How many Google reviews do you need to show up in AI results?

So why does review count come up at all?

Because it correlates with something real. When an AI assistant answers "who's a good plumber near me," it isn't counting stars. It's summarising what looks current, consistent and verifiable about businesses it has information on. Review volume, review recency and whether the owner replies are all signals of exactly that — not because there's a formula with a cutoff, but because a business that looks actively run is easier to describe confidently than one that looks dormant.

That's why, in practice, a business with a dozen recent reviews and a reply on most of them can come up ahead of a business sitting on forty reviews from two years ago with no replies at all. The second business has a bigger number, but the first one looks alive right now — and "right now" is closer to what these tools are actually trying to answer.

What seems to matter, based on how these tools work

• Volume — enough reviews to look like a real, ongoing business, not a one-off listing. There's no magic line; more than a handful is clearly better than almost none.
• Recency — reviews from the last few months read as "still trading and still good," which matters more than a pile of old ones.
• Owner replies — a listing where the owner responds reads as actively managed, which is one of the few signals a business fully controls.
• Consistency elsewhere — the same name, address, phone number, hours and service area across every directory the business appears on, so there's one clear, repeatable record to summarise rather than several conflicting ones.

None of this is a guarantee. It's a description of what's plausible given how these assistants summarise public information — not a documented ranking system, because no such public document exists.

Background reading, not our claims

Two pieces informed this page. We're linking them so you can read the underlying thinking yourself rather than take our word for it:

• A report on a live product that sells AI-visibility fixes to local service businesses — background on why this problem is being taken seriously enough that other companies sell a fix for it. Read the report
• An explainer on how AI assistants recommend local businesses based on reviews and listing consistency — background on the mechanics described above. Read the explainer

We haven't repeated any numbers from either source here, because we haven't independently verified them. What we have verified is our own method — see the exact prompts and how we run them.

The practical move, with or without an audit

You don't need our report to start on this. Two things move the signals above, starting today:

• Ask every customer, right after a finished job, for a review. A short text or a card left on-site works better than hoping people remember later.
• Reply to all of your reviews — the good ones and the awkward ones — in your own words, describing the job and the area you served. It costs nothing and it's one of the few things entirely in your control.

What we can't tell you from a blog post is whether any of that is actually enough to get your business named for your trade in your area, today. That's a question about your specific listings and your specific competitors, not a general rule — which is exactly what the snapshot checks.

Want it checked properly, not guessed at?

49 EUR, one-off. We run a fixed set of questions across ChatGPT, Gemini and Google AI Overviews, in at least three separate sessions at different times, and send a timestamped PDF naming who came up and the three to five fixes most likely to help. See what's inside the report.