Fake Review Checker: 7 Ways to Tell if Reviews Are Real

By Robinson Guerra, Founder · Published 2026-08-07 · Updated 2026-10-09

To check if reviews are fake, use this 7-point fake review checker: look at the language, the reviewer's profile, the timing, the photos, the platform's own signals, other sources, and what the "verified" badge actually verifies. Timing anomalies and reviewer profiles are the hardest signals to disguise. No single sign proves a review is fake, but several together are strong evidence.

You can run the checks on your own product pages, a competitor's listing, or a review platform you're considering. Want to check a single review instantly? Paste it into our free fake review checker tool. It scores the writing-pattern risk signals in seconds.

How to spot fake reviews: quick checklist

The old tells (broken English, repetitive phrasing, generic praise) are losing their value. In two experiments published in Marketing Letters in April 2024, researchers found that "humans cannot recognize AI-written reviews," and AI detectors were fooled too. Each AI review can be grammatically clean and different in style from the last. So before you run the full 7-point method below, start with the signals that still hold up:

• Ask who verified the review, not whether it's verified. A badge is not proof. The source behind it is. Check 7 below shows how to tell a merchant-supplied badge from a processor-backed one.
• Look at the shape of the ratings, not the average. Good products usually collect some low ratings too, from shipping mishaps, defective units or wrong expectations. A product with 500 reviews and a 4.99 average is statistically improbable. So is a profile with no negative reviews at all.
• Read the middle ratings first. The 2-, 3- and 4-star reviews are the most likely to be genuine (see below).
• Watch for bursts. 50 reviews on one day, then three months of silence, is a warning sign. Organic review velocity is fairly smooth.
• Compare review count to market position. If you're sizing up a competitor, a product with far more reviews than its market position suggests, plus an unusually high average, often has a manufactured profile. The economics of fake reviews make that pattern easy to produce and hard to hide.

Read the 2-, 3- and 4-star reviews first

Fake review operations overwhelmingly target 1 star (to attack competitors) and 5 stars (to boost products). Of the 4.5 million fake reviews Trustpilot removed in 2024, 3.4 million were 5-star and 627,000 were 1-star, against 70,000 at 3 stars (Trustpilot Trust Report 2025). Middle ratings come from real customers with mixed experiences: "Product works but setup was confusing," "Good value but shipping was slow." If a product has hundreds of 5-star reviews and zero 3-star reviews, the 5-star reviews are suspicious. Not because each one is fake, but because the distribution doesn't match how real customers behave.

AI-written reviews: the tells that still work

• Uniform sentence length. People vary sentence length naturally. AI text tends toward steady, mid-length sentences.
• No specific details. Compare "This product is great, highly recommend" with "The zipper broke after three weeks but customer service sent a replacement in two days." AI is bad at invented specifics.
• Perfect grammar, no typos. Real reviews contain occasional typos. 200 flawless reviews with zero typos is statistically unlikely.
• Generic enthusiasm. "Changed my life!" "Best purchase ever!" "You won't regret it!" Real reviews are more measured, even when positive.

No single tell is conclusive on its own. The fake review checker tool scores a review against these patterns at once and shows which signals fired.

The 7-point fake review checker

1. Language pattern analysis

Fake reviews tend to follow detectable language patterns. Run this checklist against any review:

• Overly generic language: "Great product," "Highly recommend," "Excellent service": with no specifics about the product, the purchase experience, or what made it good
• Repeated phrases: The same phrase or sentence structure appears across multiple reviews from different "reviewers"
• Marketing-speak: The review reads like ad copy, using phrases like "game-changing," "next-level," or "best-in-class" without concrete examples
• No personal context: The reviewer doesn't mention how they used the product, what problem it solved, or why they chose it
• Unnatural grammar: Awkward sentence construction that doesn't match typical customer language, often a sign of AI-generated or template-based fake reviews
• All 5-star or all 1-star: Real review distributions include middle ratings. A product with only 5-star and 1-star reviews (no 2, 3, or 4) is a red flag. See fake review statistics 2026 for the latest data.

Red flag if: 3 or more of these are true.

2. Reviewer profile investigation

Click through to the reviewer's profile (if the platform allows it). Look for:

• No profile photo or a stock/generic image, reverse image search the photo if suspicious
• Account created recently (within days or weeks of posting the review)
• Only one review ever posted, especially suspicious if it's a glowing 5-star or a brutal 1-star
• Burst of reviews in a short window. The reviewer posted 5+ reviews on the same day, often for products in completely different categories
• Reviewer name matches a pattern, "John S." "Sarah M." "Mike T." (first name + last initial is a common template for fabricated profiles)
• All reviews follow the same sentiment, either all 5-star or all 1-star across every product. Real reviewers have mixed experiences.

Red flag if: 2 or more of these are true.

