What to do about a review you believe is fake

That is not the answer owners want. The instinct is that a lie about your business should come down, and there should be somebody to call. There is not, and the sooner you accept that, the sooner you stop burning weeks on it.

What you can control is the ratio. One fabricated review sitting on top of a profile that has not moved in a year is a real problem. The same review sitting under fifteen new genuine ones is a footnote.

How do I tell fake from just harsh?

Ask whether the review describes something that could not have happened, not whether it is unfair.

A review is probably fabricated if it describes a service you do not offer, a staff member who does not exist, a location you do not operate, or a policy you have never had. Those are checkable facts.

Other signals worth noting. The account has a name and no history, or a burst of reviews across unrelated businesses on the same day. The wording matches reviews appearing on your competitors. Several arrive within a short window with no corresponding change in your operation.

A review is probably genuine and simply harsh if it describes a real night in unflattering terms, gets details roughly right, or complains about something you know is a weak point. The uncomfortable version of this is that most reviews owners are certain are fake turn out to be a real customer you do not remember.

Check your own records before reporting. Reservation system, POS, appointment log, whatever you keep. If the visit happened, the review is real, and you should answer it rather than fight it.

What is the actual reporting process?

Flag it, state a policy violation, and expect nothing.

Google is explicit about the standard it applies. Its own help documentation on removing reviews says you can report any review, but only those that violate Google policies are eligible for removal, and it adds a sentence most owners have never read: do not report a review just because you disagree with it or dislike it, because Google does not get involved in conflict between businesses and customers.

Read that as a filing instruction. Your report has to name a policy, not an injustice. The two grounds that actually produce removals are the two that are checkable without hearing your side.

Off topic. Google's prohibited content policy does not allow content containing general, political, or social commentary or personal rants. A review that argues with a news story, a sign in your window, or something the owner said publicly is off topic on its face, and that is a much stronger filing than "this never happened."

Conflict of interest. Google's policy names current or former employment, a contractual relationship, and other professional or personal affiliations, including industry competitors and family relationships. A review from a fired employee or a competitor is removable on a named ground, and this is the ground most owners never think to use.

The escalation path runs flag from the profile, then the Reviews Management Tool, then an appeal covering ten reviews at a time. The appeal is described as one time and that word is literal. Do not spend it on a weak argument you filed while angry.

Report once. Repeated flagging from multiple accounts does not increase your odds and can look like coordinated behavior.

If the review arrived with a demand attached, that is a different thing entirely. Google runs a dedicated channel for negative review extortion scams, and restaurants are the most frequent targets of them. Pay nobody, document everything, and file it there rather than as an ordinary flag.

If you have genuine evidence of a coordinated campaign, that is a different situation with a different playbook, and a review bomb should be handled as an incident rather than as a series of individual disputes.

What should I write in the response?

Something short that lets a reader draw their own conclusion.

Do not call the reviewer a liar. Do not say "we have no record of you," which sounds defensive even when it is accurate and sometimes invites a longer argument in the replies.

The structure that works: acknowledge, state a fact neutrally, offer a real contact.

"We take this seriously and we're not finding a visit matching this. If we got something wrong, please email us at the address on our profile so we can look into it properly."

That does the whole job. It tells the reader you looked. It does not accuse anyone. And it makes the reviewer's silence, if they never respond, part of the record.

Healthcare and other regulated fields need a different construction entirely, because even acknowledging that the reviewer might be a patient is a disclosure. Those responses have to be written with the privacy rules first and the reputation second.

Is threatening the reviewer an option?

No, and since October 2024 it is the one move that converts your problem into a federal one.

The old reputation management playbook was a boilerplate defamation letter fired at the reviewer, on the theory that most people fold rather than hire a lawyer. That tactic now has a rule written against it.

The FTC's Rule on the Use of Consumer Reviews and Testimonials, 16 CFR Part 465, took effect on 21 October 2024. Section 465.7(a) makes it unlawful for anyone to use an unfounded or groundless legal threat, a physical threat, intimidation, or a public false accusation to prevent or remove a review. The pre 2024 playbook is now itself the violation. Counsel reporting at finalization put civil penalties for knowing violations at $51,744 per violation, adjusted annually for inflation, so confirm the current figure before anyone relies on it.

The phrase "public false accusation" is worth sitting with, because it reaches the public reply as well as the letter. Posting "this reviewer is a competitor" under a review, when you cannot show it, is the accusation the section names.

