How many reviews does a business need? Your three targets
How many reviews does a business need? Five to convince a buyer, the median beside you to be found, and enough to steady your rating, with the arithmetic.
Written by OpenRated Team You have eleven reviews and a 4.6. The company you keep losing deals to has ninety and a 4.4. So how many reviews does a business need before a buyer stops counting and starts reading, and does the number you are chasing even matter, or is it the rating?
The answer is three numbers, because a review count does three different jobs. It has to convince the person who has already found you, it has to help the next person find you, and it has to be big enough that one bad week does not move your average. Each job has its own target, and only the first has a study behind it.
How many reviews does a business need: three jobs, three targets
Most pages that answer this question pick one number and apply it to everything. The number depends on what you need the reviews to do, so set the targets separately.
| What the count has to do | What the evidence says | The target |
|---|---|---|
| Convince a buyer who is already looking at you | Purchase likelihood is 270% higher with five reviews than with none, and the gain per extra review drops fast after five (Spiegel Research Center, 2017) | Five on the listing the buyer sees |
| Help the next buyer find you | Google says more reviews and positive ratings can help local ranking, and its help page gives no threshold | As many as the businesses shown beside you |
| Keep the rating steady | One one-star review on 20 reviews at 4.6 shows as 4.4; on 200 it still shows 4.6 (arithmetic below) | Enough that one review moves the average by less than 0.05: at 4.6, 72 or more |
Five reviews: the number that convinces a buyer
The one count in this post that traces to a study is five. The Spiegel Research Center at Northwestern University published How Online Reviews Influence Sales in 2017, and its finding is that purchase likelihood for a product with five reviews is 270% greater than for a product with none, with the marginal benefit of each additional review diminishing rapidly after those first five. The same study found the lift was larger for higher-priced items (a 380% increase in conversion rate) than for lower-priced ones (190%).
Take the curve from it and leave the decimals. The study measured product pages, so a software company selling a $400-a-month plan is closer to its conditions than a bakery or a plumber is. What carries across is the shape: the first handful of reviews does most of the persuading, and review number sixty does far less than review number four. If your business has two reviews, getting to five is the highest-value thing you can do this month. If you have forty, the count is no longer your problem.
Two limits. Five reviews on a listing nobody opens convince nobody, so the five have to sit where your buyers actually look: a software company's buyers compare products on a review site's category page, an agency's look at its Google profile and a directory or two, a shop's at Google and Maps. And five reviews from the same week, all five stars, all one sentence long, read as arranged. Five spread over a few months, with one of them a four, read as real.
How many Google reviews does a business need
Most answers to this question are about Google, and most name a threshold: ten, twenty, thirty. Google's own local ranking help page names none. It lists three factors (relevance, distance and prominence), puts review count and ratings under prominence with the sentence "more reviews and positive ratings can help your business's local ranking", and adds that there is no way to request or pay for a better local ranking. That is Google's own statement on the subject; any specific number you read elsewhere is somebody's estimate.
The practical target is relative, and you can measure it in five minutes. Search the phrase your customers use, with your town or your category, and write down the review count and rating of each business or product shown in the first results. That median is your number. An agency with twelve reviews in a city where the top three show forty, ninety and two hundred has a volume problem; the same agency in a town where the top three show six, nine and fifteen does not, and should spend its effort on the rating and on answering the reviews it has.
The limit is the one Google states itself: reviews are one input to prominence, and prominence is one of three factors. A business across town from the searcher will not outrank one next to them on review count alone.
What is a good review rating
Higher is not simply better, and the same Spiegel study is the source for that too. Purchase likelihood peaked at ratings between 4.0 and 4.7 and then began to decrease as ratings approached 5.0. A buyer reads a flawless 5.0 from thirty reviews as either a very new business or a curated one, and reads a 4.5 with a couple of three-star reviews, answered by the owner, as a company that has been around and dealt with things.
So a good review rating for a business is anything from 4.0 up, and your first four-star review is a normal event. The rating that hurts on its own is one below 4.0, the bottom of the range the study found buyers reward.
