September 2, 2026 · 10 min read

How Google's review policies actually work in 2026

Google removed 292 million policy-violating reviews in 2025. Here is what gets a single review pulled, and what gets the business itself penalised.

Most advice about Google reviews is written as if there were a single rulebook and a single punishment. There are two, and confusing them is expensive.

The first is what happens to an individual review. It gets filtered before it appears, or it gets removed after a complaint, and the business usually never finds out why. The second is what happens to the business — a Business Profile that stops being reachable, in the middle of a week when the phone should be ringing. The first is an annoyance. The second is a revenue event, and in almost every case it is caused not by a customer's review but by how the business went about collecting reviews.

This is what Google's own documentation says about both, as published at the URLs cited and accessed on 2 September 2026. One caveat worth stating up front: none of Google's policy pages carry a visible last-updated date. They change without a changelog. Anyone who tells you a specific rule "changed in April" is inferring it, and so is anyone quoting a version of a Google rule from a blog post rather than from Google.

The document that actually governs your reviews

The operative text is the Maps User Contributed Content Policy, mirrored at support.google.com/business/answer/7400114. It covers reviews, ratings, photos, videos, Q&A and place edits — everything a member of the public can add to Maps.

It is organised into roughly two dozen categories: fake engagement, rating manipulation, impersonation, misinformation, misrepresentation, harassment, hate speech, offensive content, personal information, obscenity, sexually explicit content, adult-themed content, violence and gore, restricted content, dangerous content, child safety, terrorist content, off-topic, advertising and solicitation, unclear content, repetitive content, and defacement.

Most of those never come up for an ordinary small business. Four do.

What gets a single review removed

Fake engagement

Google's line is short and unusually plain:

Contributions to Google Maps should reflect a genuine experience at a place or business. Fake engagement is not allowed and will be removed.

The policy then enumerates what fake means, and the list is broader than "we made it up." It includes content "not based on a real experience," reviews or ratings "that have been paid for, directly or in kind," content "posted from multiple accounts by or at the request of one person," and content posted using an emulator or a modified operating system to mimic genuine engagement.

The "directly or in kind" clause is the one that catches honest businesses. A free dessert is in kind. So is a discount code.

Conflict of interest

Content that is based on a conflict of interest. A conflict of interest may include current or former employment, a contractual or consultory relationship, or other professional or personal affiliations that demonstrate a conflict of interest (such as industry competitors, familial relationships, etc.).

Note what this does and does not say. It does not contain a sentence reading "business owners may not review their own business." The prohibition on staff reviews is an inference from the words "current or former employment" — a well-founded one, but if you are quoting Google to a sceptical partner, quote the conflict-of-interest clause and not a paraphrase you found on an agency blog.

Off-topic and advertising

Only post content that is based on your experience or questions about experiences at a specific location.

And separately: "Don't post content for advertising or solicitation purposes." Between them, these two cover the majority of genuinely removable one-star reviews from people who were never customers — the ones about a delivery driver, a parking dispute, or a national brand's advertising campaign left on a local franchise's profile.

Rating manipulation

The category exists in the policy and is worth knowing about, but in practice it overlaps almost entirely with fake engagement for a small business. Where it bites independently is coordinated campaigns, including negative ones aimed at a competitor.

What gets the business penalised

This is the part that matters more, and it is in the same policy under prohibited merchant conduct. Verbatim, a business must not:

• "Solicit or encourage the posting of content that does not represent a genuine experience."
• "Offer incentives – such as payment, discounts, free goods and/or services - in exchange for posting any review or revision or removal of a negative review."
• "Post content on a competitor's place or business to undermine that business' or product's reputation."
• "Discourage or prohibit negative reviews, or selectively solicit positive reviews from customers."

That last line is the whole of review gating, in Google's words, listed as something the business does rather than something the review is. It is the single most commonly violated rule in the local marketing industry, and it is violated by tools that market themselves as reputation management.

The Tips to get more reviews page repeats the incentive prohibition in stronger terms — "strictly prohibited" — and adds three constraints on how staff ask that are newer than most published guidance and easy to trip over:

• Merchants "should not require or pressure users to leave ratings or write reviews while on the premises."
• Merchants must not "request that specific content be included."
• Merchants must not request "that staff solicit a certain number of reviews," nor "that staff solicit reviews that include specific content, including content that identifies a staff member."

