How many reviews are enough to be believed

Two thresholds, not a target number

The common question is how many reviews are enough. It is the wrong question, because nobody counts. What people do is glance at a rating and a date, and both of those are thresholds rather than scores.

The first threshold is the rating. The second is recency, and it is the one almost every business underestimates, because reviews feel like an achievement that accumulates. They behave more like fresh produce.

The measured thresholds

Across 1,002 US adult consumers in 2026: 92 percent consider star ratings when selecting a business. 68 percent require a minimum of 4.0 stars and 31 percent will only consider 4.5 or better. So the gap between 3.9 and 4.1 is not a fifth of a star, it is two thirds of the market.

On age: 74 percent prioritise reviews from the last three months, 32 percent expect reviews within two weeks and 18 percent within one week. And the reading has changed shape, since 82 percent now read AI generated summaries of reviews and 23 percent are willing to decide on the summary alone. Consumers consult an average of six review sites, so these thresholds apply six times over.

The thresholds consumers actually apply, 2026

Threshold	Share of consumers
Consider star ratings at all	92 %
Require a minimum of 4.0 stars	68 %
Require 4.5 stars or better	31 %
Prioritise reviews from the last three months	74 %
Expect reviews from within two weeks	32 %
Expect reviews from within one week	18 %

BrightLocal Local Consumer Review Survey 2026, 1,002 US adult consumers surveyed via SurveyMonkey. Rows are separate questions and do not sum to 100.

What this makes worth doing

Three consequences. First, a rating just under 4.0 is worth more attention than any other single number in your marketing, because it excludes you for 68 percent of people. Second, a steady trickle beats a burst: five reviews a quarter keeps you inside the three month window that 74 percent care about, while forty reviews in one week and then silence does not.

Third, since 82 percent read summaries, the wording of your reviews matters as much as the count. A summary is built from what reviewers repeat, so the phrase five customers used will end up describing you, whether or not it is the phrase you would have chosen.

Questions and answers

So how many reviews do I need?

Enough that the rating is stable and the newest is recent. In practice that means a handful every quarter rather than a target total, because 74 percent look at the last three months.

Is 4.9 better than 4.6?

Barely, since both clear the 4.5 threshold that 31 percent apply. The difference between 3.9 and 4.1 is far larger, because 68 percent require at least 4.0.

Do old reviews still help?

They hold the average up, which matters, but they do not satisfy the recency test that three quarters of consumers apply.

Why do the words matter now?

Because 82 percent read AI generated summaries, and a summary is assembled from what reviewers repeat. Repetition, not eloquence, decides what it says.

Sources

BrightLocal, Local Consumer Review Survey 2026 1,002 US adult consumers; star thresholds, recency and summary reading
Google Search Central, Local business structured data On how review and rating data is expressed for machines