How to Report Fake Reviews on Major Platforms
How to report fake reviews on Google, Amazon, Trustpilot and your own store, what evidence each platform wants, and what to do when a report is rejected.
Last updated: October 2026 · By The Brandfeel Team
A fake review is defined as a review that does not reflect a genuine customer experience, such as an undisclosed incentivised review, competitor sabotage, a self-review or a review left on the wrong listing. Shoppers rely on reviews: 70% say they trust opinions that other consumers post online [Nielsen, 2012]. To report fake reviews successfully, cite a named platform policy and attach evidence that no matching transaction exists. You will learn the Evidence Pack, the route on each platform and what to do when a report is rejected.
- Frame every report around a named clause, because platforms remove reviews for policy breaches, never for being negative.
- Attach the Evidence Pack, an empty order search plus timestamps plus the policy clause, because it separates a successful report from a rejected one.
- Report on every surface the review appears on, because the same fraudulent review is often syndicated to two or three listings.
- Escalate once, then stop, because a public reply reaches more shoppers than a third removal attempt ever will.
- Report a sudden cluster of similar reviews as one coordinated pattern, not one by one.
To report fake reviews means submitting a policy-breach claim to the platform hosting the review, with evidence that the review does not reflect a genuine customer experience. Every major platform has a process for this, and almost every store uses it badly: teams report reviews they merely dislike, skip the evidence step, and then conclude that reporting does not work. It does work, but only on policy grounds. Omniconvert has measured how stores collect and display customer feedback across the CROBenchmark dataset of 7,000+ websites in 15+ industries, against 248+ audit criteria, over 13 years in eCommerce [CROBenchmark Report 2026, Omniconvert]. A small number of fraudulent reviews on a high-traffic listing does real damage, which is why the removal process is worth running properly. To spot them first, see how to spot fake reviews and protect your store; for a sudden burst, see how to protect your brand from review bombing.
What actually counts as a fake review
The practical test is not whether the review is true but whether a transaction happened. A customer who bought the product and hated it has written a genuine review, however wrong their reasoning. A person who never bought it has written a fake one, however reasonable it sounds.
Five patterns cover nearly everything worth reporting. Incentivised reviews, where payment or free product was exchanged without disclosure. Competitor reviews, where a rival posts to damage a listing. Self-reviews, including those posted by staff or family. Misplaced reviews, where a customer reviews the wrong product or the wrong business entirely, which is far more common than teams expect. And review bombing, where a cluster of similar negative reviews arrives in a short window, usually following a public controversy.
Reviews carry weight with shoppers, which is why a fraudulent one is worth removing: Nielsen found that 70% of consumers trust opinions posted online [Nielsen, 2012], and Bain and Company retention research puts a 5% retention lift at 25 to 95 percent more profit [Harvard Business Review, 2014].
Build the Evidence Pack before you report
The Evidence Pack is defined as the proof you attach to a report: an empty order search, timestamps and the policy clause. Most reports fail because they are an opinion. "This review is fake and unfair" gives a moderator nothing to act on. The Evidence Pack converts the claim into something checkable.
- The empty order search. Search your order system for the reviewer's display name, any email address shown, the delivery area if visible, and the product named in the review. Screenshot each search returning nothing. This is the single most persuasive artefact you can submit.
- The timestamps. Record when the review was posted and check it against your own records: was the product in stock then, had it even launched, does the review describe a variant you have never sold? A review describing a feature your product does not have is strong evidence on its own.
- The policy clause. Find the exact line in the platform's review policy that the review breaches and quote it in the report. Moderators act on policy, so handing them the clause removes the step where they have to decide which rule applies.
How to report fake reviews on each major platform
The table below maps the main surfaces, how a report is raised on each, and what tends to decide the outcome. It is qualitative rather than numeric, because published removal rates are not reliable and inventing them would be worse than useless.
| Platform | How a report is raised | What decides the outcome |
|---|---|---|
| Google Business Profile | Flag the review from the listing, then escalate through Business Profile support | Clear policy language. Off-topic, spam and conflict-of-interest cases do best. |
| Amazon | Report abuse from the review, plus a Seller Central case for a pattern | Evidence of incentivisation or a verified-purchase mismatch |
| Trustpilot | Flag the review; the reviewer is asked to supply proof of purchase | Whether the reviewer can produce documentation when challenged |
| Meta platforms | Report the recommendation or comment through the page tools | Community-standards breaches, far more than transaction evidence |
| Your own store | Moderate directly in the review app, with a logged reason | Your own published moderation policy, which you need in writing |
Google's own guidance on reporting a review that breaches policy sets out the current route and the prohibited-content list, and it is worth reading before the first submission rather than after a rejection. Shopify merchants should also check whichever review app they run, because the moderation controls differ sharply between them.
