Your competitor had forty reviews last month. This month they have two hundred and forty, all five stars, all posted within a fortnight. Your first thought — they must be generating them — may be right, and your second thought — should I? — is the one this article exists to answer. AI has made fake customer reviews cheap and fluent, and small businesses are making expensive mistakes with them. Here are the mistakes, in roughly the order businesses discover them, and the correct way to think about each.
Mistake 1: Believing “everyone does it” makes it safe
The legal position is not ambiguous, and it does not care what everyone does. In the United States, the Federal Trade Commission’s rule on consumer reviews and testimonials — the text lives in the Code of Federal Regulations as the FTC’s consumer reviews rule (16 CFR Part 465) — prohibits creating, selling, or buying fake reviews, including AI-generated ones presented as real customers, and it reaches the business that benefits, not just the vendor who typed them. Penalties run per violation, and enforcement actions against businesses using purchased praise predate AI entirely; the technology only made the violation easier to commit and easier to spot in bulk. “A marketing company offered” is not a defense. It is a description of how the liability arrived.
Mistake 2: Thinking generated reviews are hard to detect
Owners who buy generated reviews imagine them blending in. Read any seeded review section with fresh eyes and the blending fails: the same rhythm of praise, the same weightless adjectives, bursts of reviews arriving in clumps, five stars unaccompanied by a single specific incident. Shoppers have learned to read these signals — praise with no incidents, features the product does not have, bursts of identical enthusiasm — and platforms employ the same signals at machine scale, with the power to suppress the reviews, flag the listing, or suspend the business profile entirely. The scenario that should scare a small business is not the fine. It is waking up to find the whole review profile — including the genuine praise earned over years — frozen or gone because the platform’s systems classified the account as manipulated. That asset took years to build and one cheap shortcut to burn.
Mistake 3: Generating reviews “from” real customers without asking them
A subtler trap has appeared with the tools themselves: software that drafts a review for a customer to post, or worse, that turns a private feedback form, a text message, or a star rating into a polished public paragraph under the customer’s name. Even with a genuinely happy customer underneath, posting words they did not write, in a place they did not choose, crosses the same lines — it misrepresents the review’s origin to readers and to the platform, and it can convert a delighted customer into an alarmed one when they meet their own “review” online. The legitimate version of this idea exists and is simple: make it easy for customers to write their own review, in their own words, and let the words be theirs.
Mistake 4: Suppressing the bad ones instead of answering them
The fake-review rule cuts in both directions: it also prohibits suppressing negative reviews and cherry-picking which customers get invited to review. “Review gating” — sending happy customers to Google and unhappy ones to a private form — is a named violation, and platforms treat a suspiciously perfect profile as evidence rather than achievement. The businesses that handle criticism well do something less glamorous and more effective: they answer negative reviews promptly, specifically, and without defensiveness, in public, where future customers read the answer as evidence of how the business behaves when something goes wrong. A thoughtful reply to a two-star review sells harder than twenty generated five-stars, because shoppers believe it.
Mistake 5: Treating review count as the goal
Underneath all of these is a strategic error: review volume is a means, and a lagging one. What actually converts a searcher into a customer is a believable profile — a healthy count, a rating in the honest fours, recent reviews that mention the specific things the business is known for, and visible proof that real transactions keep happening. A business chasing volume buys the number and corrodes the believability, which is the asset. A business chasing believability does the boring work instead.
Mistake 6: Outsourcing to a “reputation service” and not asking questions
The modern version of buying reviews rarely arrives labeled as buying reviews. It arrives as reputation management, review acceleration, or a five-star growth package, sold by an agency that promises everything is “compliant” and “white hat.” Sometimes it is — legitimate services exist that do nothing more than send review invitations and monitor listings. Sometimes the package quietly includes the agency’s own network of accounts, generated text, and rebate schemes, and the business owner has purchased exactly the liability described above with a layer of invoices in between. The FTC rule does not grade on whether you personally wrote the fake text, and platforms do not restore profiles because a vendor did it. The due diligence is five questions, asked before signing: Will every review be written by a real customer, in their own words? Will customers be offered anything of value for reviewing? Will all customers be invited, or only selected ones? Can I see, in writing, exactly what your service posts on my behalf? And what happens to my profile if a platform flags it — who answers? A service that squirms at written answers is telling you what it sells.
If you already bought them: damage control
For the business reading this with a sinking feeling, the path back is unglamorous but real. Stop the service today — every new fake review restarts the clock and deepens the pattern. Remove what you can: reviews you control on your own website should come down, and where a platform offers a way to report or disown suspicious reviews on your own listing, use it and keep records of having done so. Do not compound the problem by suddenly soliciting a flood of real reviews to bury the fake ones; a fresh burst reads as another campaign. Instead, return to the steady program below, accept that the profile may look thinner or stranger for a while, and let time and genuine volume do what they do. If a platform does take action, appeal honestly and briefly — businesses that disclose and correct fare better than businesses that argue the fake reviews were real. The asset is recoverable. Businesses recover from this routinely; what they do not recover from is being caught twice.
The correct way: the boring program that works
Earn reviews the way you earn repeat customers, systematically. Ask at the moment of satisfaction — at the till, in the follow-up message, on the receipt — with a direct link that removes every step between the customer and the review box. Ask everyone, not just the delighted, both because gating is prohibited and because the occasional three-star review with a gracious reply makes the five-stars believable. Respond to all of them. Put real reviews, attributed with permission, in your own marketing rather than inventing testimonials for the website. And read your reviews the way your competitors’ customers will: as a report on your business, where a repeated complaint is worth more than a compliment. The practical playbook is covered step by step in how to get more Google reviews for your business — none of it is fast, and all of it compounds.
How long does the honest way take?
Longer than buying them, and the timeline depends on transaction volume rather than effort. A café serving hundreds of customers a week, asking consistently, can build a substantial, believable profile within a season. A plumber completing five jobs a week builds the same profile across a year. Both end up with the same asset, because the rate of accumulation matches the rate of real work — which is precisely what makes the profile credible to readers and to platforms alike. The businesses that get in trouble are almost never the ones with too few reviews. They are the ones whose review count stopped matching their customer count, in public, where anyone can do the arithmetic.
The competitor’s two hundred reviews, then, are best read not as a threat but as a debt they have taken on
— against the platform’s detection systems, against the regulator’s rulebook, and against the day a customer posts the screenshot. Your forty real ones are an asset with no due date. Small businesses do not lose the review game by having fewer reviews. They lose it by trading the believable kind for the generated kind, at exactly the moment when shoppers and platforms have both gotten very good at telling the difference. Protect the profile you have the way you would protect the sign over the door — it took the same years to earn, and it works on the same customers.
