How to Spot Fake Google Reviews: 7 Signals With Real Examples

Seven things give away a manipulated Google listing: a spike of 5-star reviews in a few days, reviewer profiles with nothing else on them, sentences repeated under different names, glowing ratings attached to text that describes nothing, an average that climbs faster than the traffic explains, reviewers with no local history, and reviews landing at the same hour again and again. One of those proves nothing. Two or three stacked on the same listing should change how you read its rating before you book a table, hire a contractor, or pick a dentist.

We do this at scale rather than by eye. By July 2026 we had put 241,041 reviews from 242 places through the scorer and flagged 49,047 of them, close to one in five, for at least one suspicious marker. The seven patterns below are ordered by how often we see them.

#What does a fake-review burst look like?

A spike of new reviews, nearly always 5-star, arriving days apart instead of months apart and breaking from a listing's usual pace. Organic reviews roughly track foot traffic; a wall of praise on a Tuesday does not.

Bursts sit second in our data, behind thin accounts, across 12,259 flagged reviews. They tend to follow a purchased review package, a review-gating promotion that went wrong, or a competitor inflating a rival's numbers to make the listing look bought. Checking by hand takes a minute: open the reviews tab, sort by "Newest," read the dates. Ten or more 5-star reviews inside one week with quiet on either side is the shape you are looking for.

#Why do low-activity reviewer accounts matter?

These are Google profiles with a review or two on them and nothing more, usually 5-star, usually for businesses with nothing in common, usually written in the same fortnight. It is the marker we see most: 37,061 flagged reviews carry it, more than bursts and repeated text together.

Plenty of real people review one restaurant and never write again, so a single thin profile means little. The ratio is what counts. When a large share of a listing's praise comes from accounts with no history at all, the review history looks bought or traded rather than earned. Tap a reviewer's name. If the profile holds this review and nothing else, or a handful posted the same week across unrelated businesses, weigh it accordingly.

#How can I tell if review text is duplicated or templated?

Copy a distinctive phrase and search it. If the same sentence appears under another name on the listing, or on a business across town, it was written to be reused rather than to describe an evening out.

Our data carries 5,958 flagged reviews with this marker. Reviews-for-hire operations hand writers a short menu of phrases to shuffle, which is why the wording repeats even when the photos and the star counts vary.

#Does a 5-star rating with vague text mean the review is fake?

Not by itself. Happy customers write short, generic praise all the time. But a 5-star review naming no dish, no staff member, and no moment is thinner evidence of an actual visit than one that names something, and vagueness is the default for writers who were never there.

The mismatch works in reverse too. A 1-star rating attached to a complaint that could apply to any business, with no incident described, points more toward a coordinated attack than an angry customer.

#What does a sudden rating jump indicate?

An average moving from 3.8 to 4.4 over a stretch too short for that much new traffic is worth checking against the posting dates. Pair it with the burst check: a jump with no corresponding cluster of reviews to account for it is odd.

#Real examples from our analyzed reports

Essex, a New York steakhouse with 8,934 reviews, scores 83/100, its history thick with thin accounts and spike days. Bury the Hatchet Axe Throwing in Brooklyn, past 8,000 reviews, scores 76/100 on the same markers. Each report page shows the day-by-day evidence behind the number. What neither number establishes is that either business bought anything, which review text alone cannot show. The catalog-wide view lives in our data on how common this is.

#Signal checklist

Before you book, five checks:

• Sort reviews by "Newest" and look for a burst
• Tap 3-5 recent 5-star reviewers and check their profile history
• Search a distinctive phrase from a review to check for duplicates
• Compare the current rating to what it was a few months ago, if visible
• Read the 3-star reviews — they're the least likely to be manipulated in either direction

Two or more signals stacking is worth acting on rather than shrugging at, and the reporting guide covers what to file and what happens next.

#Frequently asked questions

#Can Google detect fake reviews on its own?

Google runs automated systems and human moderators, and removes a great many. Campaigns built on thin accounts and varied wording still survive for a while before being caught, which is why checking the signals yourself remains worth the minute it takes.

#Is buying Google reviews illegal?

Yes. Paying for reviews, selling them, or having a model write them has been federally prohibited since autumn 2024, at up to $53,088 apiece. Our rundown of the rule covers what it bans and the fines issued so far.

#What's the difference between a fake review and an unfair review?

A fake review comes from someone who was never a customer, usually paid or traded. An unfair review comes from a real customer whose complaint the owner disputes. Only the first breaks a policy that Google or the FTC will enforce.

#Does a low overall rating always mean the reviews are trustworthy?

No. Attacks use the same spikes and empty accounts as inflation campaigns, aimed at 1 star instead of 5. Run the same checks whichever way the rating moved. Owners on the receiving end have a playbook of their own.

#How many reviews does a business need before I should trust the rating?

No fixed number, but a rating resting on 20 or 30 reviews bends easily under one burst. A rating built from hundreds of reviews over several years is much harder to move.

Comparing tools for other platforms, or wondering where Fakespot went? That comparison is here.

Rather than scanning by hand, hand the link to our analyzer and get all seven signals checked against the full review history at once.