You check the stars before you buy — almost everyone does. That is exactly why fake reviews work. Studies estimate that around 30% of all online reviews may be fake or manipulated, and on some product pages nearly half of the bestselling-item reviews have been flagged as unreliable by third-party analysis. Google alone blocked or removed more than 240 million policy-violating reviews in 2024 and over 292 million in 2025. Amazon removed more than 275 million suspected fake reviews in 2024. The stars are not a verdict. They are a starting point — and this guide teaches you how to read past them.
The method is the same on Amazon, Google, Trustpilot, Tripadvisor, and TikTok Shop: do not judge one review. Judge the pattern — the reviewer, the wording, the timing, and the shape of the ratings. No single sign proves a review is fake. Two or three together mean do not trust the rating. That rule protects you twice: it stops you buying on a lie, and it stops you accusing an honest business on thin evidence.
1 — Start with the scale of the problem, not one bad review
Fake reviews are an industry, not an accident. Sellers buy five-star floods to launch products, competitors buy one-star attacks to damage rivals, and brokers sell both by the hundred. The US Federal Trade Commission's Consumer Reviews Rule, in force since October 21, 2024, now bans creating, buying, or selling fake reviews — including AI-generated ones — with civil penalties of up to $53,088 per violation. In Europe, the EU's Omnibus Directive requires marketplaces to say whether reviews come from real buyers, and the UK's Digital Markets, Competition and Consumers Act 2024 allows penalties of up to 10% of turnover for fake-review practices. The law is on your side. But enforcement removes fakes after they have already influenced buyers — which is why your own two-minute check still matters before you spend $20, €20, or £20.
The most useful numbers to carry in your head: Google's fake-review rate is estimated at around 10.7%, the highest of any major platform, followed by Yelp at 7.1% and Tripadvisor at 5.2%. On Amazon, third-party analysis has flagged up to 43% of bestselling-product reviews as unreliable. And 54% of consumers say they will not buy from a business once they find fake or fraudulent reviews about it. The question is never "are there fake reviews here?" There almost certainly are. The question is whether the rating you are looking at is built on them.
2 — Check the reviewer before you believe the review
On Google, you can click a reviewer's name and see their whole public history. On Amazon, you can click a reviewer profile too. This is the single strongest check in this guide, because fake reviewers cannot fake a plausible life.
The fake fingerprints:
One review, nothing else. An account whose entire history is a single glowing review of this product or business — no photo, no other reviews, no answers — deserves suspicion. Real one-review customers exist, so this alone is not proof, but it is where every investigation starts.
Impossible geography. A genuine customer reviews businesses near where they live and travel. An account that reviewed a dentist in Chicago, a mechanic in Manchester, and a restaurant in Berlin within days is not a typical customer — it is a paid account working through a list.
Extreme-only history. A profile that posts nothing but five stars, or nothing but one stars, with no middle ground, reads as manipulation. Real reviewers leave a mix — a 4 here, a 3 there, a complaint softened by praise.
Unrelated categories at speed. Power tools, collagen powder, pet toys, and suitcases reviewed in one week is not a shopping life. It is a brief.
A stock-looking photo on a thin profile. Fake accounts often use a lifted or modelling-style photo. A polished headshot on an account with one review is a mismatch worth noting.
What a trustworthy profile looks like: several reviews over months or years, mostly local or in one hobby, with photos of actual visits or purchases, a mix of ratings, and specific detail. If the five-star reviews you are reading come mostly from accounts that fail this check, the rating is not what it seems.
3 — Read the wording like an editor, not a shopper
Fake reviewers usually never used the product or visited the business, so their writing leaks that fact in small ways. Read a handful of reviews slowly, the way an editor would.
Vague praise or vague rage. "Amazing service!!! Best place ever!!!" or "Terrible, never coming back" — with no dish, no staff name, no room, no feature described — could apply to any business on earth. Real customers mention specifics: what they ordered, who served them, what went wrong, what it cost.