A note on averages: a rating is an average, so it is only as stable as the count underneath it. That is the third job, and it has its own arithmetic.
How many bad reviews affect your business: the arithmetic
One one-star review costs you more the fewer reviews you have, which is why "how many reviews should a business have" and "how much does a bad review hurt" are the same question. The count is a cushion, and you can work out how thick yours is.
One review moves an average by roughly (your rating minus the new review's rating) divided by (your count plus one). At 4.6, that is 3.6 divided by your count plus one, so one one-star review moves the average by less than 0.05 once you have 72 reviews. Below 0.05, the one decimal most sites display does not change, provided the average sat at an exact 4.60 before the review landed; from a 4.55 displayed as 4.6, a smaller move still tips it to 4.5. So the third target is 72 at a 4.6; at a 4.4 it is 68, at a 4.8 it is 76. Under it, every bad review is visible on the page; over it, a bad review is a reply to write.
Here is the cushion at three sizes. Take a business at 4.6, the kind of rating a good company settles at, and give it one one-star review:
- 20 reviews at 4.6 hold 92 stars. Add a one-star review: 93 stars over 21 reviews is 4.43, which displays as 4.4.
- 60 reviews at 4.6 hold 276 stars. Add one: 277 over 61 is 4.54, which displays as 4.5.
- 200 reviews at 4.6 hold 920 stars. Add one: 921 over 201 is 4.58, which still displays as 4.6.
The recovery is a different sum, and it has a counter-intuitive property: the number of five-star reviews it takes to get back to the old average depends only on the rating you are aiming at, whatever your count. It is (target rating minus 1) divided by (5 minus target rating). To return to 4.6 after one one-star review: 3.6 divided by 0.4, which is nine five-star reviews, whether your listing holds 20 reviews or 200. The higher the target, the steeper that gets, which is one more reason to stop chasing a score past the 4.7 where buyers stop rewarding it.
Four steps to set your review target
The routine, in the order that finds your real gap fastest.
- Get to five on the listing your buyers open. Five on the one page where the decision is made. A software company picks the category page its buyers compare on; a shop picks Google. Below five, nothing else on this list matters yet.
- Count the businesses shown beside you. Run the search your customer runs, note the count and rating of each result on the first page, and take the median. That is your volume target, and it is different in every market.
- Keep the rating between 4.0 and 4.7. Treat a four-star review as normal and a five-star streak with no texture as a warning sign to a buyer. Reply to the critical ones.
- Set a monthly pace you can keep. Divide the gap to your volume target by the months you are willing to spend, and check that against your customer volume. A shop that serves two hundred people a week can ask for ten reviews a month without strain; an agency that closes four clients a quarter cannot, and should set the pace at one or two a month and spend the rest of its effort on the rating.
What the routine will not do is rescue a product people dislike: asking more customers for reviews when a third of them are unhappy raises the count and lowers the rating at the same time. Fix the thing they complain about first, then ask.
Why buying the count is off the table
The shortcut everyone considers is paying for the count. In the United States that is now a rule with civil penalties attached. The Federal Trade Commission's final rule on fake reviews and testimonials, announced on 14 August 2024, bans fake or AI-generated reviews, buying positive or negative reviews, undisclosed insider reviews and suppressing reviews by threat, and lets the agency seek civil penalties against knowing violators. Asking a customer for an honest review is absent from that list; paying for a five-star one is on it. This is a summary of the rule as published and is not legal advice.
That leaves the slower route, and the targets above make it finite: five on the right listing, the median count beside you, and, at a 4.6, a cushion of 72 reviews before one bad review stops moving the displayed number.
The one thing to do this week
Open the listing where your buyers decide, count the reviews on it, and compare the number with the businesses shown beside you. Under five, that is the whole plan for this month. Over five and under the median beside you, set the monthly pace from step four. Over both, stop counting and answer the reviews you have.
When the target is set, the tools that send the asks are compared row by row on the reputation and review management listing.