Read those together and a very common practice becomes non-compliant: a monthly staff leaderboard for "reviews that mention your name," collected by handing a tablet to a customer at the counter. Each of the three clauses independently prohibits part of it.

What Google explicitly does allow is the plain version: "To leave reviews, you can ask customers to visit a Google link or scan a QR code."

How much of this is machine-enforced

Google publishes annual removal figures on its own blog. These are the only enforcement numbers with a primary source; every percentage you see circulating beyond them is somebody's arithmetic.

Sources: Google, "How we're using AI to combat fake Business Profiles and reviews", 7 April 2025; and "New ways we're protecting businesses on Maps", 16 April 2026. The 2026 post also states Google published "more than 1 billion helpful reviews" over the same period, and that Gemini-based models "spot and block suggestions that violate our policies before they go live."

The operational reading of that table: enforcement is automated, it runs before publication as well as after, and its volume grew about 22% year over year. There is no human at Google weighing your intent. A pattern that looks like solicitation-with-strings will be treated as solicitation-with-strings.

When a profile gets suspended

Google's Fix suspended or disabled profiles page states simply that "we may suspend or disable Business Profiles that don't follow our guidelines," and describes the consequence: "The public can't go to the profile. The owner and managers can't act on the profile."

Two practical details from the appeals documentation:

• Once you open the evidence form, "you must submit it within 60 minutes or it won't be attached to your appeal." Gather your documents before you start.
• "Appeal reviews and decisions can take up to 5 business days."

Do not create a duplicate profile while an appeal is pending. A denied request can get one additional review; a duplicate profile complicates both.

Worth flagging, because the internet insists otherwise: Google does not document "suspended" and "disabled" as two distinct states with different visibility rules, and it does not publish a canonical list of suspension reasons. The familiar agency distinction between a "soft" and "hard" suspend appears nowhere in Google's own material.

Reporting a review you believe breaks the rules

The process, from Report inappropriate reviews on your Business Profile: open your Business Profile, select Read reviews, select Report next to the review, choose a reason, and send. The Reviews Management Tool offers the same flow.

Google states one timeline only — "Review evaluation typically takes several days" — and offers a single appeal if the outcome comes back as "Report reviewed - no policy violation," covering up to ten eligible reviews at once. Save that appeal for reviews with a concrete policy hook: the reviewer was never a customer, the review is about a different location, it contains personal information, it is a competitor. "This is unfair" is not a category.

The UK deal is a preview of the enforcement model

In January 2025 the UK's Competition and Markets Authority secured formal undertakings from Google. The CMA's announcement of 24 January 2025 describes what businesses caught using fake reviews face: "warning" alerts on their profiles, review functions deactivated so they "cannot receive any new reviews," and for repeat offenders, all reviews deleted "for 6 months or more." Google reports to the CMA on compliance over a three-year period.

Google said at the time it intended to extend the sanctions programme beyond the UK. I could not find a primary source confirming that extension happened, so treat it as a stated intention rather than a fact — but the design tells you where the enforcement is going. The penalty is not a fine. It is losing the ability to collect reviews at all, which for a business that depends on local search is worse.

What a compliant ask looks like

Everything above narrows to a short list. Ask every customer, not the ones you expect to be happy. Ask once, follow up at most once. Offer nothing in return. Do not script the content. Do not put a number on a staff member's head. Send the person to the public Google link and let them write whatever they write.

Software helps here mainly by making the boring version easy: sending the same request to everyone on the same schedule, and keeping a record of who was asked and when if you ever need to show your process. ReviewHero, for example, sends every customer the same email with the same public Google link and stops following up the moment the link is opened — which is a deliverability and courtesy feature, not a filtering one. Any tool that offers to route unhappy customers somewhere other than the review link is selling you the thing Google's policy names.

What to do next

1. Open the prohibited content policy and read the merchant conduct section. It is four bullets and takes two minutes.
2. Audit how your staff currently ask. Kill any script that requests specific wording, any staff review quota, and any on-premises tablet handoff.
3. Remove every incentive — including the informal ones nobody wrote down.
4. Check whether your review tool asks a satisfaction question before showing the Google link. If it does, that is selective solicitation, and it is your business's exposure, not the vendor's.
5. Flag the reviews on your profile that have an actual policy hook, and keep the one-time appeal in reserve for the strongest few.