Wrong details. The single most reliable wording tell: a review that describes a service the business does not offer, a staff member who does not work there, or a product it does not sell. One factual mismatch outweighs ten vague compliments. A classic example: a review praising the craft cocktails at a vegan bakery with no bar.
Copy-paste language. Several reviews using the same sentence structures, the same unusual phrases, or the same stock wording — "highly recommended," "best ever," "fantastic service," "I was skeptical but…" — repeated verbatim point to one writer producing many reviews. Openings like "as someone who…" and "product was a game-changer" stacked with generic benefits (quality, value, convenience) are template tells.
Flawless grammar with zero detail. Polished, generic, perfectly structured praise that says nothing concrete reads like it was written to a brief — because it was.
Counterpoint to keep you honest: some real customers write badly, briefly, or emotionally. Bad writing alone is not fakery. Vague plus wrong-details plus copied wording together is.
4 — Look at the timing: bursts beat drips
Real reviews arrive as a steady drip — a few a week for a busy restaurant, a few a month for a quiet tradesperson, a steady stream for a popular product. Fake reviews arrive as a burst.
A sudden wall of five stars. Fifty reviews in a week for a quiet local shop, then nothing for months, usually means a purchased batch. Sort by most recent and scroll the dates: if the rating was built in one week six months ago and has been quiet since, be sceptical.
A sudden wall of one stars. The same pattern in reverse is a coordinated attack — often a competitor, an ex-employee, or an extortion attempt. Several negative reviews within hours, with similar phrasing or the same unusual detail, is the cluster signature.
Reviews before the product existed. Reviews dated before a product was widely available, or dozens of reviews landing within 24–48 hours after weeks of silence, are a promo or pay-for-review campaign, not organic demand.
Reviews after a specific incident. An unusual influx right after a news story, a viral post, or a dispute often means people who were never customers piling on. Discount them before any platform acts.
Use the dates, not just the average. A 4.8 built steadily over three years is a different business from a 4.8 built in one fortnight.
5 — Read the shape of the ratings, not just the average
The average star rating is the least informative number on the page. Open the rating breakdown and look at its shape.
Almost all five stars, almost no middle. A distribution that is 95% five-star with barely a 4, 3, or 2 in sight is the classic manipulation shape. Real businesses annoy someone sometimes; real rating curves have a middle.
A U-shape with nothing in between. Hundreds of five stars and dozens of one stars, but almost no 2s, 3s, or 4s, usually means two campaigns fighting — bought praise versus a hostile burst — with real customers squeezed out.
The middle is where the truth lives. Read the 3-star reviews first. They are where real buyers hide: specific, balanced, annoyed about one thing and happy about another. If the 3-star reviews describe a different business from the one the 5-star reviews describe, trust the middle.
Cross-check one other platform. A business at 4.9 on Google and 3.2 everywhere else has a Google problem, not a customer problem. Check Trustpilot, Yelp, Tripadvisor, or a booking site and see whether the shape matches.
A useful habit for products: compare the verified-purchase reviews against the unverified ones. On Amazon, a listing whose unverified reviews are glowing while its verified-purchase reviews are middling is telling you exactly where the manipulation sits.
6 — Spot the AI-generated reviews
AI can now write thousands of convincing five-star reviews in minutes, and they are spreading across Google, Amazon, and Trustpilot alike. Surveys show 87% of consumers struggle to tell AI-written reviews from human ones, and 46% suspect a review is fake when it reads like it was written by AI. The tells:
Too perfect, too generic. AI reviews are grammatically flawless, warmly enthusiastic, and oddly empty — every sentence sounds like a brochure, none sounds like a person who was actually there.
The same structure, many times. AI batches repeat a template: an opening compliment, two vague positives, a recommendation, an exclamation. Read five in a row and the rhythm gives them away.
Praise without a scene. A human writes "the waiter forgot our starters but the manager fixed it." An AI writes "the staff were wonderful and the atmosphere was amazing" — no scene, no incident, no friction.
