Google's review policy explained: what can be removed, what cannot, and how the decision is made

Most owners discover Google's review policy the day they need it: a bad review is up, the "report" button has been clicked, and now they wait. This guide explains what is behind that button, so your next report is written for the rules Google actually enforces.

Two different questions

Google asks two separate questions about every reported review:

• Does the content break a content policy? (spam, off-topic, hate speech, personal data and so on)
• Was the review posted by an account that breaks the account rules? (fake accounts, review rings, paid reviews)

A report can win on either. Many owners only argue the first and miss the second, which is often the stronger case.

The categories, in plain language

Spam and fake content. Reviews that do not reflect a real experience: posted to inflate or damage a rating, copied across several listings, written from accounts that review implausible numbers of businesses, or clearly generated in bulk. This also covers reviews left in exchange for money or discounts, by the business or against it.

Off-topic. The review is about something other than a customer's experience at that place. Political commentary, a complaint about a different branch, a rant about the industry, a reaction to a news story about the owner. The test Google applies: would a customer reading this learn anything about visiting this location?

Restricted content. Promotion of alcohol, tobacco, gambling, weapons and similar regulated goods in the review text. Rare, but it is one of the easiest categories to win because the rule is mechanical.

Illegal content. Anything that would be illegal to publish: copyright violations, content involving minors, or content that a court or a valid DSA notice has identified as unlawful in a given country.

Terrorist content and dangerous content. Self-explanatory and very rarely relevant to a local business.

Sexually explicit content. Includes sexualised comments about staff. This comes up more than you would think in reviews of salons and clinics and is removed quickly.

Offensive content and hate speech. Slurs, attacks based on origin, religion, gender, disability. A review that insults a waiter by name with obscene language usually falls here.

Harassment and bullying. Targeted attacks on a specific person, especially repeated ones from the same account.

Impersonation. The author pretends to be someone else, including pretending to be a customer of yours while working for a competitor, or pretending to be your business in a reply.

Conflict of interest. Reviews by current or former employees, by the owner, by family members, by competitors, or by anyone paid or asked to post. This is the single most useful category for a local business and the most often rejected, because it has to be proven.

Personal and confidential information. Full names together with accusations, phone numbers, home addresses, number plates, medical information, financial details. If a review names your receptionist and calls her a thief, it is reportable under this category regardless of whether the rest is a "real experience".

Why reports get rejected

• The wrong category was chosen. A real customer's angry review reported as "fake" will be rejected every time. Reported as "personal information" because it names and insults an employee, it has a chance.
• No evidence. Google cannot see your bookings. If you claim the author never visited, say what you checked. If you claim it is a competitor, point to the account's public reviews.
• The review was edited after your report. Reviewers sometimes soften the text once they see a reply. Google judges the current version.
• It was reported too many times from unrelated accounts. That pattern is itself treated as manipulation.
• The review is simply allowed. Negative, unfair, one-sided, exaggerated: all allowed. Google is explicit that it does not arbitrate disputes about facts.

What Google looks at that you do not see

Google's moderation combines automated checks with human review. The automated side weighs the reviewer's account history: age of the account, how many places it reviews, whether it reviews from the same device as other accounts, geographic plausibility, timing. A one-star left by an account created yesterday that has reviewed four competing dentists in one afternoon is often removed before any human looks at it. A one-star from a three-year-old account with photos of holidays and normal restaurant reviews is almost never removed on an "it's fake" claim alone.

That is why the strongest reports describe the account, not just the text.

Appeals and the Reviews Management Tool

Since 2023, Google Business Profile has a separate Reviews Management Tool where the listing owner can check the status of reports and, if a report was rejected, submit one appeal with additional information. Use it. Reports submitted from the normal profile interface without follow-up disappear into a queue with no way to add evidence later.

The appeal is a single shot per review. Write it only when you have something new: a screenshot of the author's profile, a timeline, a booking record, a document showing the author was an employee.

What the policy does not cover, and what does

Three things people ask about that the review policy does not handle:

• Defamation. A review that makes a false factual claim ("they use expired products") is a legal matter. In the EU, the Digital Services Act gives you a notice-and-action route for illegal content, and Google has a legal removal form. This is separate from the policy report and should be used alongside it, not instead.
• Data protection. If a review publishes personal data about you or your staff, the GDPR gives the affected person a right to request erasure from the platform. Again a separate process with its own form.
• Old reviews. Age is not a reason for removal. A four-year-old one-star stays until it breaks a rule or the author deletes it.

If you are not sure which route fits your review, start with a policy report for the strongest category, and keep the legal routes for cases where real damage is being done. Our guide to removing fake reviews walks through the exact steps.