3. Timing analysis

The timing of reviews tells a story. Check:

• Review spike in a short window: 20+ reviews posted in a single day or weekend, especially after a period of zero reviews
• Reviews posted before product launch: Self-explanatory. The reviews can't be genuine if the product wasn't available
• Reviews on the same date across products: The same "reviewer" posted reviews for different products on the exact same date, a sign of a paid review campaign
• No reviews during known purchase periods: If the business had a major sale or promotion and review volume didn't increase, existing reviews may not reflect real purchase patterns
• Reviews stop abruptly: A product with weekly reviews that suddenly drops to zero: could indicate the end of a paid review campaign

Red flag if: 1 or more of these are true. Timing anomalies are one of the strongest fake review signals.

4. Photo and media verification

Photos can be faked, but most fake reviewers don't bother. Check:

• Review includes original photos: Reviews with authentic-looking, non-stock photos are more likely to be real, but reverse image search them to be sure
• Photos show the product in context: A real customer's photo usually shows the product in their home, on their desk, or in use, not a perfect product shot
• Same photo appears in multiple reviews: A clear sign of fabrication. The same "customer photo" used by different "reviewers."
• Photos are too polished: Professional lighting, perfect composition, multiple angles, most real customers snap one quick photo, not a photoshoot
• Video reviews: Video reviews are much harder to fake than text or photos. A product with multiple video reviews is less likely to have a fake review problem.

Red flag if: 2 or more of these are true.

5. Platform signals

The review platform itself provides signals. Look at:

• "Verified Purchase" badge: If present, what does it actually mean? Click it. "Verified Reviewer" usually means email verification only. "Verified Purchase" should mean the platform confirmed a transaction.
• Platform reputation: Is the review platform known for aggressive fake review prevention? Or is it a free-for-all with no verification?
• Review moderation policy: Does the platform publish its moderation policy? Do they explain how they detect and remove fake reviews? Transparency is a positive signal.
• Can anyone post? Platforms that allow anyone to review any business without proof of purchase are inherently vulnerable to fake reviews, both positive and negative.
• Does the platform use cryptographic signing? A few platforms cryptographically sign reviews so the content is tamper-evident. If a review is signed, look for a public verification page where you can confirm it hasn't been altered after submission.

Red flag if: The platform has no verification mechanism and allows unrestricted posting.

6. Cross-reference with other sources

Don't rely on one platform's reviews alone:

• Check multiple platforms: Does the product have similar ratings and review patterns on Google, Amazon, Trustpilot, and the business's own site? Dramatic differences suggest manipulation on one platform.
• Check the business's response to negative reviews: Real businesses engage with criticism. Businesses that only respond to positive reviews, or that argue with every negative reviewer, may be managing their reviews aggressively.
• Look at the business's overall online presence: How long have they been operating? Do they have a physical address? Are they active on social media with real customer interactions? A business that only exists on a single review platform is suspicious.
• Check for review removal patterns: Use tools like the Wayback Machine or review monitoring services to see if reviews were removed. Legitimate businesses remove very few reviews (only those that violate policies). Removing a large number of negative reviews is a strong manipulation signal.

Red flag if: 2 or more of these are true.

7. The verification badge deep-dive

Verification badges are not all equal. When you see a "Verified" badge:

• What does the badge actually verify? Click it. Read the tooltip or linked page. If it says "This reviewer confirmed their email address," that's email verification, the weakest form.
• Who is the verifying party? Is it the review platform itself, or an independent third party (payment processor, identity provider)? Self-attested verification is weaker.
• Is there a transaction reference? The strongest verification badges tie each review to a specific transaction record from the payment processor (e.g., a Stripe charge such as ch_3QabcDEF...), not to a customer list. Check whether the badge links to a page that shows what was matched.
• What happens on refund? If a verified review stays up after the customer gets a refund, the verification is misleading. Transaction-based verification should automatically flag or hide refunded purchases.
• Can the business remove reviews unilaterally? If the business can delete any review they don't like, the "verified" badge loses meaning: verification should protect reviews from removal as well as fabrication.

Merchant-supplied or processor-backed? Most "Verified Buyer" badges sit at Level 3 on the verification spectrum: the business said this person was a customer. Trustpilot invited reviews, Yotpo, Judge.me and most Shopify apps work this way. Level 4 ties each review to a charge record from the merchant's payment processor instead of a customer list the merchant typed in. Only a handful of platforms operate at Level 4. If a platform doesn't make the distinction clear, ask why.

Red flag if: The badge provides no specific information about what was verified or by whom.

Fake review checker scorecard

Use this scorecard to evaluate any set of reviews. Tally the red flags:

Score interpretation:

• 0–4 red flags: Reviews are likely genuine. Minor anomalies are normal, real review patterns aren't perfectly smooth.
• 5–9 red flags: Mixed signals. Some reviews may be fake. Investigate further, especially the timing and reviewer profile checks.
• 10–16 red flags: Strong evidence of fake reviews. Multiple dimensions confirm suspicious patterns.
• 17+ red flags: Almost certainly a manipulated review profile. The product or business has systematic fake review activity.

What to do if you find fake reviews

If they're on your competitor's product

Document what you found. If the competitor is using fake reviews to gain an unfair advantage, you may be able to:

• Report them to the platform (most platforms have fraud reporting mechanisms)
• Report them to the FTC (its 2024 Consumer Reviews and Testimonials Rule bans fake reviews)
• Use it as competitive intelligence, if their reviews are fake, your genuine verified reviews are a stronger trust signal

See FTC fake reviews rules: what every business owner needs to know for the legal landscape.