Suspicious polish on a throwaway profile. A one-review account writing like a copywriter is a double signal: fake profile plus AI text.
If a review reads strangely polished or oddly generic, paste the review text into our fake review checker to test it for AI-generated or templated patterns before you trust the stars. Our checker works on review text you paste in — it does not fetch product or business links, and it cannot look up a reviewer's history, so use it alongside the profile and timing checks above, not instead of them.
7 — Platform by platform: where each check matters most
The same method bends slightly by platform, because each one hides and reveals different things:
Amazon. Check verified-purchase status, reviewer profiles, and review timing first; watch for review hijacking, where reviews for one product are pooled onto another (the "zipper broke on the backpack" review sitting on a phone-case listing). Our companion guide on the Amazon fake reviews checker covers the tool side in depth.
Google. Profiles are your superpower — click every suspicious name. Our companion guide on how to tell if a Google review is fake covers the shopper-side checks for local businesses, and our owner-side guide covers reporting and removal for businesses.
Trustpilot. Trustpilot removed 4.5 million reviews in 2024 (about 7% of its total). Read the 1- and 2-star reviews for patterns (refund refusals, wrong items) rather than one-off rants, and compare the Trustpilot score with Google before deciding.
Tripadvisor and booking sites. Tripadvisor removed 2.7 million reviews in 2024. Watch for bursts around a single property and for reviews that describe a different hotel or restaurant than the photos show.
TikTok Shop, Shein, and Temu. On marketplace apps, the listing review and the seller are different things: a five-star flood on one listing says nothing about the seller's other listings. Our Temu guide covers the marketplace-specific traps — coupon-code reviews, photo-less praise floods, and all-five-star reviewer profiles.
Small websites and Facebook ad stores. If you have never heard of the store, check the store before you check its reviews — a fake store's reviews are fake by definition. Run its domain through the website scam checker first.
8 — The two-minute routine, and what to do when you spot fakes
Put the checks together into one quick routine, in order, before you spend money — a meal, a hotel night, a product over $20 or €20, a deposit to a tradesperson:
1. Open the rating breakdown. If it is nearly all five stars with no middle, slow down.
2. Click three five-star reviewers. One-review accounts, impossible geography, or extreme-only histories on two of the three is a red flag.
3. Read five 3-star reviews. If they describe a worse, more specific business than the five stars do, believe the middle.
4. Scan the dates. A burst beats a drip — a wall of reviews in one week is a batch, not a reputation.
5. Cross-check one other platform. Trustpilot, Yelp, Tripadvisor, or a booking site — a big gap between platforms is information.
6. Paste anything suspicious. If a review reads like AI or a template, paste the review text into our fake review checker before you trust it.
When you spot fakes, act on it correctly. As a shopper, report the review through the platform's own reporting route — Google Maps or Search, Amazon's report link, Trustpilot's flag — picking the reason that matches the violation (spam or fake content for a reviewer who was never a customer, conflict of interest for a competitor or employee). Business-owner reports carry more weight, but user flags still feed the filters. If the business itself looks fake, report it to the FTC at ReportFraud.ftc.gov in the US, or to your national consumer authority in Europe. Do not accuse publicly on thin evidence: a single vague review is not proof of a bought campaign, and a public accusation against an honest small business does real damage. Report quietly, let the pattern make the case, and spend your money with the business whose reviews survive the same routine.
The stars are a starting point, not a verdict. A product or business that survives these checks has earned your order. One that fails three of them has told you everything you need to know — before you spend a dollar, a pound, or a euro.
FAQ
Q: How can you tell if a review is fake?
A: Stack the signals: check the reviewer's profile (one review, impossible geography, extreme-only history), read the wording (vague, wrong details, copy-paste phrasing), check the timing (bursts, not drips), and read the rating shape (almost all five stars with no middle). Two or three signals together mean do not trust the rating; one alone means look closer.
