The 20-review line

Open your Google Business Profile on your phone. Do it before you read the next paragraph. Read two numbers out loud: your review count, and the date on the newest review. That is the whole test. About forty seconds, and you have already done the hard part.

Now here is the number this piece rests on.

47% of consumers say they will not use a business with fewer than 20 reviews.

That comes from BrightLocal’s Local Consumer Review Survey, published 11 February 2026: 1,002 US adult consumers on a SurveyMonkey panel, spread across ages 18 to 60-plus. It is a research company’s own published survey, not a vendor’s marketing page.

Here is the part that should stop you.

On 21 October 2026 — last week — the same research company published a survey of 1,002 US owners of locally operated businesses with 1 to 50 employees. 47% of them have fewer than 20 reviews.

Two surveys. Two different groups of a thousand people. The same number standing on both sides of the counter.

What you just read off your phone

If that count was under 20, you are in the second 47%.

That is not a verdict on your work. Before this goes any further I want to be clear about what these two numbers do not prove, because there is a very tempting piece of arithmetic sitting right here and it is wrong. I will get to it.

But the mechanism is worth understanding, because it has almost nothing to do with how good you are.

Where the line actually sits

The review count is the first thing on your listing that a person can read without thinking. It sits next to your name in the search result. It is a single integer, and integers get compared instantly — faster than photos, faster than a price, faster than the three words of your business description.

The same consumer survey found only 9% of consumers would use a business with five or fewer reviews. So the curve is not gentle. Somewhere between five and twenty, a business stops reading as new and starts reading as established, and most of that shift happens before twenty.

There is a second line in the same survey, and almost nobody checks it: 74% of consumers look for reviews written in the last three months. Thirty-two percent want something from the last two weeks — up from 20% the year before. Eighteen percent only count the last seven days.

That second number is the date you read off your phone. A business with forty reviews, the newest one from last spring, is below a line it does not know exists.

And the stars matter more than they used to. Sixty-eight percent of consumers in that survey said they would only use a business with four or more stars, up from 55% the previous year. Thirty-one percent set the bar at 4.5 — and that is the one that moved hardest, from 17% a year earlier. It very nearly doubled while nobody was watching.

The confidence gap

The owner survey asked people how sure they were that customers could find them online. Fifty-eight percent said very confident. Another 26% said fairly confident. That is 84% confident.

Hold 84% next to 47%.

If 84% of owners are confident and 47% are below the review line, then at least 31% of owners are both confident and below the line at the same time. That 31% is my own arithmetic, not BrightLocal’s — it is the minimum overlap two percentages that large are forced into, nothing more. It is not a measured group and nobody surveyed it. But the floor is real: roughly one owner in three is sure they are findable while sitting under the number almost half of consumers use to rule a business out.

There is a sharper version of the same gap in that survey, and it is the one I would watch. Fifty-four percent of those owners said they were confident they show up in AI search. I spent last week on what AI answers actually do with local businesses: ask the identical question twice, sixty seconds apart, and only 21–33% of the named businesses come back. Being confident about a surface that re-rolls every time somebody asks it is confidence about nothing in particular.

Confidence is not a signal. It is the absence of a signal. Nobody tells you that you were skipped.

What it costs you, specifically

Here is the shape of it, if you braid hair or cut it or clean houses or fix cars.

Somebody in your city opens their phone on a Thursday night. They are ready. They have the money, they have the date, they have a picture saved of exactly what they want. Three shops come up. Yours is one of them, and your work is the best of the three — genuinely, no flattery.

They read three numbers: 11, 64, 203.

They tap the 203.

You did not lose a customer. You never had one. There was no conversation, no DM, no quote you could have followed up on. The Saturday slot stays open and you have no idea why, because nothing happened — and nothing is very hard to notice.

Nobody turned you down. A number did.

That is the real cost, and it is quiet. An empty chair on a Saturday does not send you feedback.

The honest limit of all this

Now the arithmetic I said I would not do.

Two 47%s are sitting next to each other and it is very tempting to multiply them into something like “half of small businesses are invisible to half their market”. That sentence is false and I am not going to write it as though it were true.