If they're on your own product

Fake positive reviews can appear on your product pages without your knowledge: from overzealous marketing agencies, well-meaning employees, or even competitors trying to make you look bad. If you find them:

1. Don't ignore them. Fake reviews on your pages can trigger FTC penalties even if you didn't commission them.
2. Report them to the platform for removal.
3. Audit your review sources. If you use a review collection service, verify their practices.
4. Switch to transaction-verified reviews. When every review is tied to a real, completed Stripe charge, no one can post without buying first, and each fake costs real money and leaves a payment record.

How to prevent fake reviews on your own pages

The best defense against fake reviews is structural, not procedural. Instead of trying to catch fakes after they're posted, use a system where nobody can post a review without a real purchase:

1. Use transaction-verified reviews: Require a real, settled payment before a review can be submitted. This makes fake reviews economically irrational, each one costs real money.
2. Use cryptographic signing: Sign every review so the content is tamper-evident. Anyone can check on a public verification page that the review hasn't been altered.
3. Automate refund handling: When a customer gets a full refund or files a dispute, their review should automatically come down. Manual processes fail.
4. Publish your verification methodology: Be transparent about how your reviews are verified. The more specific, the more credible.

Signed Reviews implements all four of these by design. See how Stripe review verification works for the technical architecture, or visit the pricing page for plan details.

Show shoppers your reviews are real

Shoppers are getting more skeptical. Beyond keeping fakes out, make your real reviews easy to check:

• Show the chain, not just the word "verified." Purchase, charge record, invitation, review, signature. The more links a shopper can check, the more each review is worth. How Signed Reviews works shows one example.
• Say who can review. Publish who can leave a review, how you check it, and what your incentive policy is. A clear review policy is a trust signal on its own.
• Display verified and unverified reviews separately. If you collect reviews from several sources, don't mix them. A verified review loses its trust value when it sits among unverified ones.
• Keep negative reviews up unless they break your policy. A 4.5 average with some thoughtful 2-star reviews is more credible than a 5.0 with no complaints. Answer them professionally. They are a conversion asset, not a liability.

FAQ: fake review checker

Can AI detect fake reviews?

Yes, to an extent. Machine learning models can identify language patterns, reviewer behavior anomalies, and timing irregularities that humans miss. However, AI-generated fake reviews are getting better: large language models can now produce reviews that are nearly indistinguishable from human-written ones. The arms race between fake review generators and detectors means no automated system is 100% reliable. The best approach combines automated detection (patterns, timing, metadata) with human judgment (context, nuance, specific details). See the how to spot fake reviews checklist above for the AI-writing tells that still work.

What's the most reliable sign of a fake review?

Timing anomalies are the hardest to disguise. A product with 30 reviews posted on the same day after months of silence is suspicious regardless of how well-written the reviews are. Reviewer profile patterns (account created the same day as the review, no other review history) are also strong signals because they're hard to fabricate at scale without being obvious.

Are all negative reviews fake?

No. Genuine customers leave negative reviews. In fact, a profile with only positive reviews is more suspicious than one with mixed sentiment. The question is whether the negative reviews follow the patterns above. A detailed, context-rich 2-star review from a reviewer with a genuine profile history is likely real. A one-sentence 1-star review from a newly created account is suspicious.

How does transaction verification prevent fake reviews?

Transaction verification requires a real, settled payment through a regulated payment processor (Stripe) before a review can be submitted. Each fake review would cost real Stripe processing fees (2.9% + 30¢ per domestic card charge on Stripe's standard pricing), and the charge would appear permanently in the merchant's Stripe dashboard. The combination of real cost and permanent audit trail makes systematic fake reviews structurally irrational. It's not that they're impossible. It's that they don't make economic sense.

Is there a fake review checker tool I can use?

Third-party checkers analyze review text, reviewer profiles, and rating distributions to estimate how many reviews might be fake. The best-known browser extension, Fakespot, shut down on July 1, 2025 (Mozilla), so confirm that any tool you rely on is still maintained. These tools are useful for quick checks but have limitations: they primarily work on major platforms (Amazon, Yelp, Trustpilot), they can produce false positives, and they can't access the underlying transaction data that would conclusively verify or disprove a review. For your own business, the better approach is to use a verification system that stops non-customers from posting reviews in the first place.

Is there a real review checker that actually works?

Yes, in two different senses. A real review checker for shoppers scores review text for the writing-pattern signals fakes leave behind (generic praise, superlative density, repeated phrasing, missing concrete detail) and then shows what it found, instead of hiding behind a single opaque score. A real review checker for businesses does the same thing against reviews of their own product pages. Either way, a real one has to disclose its limits: pattern scoring is a screen, not proof, and no text checker can see the transaction behind a review. Our free fake review checker tool scores any pasted review and explains every signal it flags. For platform-wide questions, see our FAQ page.

← Back to blog