Q: What percentage of reviews are fake?
A: Studies estimate around 30% of all online reviews may be fake or manipulated. Google's fake-review rate is estimated at about 10.7%, Yelp at 7.1%, Tripadvisor at 5.2%, and third-party analysis has flagged up to 43% of bestselling-product reviews on Amazon as unreliable. Platform removal figures differ from prevalence — Google removed over 292 million policy-violating reviews in 2025, Amazon over 275 million suspected fakes in 2024.
Q: Can you spot an AI-generated review?
A: Often, yes: AI reviews are grammatically flawless but oddly empty — brochure-style praise with no scene, no incident, and no specifics — and they repeat the same structure across many reviews. If a review reads that way, paste the review text into our fake review checker to test it for AI-generated or templated patterns.
Q: Are fake reviews illegal?
A: Yes, on both sides of the Atlantic. The US FTC's Consumer Reviews Rule (in force since October 21, 2024) bans creating, buying, or selling fake reviews, including AI-generated ones, with civil penalties up to $53,088 per violation. In Europe, the EU Omnibus Directive requires marketplaces to disclose whether reviews come from real buyers, and the UK's 2024 consumer protection act allows penalties up to 10% of turnover.
Q: Should you trust a product with only 5-star reviews?
A: Be cautious. A rating distribution that is almost entirely five stars with almost no 2-, 3-, or 4-star reviews is the classic manipulation shape. Real products collect a middle. Read the 3-star reviews and cross-check the product on another platform before buying.
Q: What should you do when you spot a fake review?
A: Report it quietly through the platform's own reporting route, picking the reason that matches the violation. Do not accuse publicly on thin evidence. If the business itself looks fake, run its domain through the website scam checker and report it to the FTC (US) or your national consumer authority (Europe).
Sources: wiserreview.com "45 Fake Review Statistics Every Business Must Know (2026)" (around 30% of reviews fake/manipulated; up to 43% of bestselling-product reviews flagged unreliable; Google 10.7%, Yelp 7.1%, Tripadvisor 5.2%; Google 240M removed 2024; Amazon 275M removed 2024; Trustpilot 4.5M removed, 7%; Tripadvisor 2.7M removed; 46% of fake reviews are 5-star); shno.co "Reviews and Ratings Statistics for 2026" (87% struggle to tell AI reviews from human ones; 46% suspect a review is fake when it reads AI-written; 54% will not purchase after finding fake reviews; 40% found fake Google reviews in 2025); reviewflowz.com "112 Online Review Statistics (2026)" (Google 292M+ policy-violating reviews blocked/removed in 2025, 221M under fake-engagement policy alone; 782,000+ accounts restricted, 13M+ fake Business Profiles removed in 2025); legalclarity.org / auditsocials.com / merchantfraudjournal.com FTC Consumer Reviews Rule coverage (16 CFR Part 465, effective October 21, 2024; bans fake reviews including AI-generated, buying/selling reviews, insider reviews without disclosure, review suppression, fake social indicators, company-controlled review sites; civil penalties up to $53,088 per violation; FTC warning letters to 10 companies December 22, 2025); elitecontentmarketer.com "59 Online Review Statistics for 2026" (US penalties up to $53,088; UK penalties up to 10% of turnover under the Digital Markets, Competition and Consumers Act 2024); trypriceai.app "How to Spot Fake Reviews Before You Buy (and What the FTC Bans)" (FTC red flags: bursts, single-review accounts, too-positive reviews, fake negative reviews; FTC checks: dates, reviewer history, source, incentives); consumeraffairs.com "How to spot fake and AI-created reviews: the 60 second test" (same-wording openings, 24–48h bursts, cross-category profiles, wrong-item descriptions); socialpilot.co fake Google reviews guide (wrong-detail example, copy-paste signals). All facts verified in live search results on 2026-10-25; no invented figures; search volume unverified throughout.
Explore the evidence behind your next question.
Start a website check