Review gating: what changed, and what to do about it

If your review software asks customers how it went before deciding who gets the Google link, that is review gating — and it is the thing to turn off. Here is what the rules actually say, where the real risk is, and how to check your own setup.

Last checked 12 September 2026. Every date and figure below is the one on the source as of that day.

What review gating is

It has a different name in every product: sentiment routing, rating gate, feedback funnel, smart routing, private feedback filter, NPS pre-screen. The mechanics are always the same.

After a job, the customer gets a message asking how it went.
Answer positively and you get the link to the public review site.
Answer negatively and you get an internal feedback form instead.

The public rating that results is not what customers think it is. It is the part of the opinion the business chose to publish.

What Google’s policy says

Gating has been against Google’s policy for years, not since 2026. Google’s content policy prohibits “discouraging or prohibiting negative reviews, or selectively soliciting positive reviews from customers”.

What changed in April 2026 is enforcement. Google made two changes, on 16 and 17 April 2026.

Moderation powered by Gemini that screens reviews before they are published, rather than reacting to reports afterwards.
Newly explicit under the Rating Manipulation policy: staff review quotas, asking a customer to name a specific employee, and on-premises review pressure — asking at the counter, review kiosks, tablets in the waiting area.

That last one catches businesses who thought they were doing everything right. A tablet by the till is now a policy problem.

Google’s consequences are profile-level and customers can see them: losing the ability to receive new reviews, existing reviews being unpublished, and a public banner on the profile saying fake reviews were removed. No fine looks as bad to a customer as that banner.

What the FTC rule says — and the part most articles get wrong

The FTC’s Rule on the Use of Consumer Reviews and Testimonials (16 CFR Part 465) took effect on 21 October 2024. It bans fake and AI-generated reviews, undisclosed insider reviews, buying positive or negative reviews, incentives conditioned on a review expressing a particular sentiment, fake social-media indicators, and review suppression — both threatening someone into taking a review down, and publishing only a flattering subset of the reviews you hold.

It does not, on its face, ban review gating. The rule carries an exception at §465.2(d) for reviews resulting from generalized solicitations, and the legal summaries are consistent that it contains no specific prohibition on selective solicitation.

That is not permission. The exposure is real — it runs through three doors rather than one.

The FTC Act’s general prohibition on deception, and the Endorsement Guides, which carry a published example that is almost exactly the feedback funnel: a marketer asks purchasers for feedback, invites the very positive ones to post reviews on third-party sites, and merely thanks the less pleased ones.
Rule §465.6 on suppression — the moment gating becomes publishing a filtered set, such as testimonials on your own site or a star average you control, it is inside the rule.
Misrepresenting that your ratings reflect all customers, which is the theory the FTC actually used in the case below.

Civil penalties run to $53,088 per violation — the FTC Act maximum, set on 17 January 2025 and unchanged for 2026. It is adjusted for inflation most years, so check the current figure rather than trusting this sentence in a year’s time.

The precedent worth knowing if you are choosing software

Fashion Nova, January 2022, $4.2 million. The FTC alleged the retailer’s website showed only some of the reviews customers had submitted.

The mechanism is the part that matters here. It used a third-party review-management interface that automatically posted four- and five-star reviews and held everything lower for approval — and from late 2015 to November 2019, the lower-rated reviews were never approved. It was the FTC’s first case about concealing negative reviews.

The tool did it. The business paid.

Two things this case is not: it predates Part 465 and was brought under the FTC Act, and it was about suppressing reviews on the company’s own website rather than gating who gets asked for a Google review.

Enforcement so far

On 22 December 2025 the FTC sent warning letters to 10 companies, whose names were not made public, over possible violations of the Consumer Reviews Rule. They are the first public enforcement step under it, and each letter told the recipient to stop and to confirm in writing what it had done about it.

Reporting on the letters describes them as concerning fake and AI-generated reviews rather than gating. They are still the clearest signal so far that the rule is being enforced rather than explained.

What to do about it, whatever your tool is called

Do not go looking for a setting called “review gating”. No product calls it that. Look for the behaviour instead.

Does anything ask how the customer feels before they are given the review link?
Does any answer send some customers somewhere other than the public review site?
Is a private feedback form offered instead of the public ask, rather than alongside it?
Is anything offered in exchange for a review — a discount, a free coffee, entry into a draw?
Does anyone ask for reviews on the premises, set staff targets, or ask customers to name a particular employee?

Then do the thing that actually settles it: ask your vendor, in writing, whether any customer is routed anywhere other than the public review link, and keep the answer. A support reply in your inbox is worth more than a setting you think you understand, and it is the only check that survives the product being renamed or redesigned.

Several review tools can be configured compliantly. The question is not which vendor you use, but whether the funnel exists in your setup.

Check your setup in five questions

What compliant asking looks like

Everyone gets asked, every time — no screening step.
The same message to every customer, whatever you expect them to say.
No incentive, for a review or for a particular rating.
Private feedback offered in addition to the public ask, never instead of it.
Ask afterwards, by email or text — not at the counter, not on a tablet in the waiting room.

Where we come into it

GossipGarden was built this way from the start. There is no sentiment step, no “how did we do?” pre-screen, and no way to route an unhappy customer somewhere other than the review site — not as a setting that is switched off, but as something the product cannot do. We have never had a gate to turn off.

See how it works

Sources

Every claim above links to the source that carries it. Where a figure changes over time — the penalty maximum especially — the date it was checked is printed beside it.