Review gating: why filtering for happy customers is against the rules

You have almost certainly received one. A text after a job asking how it went, one to five stars. Tap four or five and you get a link to Google. Tap one, two or three and you get a form that goes to the owner's inbox.

It feels reasonable. It is sold as reasonable. It is called review gating, and it is not allowed.

What the rules actually say

Google's policy on user contributed content prohibits soliciting reviews selectively, and prohibits discouraging or suppressing negative reviews. The mechanism above does both at once, which is why it is not a grey area. It is the named example.

In the United States, the Federal Trade Commission's rule on consumer reviews and testimonials covers the same conduct from the consumer-protection side, and it carries civil penalties. Consumer law in Australia and the EU addresses it too, on the grounds that a rating built from a filtered sample misleads the people relying on it.

Why it is a bad idea even if nobody catches you

Set the rules aside for a moment. Gating does not do what the person buying it thinks it does.

A perfect rating is not persuasive. Shoppers who read reviews are looking for the shape of the risk, and a page of five-star ratings with no texture reads as either a very new business or a managed one. The reviews people spend the most time on are the middling ones, because that is where the specifics live.

There is a second cost that is harder to see. A gated complaint goes into an inbox and dies there. An ungated complaint is public, which means it gets answered, and a well-answered complaint is the single most convincing thing on most listings. You are trading your best sales asset for a decimal point.

The customer who complained in public and got a good answer sells more work than the customer who left five stars and one word.

What you are allowed to do

Quite a lot, as it turns out. The rule is not that you cannot ask. The rule is that you cannot ask selectively.

• Ask every customer, in the same way, at the same point in the job.
• Ask in person, by text, by email, on the invoice, on a card in the van. Channel is not the issue.
• Make it easy: a short link straight to the review box, not a homepage and a treasure hunt.
• Remind once. One follow-up is a reminder. Three is pressure.
• Ask a dissatisfied customer for a review too. That sounds mad and it is the whole point.

That last one deserves a sentence. If you have fixed the problem, the customer who had one and watched you sort it out is often the most willing reviewer you have, and the review they write does more work than a dozen ordinary ones.

The survey is not the problem, the branch is

Plenty of businesses want to know how a job went before it becomes public, and there is a legitimate version of that instinct.

Run your internal satisfaction survey if you find it useful. Read the answers. Ring the unhappy ones. Then send the review request to everybody, regardless of what they said. The survey is a management tool. It is not a gate, and the moment it decides who sees the Google link, it becomes one.

If you are already doing this

1. 01Turn off the branching today. It is usually a single toggle in whatever tool sends the message.
2. 02Send the same request to everyone from now on, with no scoring step in front of it.
3. 03Expect your average to fall slightly. That is the sample correcting, not a decline in your work.
4. 04Answer every review that is not five stars, in public, within a couple of days.
5. 05Keep asking. Volume and recency are what a rating is made of, and both are things you control.

If you are working out how many reviews you actually need to compete, we wrote about that here. Reviews are one part of what Google calls prominence, and the off-page side of local SEO covers the rest of it, alongside the full local SEO checklist.