the place. The owner is one contributor among four, and the owner's contribution is a submission rather than a publication.

The documentation is silent on how conflicts between those four inputs are resolved. When a crawled page, a licensed data feed, a user suggestion and an owner's entry disagree about an address or a set of opening hours, no published rule says which wins. That absence should be recorded rather than filled: the resolution procedure is not disclosed.

The volume of third-party editing

The enforcement figures give the scale. Alongside the 79 million blocked edits for 2025, Google reported removing over 13 million fake Business Profiles in the same year, and stated that the Maps community shared more than 1 billion reviews and suggested 80 million updates to business information. Google's own wording is "more than 1 billion", and it should not be rendered as a precise count.

The prior year gives the trend. For 2024 Google reported more than 70 million policy-violating edits to places blocked or removed, more than 12 million fake profiles removed or blocked, and posting restrictions applied to more than 900,000 repeatedly violating accounts. Blocked edits rose from over 70 million to 79 million, and fake profile removals from over 12 million to over 13 million.

Set the two 2025 figures side by side and the shape of the system is visible. The community suggested 80 million updates to business information; Google blocked 79 million edits as inaccurate or unverified. Both quantities are external to the businesses concerned. The organisation named in the record is not the party doing most of the writing to it, and it is not the party adjudicating it either.

The record is read, and it outranks the website

The audience for that record is close to universal within the category. BrightLocal's Local Consumer Review Survey 2026, a February 2026 survey of 1,002 US adult consumers, reports that 97% of consumers read reviews for local businesses and that 31% will only use a business with 4.5 stars or more. The same survey finds that 89% expect owners to respond to reviews, with 80% saying they are likely to use a business that responds to all of them. These are self-reported figures from a consumer panel.

An earlier BrightLocal report, from October 2023 and based on 1,138 US consumers, asked which sources people trust for information about a local business and recorded Google at 66%, Google Maps at 45% and the business's own website at 36%. On that measurement the surface the organisation does not control is trusted at nearly twice the rate of the one it does.

How often the record is wrong

Direct evidence here is thin and dated, and the distinction between self-report and audit matters. BrightLocal's Business Listings Trust Report, published in September 2021 from a survey of 1,141 US consumers, found that 85% said they had encountered incorrect or incomplete information on a business listing in the past year and that 77% saw conflicting information on a business across different online directories. That is what consumers report noticing, not a measurement of the listings themselves.

One audit-style study exists and it is seven years old. Uberall analysed 73,000 business locations across Google, Yelp and Bing and reported in April 2019 978,305 opening-hour errors across the set, with only 3.82% of locations correct on all three directories and an average readiness score of 44.12%. Three caveats travel with it: the study is from 2019, Uberall's own page for it now returns a 404 so only the wire copy survives, and "correct" is Uberall's proprietary rubric rather than an independent standard. The error count is the more defensible half of the finding.

No current audit of listing accuracy at scale could be found. Figures circulating for the share of business profiles that are claimed or unclaimed trace to stat roundups with no traceable original and are not usable.

Implication

The operative description of a business is maintained by parties other than the business, at a volume the business cannot match. That is not a claim about neglect. It is arithmetic: tens of millions of external edits a year against one owner's occasional submission.

Two consequences follow for how the record should be treated. It is a monitored surface rather than a published one, so the useful cadence is checking what it currently says rather than setting it once. And accuracy of the underlying facts is worth more than presentation, because those facts are what a contradicting third-party edit is checked against.

The gap in the evidence should be stated rather than covered. There is no recent measurement of how accurate local records are, no published resolution rule for conflicting sources, and no figure for how long an incorrect entry persists before correction. What is measured is the volume of editing pressure, and that volume is very large.