How many 5-star reviews do you actually need?

Exact maths, not an estimate. The same sum Google runs on your profile.

Review rating calculator

Platform
Leave at 5 unless you expect a mix.

You need

10

more 5-star reviews

To go from 4.3 to 4.5 on 23 reviews, you need 10 more at 5 stars. That puts your true average at 4.51.

Google rounds to one decimal, so you display as 4.3 today. Just 2 reviews tips that to 4.4, which is the cheapest visible win you have.

If one 1-star arrived tomorrow

Your average would go from 4.30 to 4.16, dropping the rating people see from 4.3 to 4.2. Undoing it would take 5 five-star reviews.

What one review does to the rating people see

The average moves in hundredths. The number on your listing moves in steps. This is the same business before and after crossing the next step.

Today

4.3

True average 4.30

2 more 5-star reviews

After that

4.4

True average 4.36

Google rounds to one decimal, so any true average from 4.35 to 4.44 shows as 4.4. Landing anywhere in that band looks identical to a customer.

Does this work for your platform?

Only where the rating is a plain average. Two of the five below qualify, and pretending otherwise would give you a confident number that is simply wrong.

The maths is exact

Google

Google

A plain average of every published review, shown to one decimal. Every review counts the same however old it is, so the sum above is exactly the sum Google is running.

Yelp

Yelp

Also a plain average, but Yelp rounds to the nearest half star, so anything from 3.75 to 4.24 shows as 4.0. Same sum, bigger visible steps. Pick Yelp above and the threshold line adjusts.

Different maths, so no exact answer

Trustpilot

Trustpilot

Not a plain average. Trustpilot's own documentation says newer reviews count for more, and the formula is not published. A score here can fall through inactivity alone.

Tripadvisor

Tripadvisor

A summary score drawn from the written reviews as well as the scores, shown in half bubbles. Not an average you can work backwards from.

Facebook

Facebook

Recommendations, a yes or a no, rather than a star average. There is no mean here for a calculator to move.

Why ten reviews are worth more than your next hundred

One 1-star review. Two businesses, both sitting on 4.8. The only difference is how many reviews they had already collected.

6 reviews

4.84.3

The public rating falls five whole steps. Anyone comparing you against the shop next door sees a different business today than they did yesterday.

112 reviews

4.84.8

The average moves by 0.03 and the rating people see does not change at all. The review is still there, and it costs nothing.

Worth being precise about this, because it is where most advice gets it wrong: getting the true average back takes 19 five-star reviews in both cases. What a long review history buys you is not a cheaper recovery. It is not looking damaged while you recover.

The sum itself

No estimate, no industry benchmark, no weighting anyone invented. Your rating is a plain mean: add up every star and divide by the number of reviews. So the number of new reviews you need is fixed by three things, and you can check it on paper.

reviews needed = how many you have, times the gap you are closing, divided by how far a 5-star sits above your target

That last division is why the answer grows so fast as you approach 5. Moving from 4.3 to 4.5 means each new 5-star review buys you half a star of headroom. Moving from 4.3 to 4.9 leaves a tenth, so the same climb costs six times as many reviews.

Why your history is the weight

Two businesses both sit on 4.3 and both want 4.5. The one with 23 reviews needs 10 more. The one with 112 needs 45. Nothing about the second business is worse. It simply has more history holding the average where it is, and every review it ever collected is now pulling against the move.

This cuts both ways, and it is the most useful thing on the page. A long review history is slow to improve and almost impossible to damage. A short one moves fast in whichever direction it is pushed.

The rating people see

Google displays one decimal place and rounds to it. An average of 4.44 shows as 4.4. An average of 4.45 shows as 4.5. