What reviews does Google remove?
Google removes reviews that break its Maps user-generated content policy. The rules are on one page, Prohibited & restricted content. It has 21 headings in two groups: Fake & Misleading Content & Behavior, and Inappropriate content & behavior. Google does not remove a review because it is harsh, one-sided or wrong about a detail. Its report page says Google “doesn’t get involved in conflict between businesses and customers”.
The same rules cover text reviews, ratings, photos, videos, captions and links, and Google’s About our policies page says they apply whether a person or an AI wrote the content. Google uses machine learning and trained reviewers to find violations. It also says it generally removes the full post when one part breaks a rule, so an offending photo or sentence can take down the whole review.
What removed reviews look like in practice: across 1,170 Google reviews that came down after we filed them and that we hold a copy of (TrustReviews data, October 2026), 86.6% were one-star, 6.8% had no text, 36.8% had an owner reply and 1 in 3 was over a year old when we filed it. So a reply, a missing text or an old date does not by itself keep a review up. The heading it breaks decides. You can read the full numbers in our Google review data study.
Every heading in Google’s review policy
The table uses Google’s exact headings, in the order of the policy page. Each heading links to Google’s own definition where one exists. Unclear and Repetitive Content has two parts, so it takes two rows. The last column is our reading of how easy a break of that heading is for Google to see from the review and from public information. It is not a removal rate, and Google decides each case.
| Heading | What it covers | Example review | How strong the case usually is |
|---|---|---|---|
| Fake Engagement | Content not based on a real experience, reviews paid for directly or in kind, content posted from several accounts by or for one person, and device tampering. Since February 2026 it also bars users, not only merchants, from posting on a competitor’s place to hurt it. | “Never been here and never will. Everyone knows they rip people off.” | Strong when the review says the person never visited, or when several new accounts post at once. Weak when the reviewer could be a customer you have no record of. |
| Rating Manipulation | Reviews given for an incentive, unusual volumes or patterns of reviews meant to move a rating, and reviews based on a conflict of interest: current or former employment, a contractual or consultory relationship, industry competitors or family. It also limits how a merchant may ask for reviews. | “Worked here two years. Management is a joke and they cut corners on every job.” | Strong when Google can see the link from public information, such as a reviewer name that matches a former employee or a competitor’s owner. Weak when the report only says the link exists. |
| Impersonation | Content that pretends to be another person, group or organization, or a verified authoritative source. | A review from an account named “Bright Smile Dental Owner” that says “We know our prices are a scam.” | Strong when the account name or photo copies yours. Rare in practice. |
| Misinformation | Harmful, deceptive health or medical information, deceptive information about civic processes, and manipulated content about news events. | “This pharmacy swaps flu shots for saline to save money.” | Narrow. It fits false health claims, not disputed facts about a visit. |
| Misrepresentation | False or misleading accounts of a good or service, facts distorted to scam users, content based on a conflict of interest, and phishing. | “They charged me $900 for a job they never did,” when your records show the job was done. | Weak on its own, because Google cannot check your invoices. Its conflict-of-interest line repeats the one under Rating Manipulation. |
| Harassment | Specific threats of harm, doxxing, and unwanted sexualization of a person. | “Tell Mike at the counter I know where he parks.” | Strong. A threat aimed at a named person is one of the clearest grounds in the policy. |
| Hate speech | Calls to violence against, or content that dehumanizes, belittles or vilifies, people on a protected characteristic. | A complaint about a long wait that ends with a slur about the owner’s religion. | Strong when the slur or attack is explicit. |
| Offensive content | Attacks on people, deliberately provocative content, and “unsubstantiated allegations of unethical behavior or criminal wrongdoing”. Negative experiences described respectfully are allowed. | “The owner is a con artist who launders money through this shop.” | Moderate to strong for a specific accusation of a crime with nothing behind it. Weak for harsh opinions about the service. |
| Personal information | Another person’s full name or last name, face, or other details posted without consent, and financial, medical or ID information. Names that are part of the business, public-facing professionals and known executives are allowed. | “The receptionist, Dana Whitfield, lives off Elm Street and drives a red Civic.” | Strong for a non-public employee’s full name with personal details, and for medical or financial details. Weak for a doctor or lawyer who works under their own name. |
| Obscenity & profanity | Profanity or obscenity used to offend or to add weight to criticism. | A two-line review that calls the staff a string of obscenities. | Strong when the swearing is most of the review. Weak when one word sits in a long, calm review. |
| Sexually explicit content | Pornography, depictions or descriptions of sexual acts, and genital nudity. | A photo with nudity attached to a review of a gym. | Strong, and rare in reviews of local businesses. |
| Adult-themed content | Non-explicit sexual content, sexualized objects and sexually suggestive poses. | “Book the massage and ask for the ‘extras’, if you know what I mean.” | Strong, and rare. A sexual remark about a named person also fits Harassment. |
| Violence & gore | Graphic violence, serious injury or death, and animal cruelty meant to shock. | A close-up photo of a bleeding wound attached to a review of a tattoo studio. | Mixed. A graphic photo can come down while a real complaint in the text stays. |
| Restricted content | Offers, links and calls to action for regulated goods such as alcohol, gambling, tobacco, guns, medical devices, pharmaceuticals, adult services and financial services. Incidental mentions are allowed. | A review on a bar’s profile: “Cheaper vapes at this link,” followed by a web address. | Strong when the review carries an offer or link. It does not apply to an ordinary complaint. |
| Dangerous content | Content that encourages serious harm, or gives instructions for misusing or making dangerous items. | A review of a gas fitter that explains how to bypass a shutoff valve yourself. | Strong, and rare. |
| Child safety | Content that exploits, abuses or endangers children. | A comment that sexualizes a child seen at a daycare. | Strong. Google says it may also report it to the authorities. |
| Terrorist content | Content that promotes or celebrates terrorism, or is made for terrorist groups. | A review that praises an attack that happened near the business. | Strong, and very rare. |
| Off-topic | Content not based on an experience at that place: general, political or social commentary and personal rants. | “Boycott every business in this town after what the city council did.” | Strong for reviews about politics, the news or a different business. Weak when the complaint is about the visit. |
| Advertising & solicitation | Promotional content, and reviews with email addresses, phone numbers, social media links or links to other websites. Since September 2026 it also sets separate rules for merchants’ listings and posts. | “Awful plumber. Call 555-0134 for one who shows up.” | Strong when the text holds a phone number, email or link, because Google can see it. |
| Unclear content | Content with no meaning in any language, such as a random group of characters. | “kjsdf kjsdf ;;; 11” | Strong. |
| Repetitive content | The same content posted more than once, from one account or several. | The same “bait and switch” paragraph posted from four accounts in one day. | Strong. Report it under Fake Engagement or Rating Manipulation as well. |
| Defacement & Mischief | Edits that damage a place’s listing as a prank or to express views. | Not a review: a suggested edit that renames your business to an insult. | Does not apply to reviews. Fix the edit in your Business Profile. |
What changed in 2025 and 2026
Google’s policy page has no change log. We dated each change by comparing copies of the page saved by the Internet Archive, so each date is a window between two saved copies. Some articles date the staff-name rule to April 2026; the archived copies show it was live by 22 February 2026.
| When | What changed | Evidence |
|---|---|---|
| Between 4 and 17 June 2025 | Personal information spelled out who may be named. It already allowed names that are part of the business and public-facing professionals; it now lists doctors, lawyers, realtors, financial planners and contractors, and adds CEOs and executives whose link to the place is public. Misrepresentation now defines a conflict of interest, naming current or former employment. | 4 June and 17 June copies |
| 6 November 2025 | Google announced a form for businesses to report negative review extortion: a flood of fake one-star reviews followed by a demand for money. | Google advisory, report form page |
| Between 15 and 22 February 2026 | The group “Deceptive content & behavior” became “Fake & Misleading Content & Behavior”. A new Rating Manipulation heading took in incentivized reviews, unusual review patterns and conflict of interest, now with industry competitors and family named. It bars merchants from pressuring customers to review on the premises, from setting staff quotas for reviews, and from asking for reviews that identify a staff member. Fake Engagement now covers reviews “paid for, directly or in kind” and bars users, not only merchants, from posting on a competitor’s place. | 15 February and 22 February copies |
| 16 April 2026 | Google said it blocked or removed more than 292 million policy-violating reviews in 2025. It also said that when it sees a sudden spike of spam reviews, it removes them, pauses new reviews on the profile, alerts the owner and shows a banner. | Google blog |
| Between 16 and 21 September 2026 | Advertising & solicitation split into rules for users and for merchants. A listing may not carry time-bound promotions, pricing or direct calls to action. A merchant’s posts may not carry contact details or links unrelated to the business. | 16 September and 21 September copies |
How to use the headings that matter most
Conflict of interest: former staff, competitors and family
Conflict of interest is no longer a heading of its own. It sits under Rating Manipulation and again under Misrepresentation. Google lists current or former employment, contractual or consultory relationships, industry competitors and family. Google has to see the link, and a report that only says “ex-employee” gives it nothing to check. Our guides on reviews from former employees and competitors leaving fake reviews cover what evidence to collect.
Rating-only reviews
A one-star rating with no text gives Google nothing to judge for profanity, personal information or off-topic content. Only headings that do not depend on the text can apply: Fake Engagement, Rating Manipulation and Repetitive content, usually through a pattern across several reviews. We cover them in one-star reviews with no text.
Review bombing and extortion
A burst of one-star reviews followed by a message asking for money is extortion. Google handles it through its merchant extortion report form, not the normal flag, and tells owners not to pay or engage. Our guide to Google review extortion covers what to save and how to file.
Accusations of crimes and defamation
Offensive content covers “unsubstantiated allegations of unethical behavior or criminal wrongdoing”, and Google’s policy overview names defamatory language as a violation. A review that says your staff steal from customers fits this heading. A review that says your prices are too high does not. Separately, Google’s Legal Removals page lets anyone report content they believe breaks local law, which Google judges under that law. This is general information, not legal advice; read can you sue for a fake review and talk to a lawyer licensed in your state before you act on it.
Google review policy on employee names
A review may name an employee and criticise their work. The Personal information heading bars another person’s full name or last name posted without consent, so a review that gives a non-public employee’s full name, especially with other details, can be reported. Since June 2025 the policy also lists who may be named.
- Usually stays: “Jessica at the front desk was dismissive and didn’t help us.” A first name with criticism of the service is not one of the details the policy lists.
- Can be reported: “Jessica Moreno at the front desk is the worst.” A receptionist’s full name, posted without her consent.
- Can be reported: a review that adds a staff member’s phone number, home area, car or health information.
- Stays: “Dr. Patel rushed my appointment.” Doctors, lawyers, realtors, financial planners and contractors who work under their own names may be named.
- Stays: a review that names your CEO or another executive whose name and role are public.
- Report under Harassment instead: a threat or a sexual comment about a named employee, with or without a full name.
The employee can report the review too. The policy counts information “reported as having been posted without consent”, and tells a person who finds their own details in a review to flag it.
Asking customers to name your staff
Since February 2026, Rating Manipulation bars merchants from asking staff to collect reviews that include specific content, “including content that identifies a staff member”. It also bars setting staff a number of reviews to collect, and requiring or pressuring customers to review while they are on the premises. A card that says “Mention Jessica in your review!” now breaks the policy. A customer who names a staff member on their own does not, because the rule is about what the business asks for.
Google says Rating Manipulation content is “not allowed and will be removed from Maps”, so reviews gathered this way can come down. Its Business Profile restrictions page also lists what can follow when a business breaks the Fake Engagement policy: no new reviews for a set period, existing reviews unpublished for a set period, and a warning to customers that fake reviews were removed. Ask every customer the same way, with no script and no reward.
Names in your replies
Google checks owner replies against the same content policies before it posts them (Manage customer reviews). Do not put a reviewer’s full name, appointment details or account history in a reply. Our guide on responding to negative reviews covers the HIPAA and bar rules for medical and legal practices.
What the policy does not cover
- A real customer’s low rating with an honest complaint, written respectfully.
- A dispute about price, refunds or what was promised, where Google cannot tell who is right.
- A review you believe exaggerates, as long as it describes a real visit and breaks no heading.
- A review you already replied to. A reply does not protect the review or block a report (what if I replied to the review).
For these reviews, a calm public reply is the useful step. Google can also restore content after removing it, and its report page for Maps users lists “Content restored” as a status (see Google review appeal status). Read can removed content come back for what that means for you.
How to report a review under the right heading
- Read the review against the table and pick the one heading it breaks most clearly. If it breaks two, choose the one Google can verify from what it sees.
- Take a screenshot of the review first. The reviewer can edit it after you report.
- Report it from your Business Profile (Read reviews, then Report) or from the Reviews Management Tool. The form uses short reasons such as “Spam” or “Profanity”; choose the one closest to your heading.
- If Google finds no violation, you can send one appeal per review, with up to 10 reviews in one appeal. Our guide to Google review appeal status explains each status.
- If a review names you or a staff member in a way the policy allows, reply instead of reporting it again.
Google may tell the reviewer that their review was removed. Its removed content page says contributors “may be informed” and can see the review marked “Not posted” or “Private”. More detail is in does Google tell the reviewer. For the full process from the start, read how to remove a Google review.
Sources
- Google: Prohibited & restricted content (Maps user-generated content policy)
- Google: Maps user-generated content policy overview
- Google: About our policies
- Google: Rating Manipulation definition
- Google: Incentivized or Biased Reviews definition
- Google: Personal information definition
- Google: Report or Fix Content
- Google: Removed & rejected content for policy violations
- Google: Feature access restrictions for policy violations
- Google: Legal Removals
- Google Business Profile Help: Report inappropriate reviews
- Google Business Profile Help: Business Profile restrictions for policy violations
- Google Business Profile Help: Report negative review extortion scams
- Google Business Profile Help: Manage customer reviews
- Google: November 2025 frauds and scams advisory (6 November 2025)
- Google: New ways we’re protecting businesses on Maps (16 April 2026)
- Internet Archive: policy page on 15 February 2026, before Rating Manipulation
- Internet Archive: policy page on 22 February 2026, with Rating Manipulation