They are two different surveys, 1,002 people each, asked two different questions. One measures what businesses have. The other measures what consumers say they would do. A stated preference is not an observed walk-away. People say they will not use a business under 20 reviews and then book it anyway because it was open, or close, or a friend mentioned it. Survey panels skew toward people who answer surveys. Neither study reports field dates.

And this is the part that matters most: nothing here shows that reviews cause bookings. Businesses with more reviews have been around longer and served more people. That is most of the correlation right there. Getting to 20 reviews does not prove you are good and will not make you good.

What it does is remove a reason to be skipped. That is a smaller claim than the one you will see elsewhere this week, and it is the one I can defend.

One more thing I should say plainly, since I sell websites: a website will not get you reviews. Nothing I build will. Asking gets you reviews. If you take one thing from this, take that, and ignore the rest of what I do.

What you can actually control

Four things, in order of how much they move the number.

Ask at the moment of visible result. Not by text three days later. When the client is looking in the mirror and happy, that is the window. BrightLocal’s press release on the owner survey says 64% of owners actively request reviews, and among those who ask, 57% sit in the 11–50 review band versus 50% of those who don’t ask. That is a seven-point gap — my subtraction, and it is a modest gap, not a transformation. Asking is necessary. It is not magic.

Reply to every one. Eighty-nine percent of consumers expect owners to respond, and 42% say they are unlikely to use a business that never replies. Half are put off by replies that read like a template. Nineteen percent now expect a same-day response, up from 6% the year before.

Keep the newest one recent. Three reviews a month beats thirty reviews from last year, under the recency numbers above. This is the line almost nobody is managing.

Own the surfaces that answer for you. In a separate BrightLocal study published 16 September 2026, which ran roughly 200,000 non-branded local prompts through ChatGPT, Google AI Mode and Google AI Overviews across about 1,300 business locations over 60 days, business websites made up 93% of the unique domains those AI answers cited, and 42% of all citations. Google Business Profile was about 29%. The two surfaces doing the answering are your profile and your website.

Run it yourself

Then tell me what you got. Not to buy anything — I want to know whether that 47% holds up on this street. Reply with your two numbers and your trade. “Braider, 11, August” is a complete answer.

If enough people send theirs, that is better data than either survey, because it is from your city and not a panel.

How I build, briefly

That last point about surfaces is the only place my own work belongs in this. Sedjro Digital builds complete business websites — booking with deposits, an owner’s dashboard, mobile-first — and the client owns the domain, the hosting, the source code and every account outright. No monthly website subscription. Fixed scope, predictable cost, every account in your name.

It will not get you reviews. It gives the asking somewhere to land, and it keeps the surface that AI answers cite most in your name rather than someone else’s.

Method, and what I am not claiming

Four BrightLocal publications carry this piece, each read on its own page rather than in coverage of it. The Local Consumer Review Survey 2026 (11 February 2026) is 1,002 US adult consumers on a SurveyMonkey panel — every consumer threshold above comes from it. The Small Business Marketing Report 2026 (21 October 2026) is 1,002 US owners of locally operated businesses with 1–50 employees, and is the source of the 47% review figure and the confidence numbers. Neither page states field dates, weighting, margin of error or survey vendor.

The asking figures — 64%, and 57% against 50% — are from BrightLocal’s own press release of 22 October 2026 rather than the report page, because they do not appear on the page itself. I have attributed them to the release for that reason. The AI citation shares come from the Local AI Visibility Study (16 September 2026); that page contradicts itself on its prompt and location counts, so I have used rounded figures and left the rounding visible.

What is mine rather than theirs. Two figures. The “at least 31%” is derived as 84 + 47 − 100 — the minimum overlap two percentages that size are forced to share, a bound rather than a measured subgroup, and the true figure could sit anywhere between 31% and 47%. The seven-point asking gap is 57 − 50, plain subtraction.

What I deliberately did not compute. I did not multiply the two 47% figures, or combine them in any way. They come from separate surveys of separate populations answering different questions, and any joint figure would be invented. Nor does anything here claim that reviews cause bookings, that any business was shown to lose work, or that what consumers said they would do is what they did.

Common questions