California operators carry a second layer. Civil Code section 1670.8 makes it unlawful to include a clause waiving a consumer's right to comment in a contract for consumer goods or services, to threaten to enforce one, or to otherwise penalize a customer for a statement. Inherited intake or membership boilerplate that says a client agrees not to publish disparaging remarks is void, and its presence alone is the violation. Threatening to enforce it is both a California violation and a groundless federal legal threat at the same time.

The rule cuts the other way too. Section 465.2 makes it unlawful to write, create, sell, buy, or disseminate a review that misrepresents that the reviewer exists or that the reviewer had experience with the business. If a review really is fabricated, the liability sits with whoever produced it. That does not get it removed any faster, and it is not a reason to send a letter.

What actually fixes it?

New reviews.

A fake one star review moves your average because your denominator is small. Increase the denominator and the same review stops mattering. This is arithmetic, not strategy, and it works on a timeline you control.

Here is the arithmetic, with numbers you can replace with your own.

A profile carries 64 reviews at a 4.7 average. The stars sum to 300.8. One fabricated one star lands, so the profile now holds 65 reviews summing to 301.8, and the displayed average falls to 4.6. That is a tenth of a star.

To climb back above 4.7 you need new five star reviews. Solve for the count: 301.8 plus five per new review, over 65 plus the count, has to clear 4.7. That works out at 13 new reviews. At a cadence of five a month, you are back inside three months, and you got there by doing the thing you should have been doing anyway.

Now run the same fake review against a thin profile. 12 reviews at 4.8 sum to 57.6. The one star drops the displayed average to 4.5, three times the damage, and clearing 4.8 again takes 19 new reviews. Same review, same arithmetic, completely different exposure.

Your review count is the shock absorber. A fake review hurts in exact proportion to how few real ones you have.

Dilution also fixes the ordering problem. Reviews sort so that recent activity is visible, which means a stream of new genuine reviews pushes the fabricated one down the page where fewer people read it.

There is a ranking cost to the alternative that owners do not see coming. Whitespark's Darren Shaw, who runs the annual Local Search Ranking Factors survey, wrote in May 2025 that the moment you stop getting new reviews, your local rankings start to slip. Recency counts regardless of sentiment. So the three weeks you spend on flag reports and support forums, collecting nothing, cost you position on top of the review you were angry about.

The durable version of this is a collection process that keeps running whether or not the owner is currently upset about something. That means it cannot depend on one motivated manager, which is the whole argument for building a review system that survives a staff change.

Where does the dilution advice break down?

Three places, and the second one is the one that catches people.

When the profile is too thin to dilute anything. Below roughly twenty reviews, every new review moves the average visibly and a single one star is a real wound. The answer is still volume, it just takes longer, and pretending otherwise is how owners lose faith in the plan by week three.

When you try to dilute too fast. Google's Rating Manipulation policy names content exhibiting unusual volumes or patterns of review contributions as a violation in its own right. Blasting two years of customers to bury one fake review is the exact pattern the policy describes. Google can respond with a public consumer alert banner on your profile, a posting restriction, or hidden reviews, and a posting freeze stops your recovery dead while competitors keep accruing. Ramp, do not blast.

When the fake reviews keep coming. Dilution assumes a fixed problem and a growing denominator. If new fabrications arrive weekly, you are not managing a review, you are managing a campaign, and the escalation and documentation path matters more than the cadence.

What to do this week

Check your records to confirm the review is actually fabricated. Do this first, because it changes everything that follows.

If it is, flag it once with the correct violation category and write the neutral response. Off topic and conflict of interest are the two grounds worth arguing.

Then set a weekly review request target and start it immediately, this week, while you are still annoyed. The annoyance is useful energy and it is about to be wasted on a support form otherwise.

Do the two line arithmetic above with your own review count and average. Knowing that the fix is thirteen reviews rather than an unknown quantity of outrage is what makes people actually start.

While you are in the profile, fix the things that quietly cost you visibility and take ten minutes. Hours and holiday hours left wrong will cost you more customers this quarter than one fake review will.

Questions about a review you think is fake?

Email me at eric@seod.com with the review text and the reviewer's profile name. I will tell you whether the pattern looks fabricated or looks like a customer you do not remember, which violation category to file it under, and give you the response to post. I will also tell you when the honest answer is that it is real and you should answer it.

I would rather save you three weeks on a support forum than let you find that out yourself.

Otherwise keep reading the reviews and reputation library.