Review bombing is a different problem
Reporting twenty bombing reviews individually usually fails, because each one taken alone looks like an ordinary unhappy customer. The signal is in the aggregate: twenty reviews in six hours against a baseline of three a week, most from accounts with no other review history, many using similar phrasing.
What a marketing manager should do this week
- Export ninety days of reviews across every surface and flag anything matching the five patterns.
- For each candidate, run the order search and screenshot the empty result. Discard any that turn out to be genuine.
- Quote the platform policy clause in every submission. Never argue fairness.
- Write and publish a moderation policy for your own store's reviews, and apply it to positive and negative alike.
- Set a weekly check on your top three surfaces. A cluster caught on day one is reportable as a pattern; caught on day thirty it is just history.
- Reply publicly to anything that survives the process, because that reply is what future shoppers read.
At volume this becomes a monitoring problem rather than a reporting one. Brandfeel by Omniconvert, the sentiment-intelligence layer, aggregates reviews, NPS responses and support-ticket themes across surfaces and surfaces an abnormal cluster against the store's own baseline, which is the detection step that makes a pattern report possible while the pattern is still fresh. Brandfeel is a module inside Omniconvert Commander (formerly Nexus), the AI for eCommerce growth engine that unifies commerce data and queues the experiments most likely to close a gap, where you approve what goes live. The strategic frame sits in Omniconvert's work on customer value optimization, which treats review health as a retention input rather than a reputation chore.
Frequently asked questions
What counts as a fake review?
A fake review is one that does not reflect a genuine customer experience. That covers reviews bought or incentivised without disclosure, reviews posted by competitors, reviews written by the business about itself, reviews left on the wrong listing, and coordinated review bombing. A review that is simply harsh, unfair or factually mistaken is not fake, and platforms will not remove it. The distinction matters because reporting a genuine negative review as fake damages your credibility with the moderation team.
How long does it take for a platform to remove a fake review?
It varies widely by platform and by how clear the policy breach is. An obvious case, such as a review containing profanity or a competitor advertisement, is usually handled by automated moderation quickly. A contested case that depends on proving the reviewer was never a customer takes longer and often needs a second escalation. Plan for the process to take days rather than hours, and never let a removal request block your public response to the review in the meantime.
What should you do if a platform rejects your report?
Escalate once with better evidence, then stop and respond publicly instead. The second submission should cite the specific policy clause the review breaches rather than arguing that the review is unfair, and it should attach order records or timestamps that show no matching transaction exists. If the second attempt fails, the most effective remaining move is a calm public reply that states the facts, because that reply is read by every future shopper who sees the review.
Can you get reviews removed just for being negative?
No. Every major platform explicitly protects genuine negative reviews, and attempting to suppress them can breach the platform terms and consumer protection rules. The grounds for removal are policy breaches: no genuine transaction, conflict of interest, incentivisation without disclosure, personal information, hate speech, spam or content posted on the wrong listing. Frame every report around one of those grounds, and treat a genuine complaint as a response problem rather than a removal problem.
How do you prove a reviewer was never a customer?
Search your order records for the reviewer name, the email address, the delivery postcode and the product named in the review, and record that each search returned nothing. Note the date the review was posted and whether the product was even in stock in that period. A screenshot of an empty order search, paired with the timestamp, is the evidence most platforms respond to. Avoid speculation about who wrote it, because an unproven competitor accusation weakens an otherwise solid report.
Teams reporting fake reviews often submit the same complaint three times, worded more angrily each time. The platform sees the repeats as noise and the claim never names a policy clause, so the cases close as not violating guidelines. One report with the clause quoted and the order search attached does more than three emotional ones.
In August 2024 the US Federal Trade Commission announced a final rule that bans fake or false reviews and testimonials, paying for reviews based on their sentiment, and certain forms of review suppression, and that lets the agency seek civil penalties against knowing violators. [Federal Trade Commission, 2024]
Report on policy grounds, then answer in public
Reporting a fake review works when it is a policy case and fails when it is a complaint. Establish first that no transaction happened, then build the Evidence Pack: the empty order search, the timestamps, and the exact policy clause the review breaches. Report on every surface the review appears on, handle a coordinated cluster as one pattern submission rather than twenty flags, and escalate a rejection once before switching to a calm public reply. The durable protection is upstream of all of it: a steady flow of genuine reviews dilutes an attack that would otherwise define the listing, so the collection system matters more than the removal process.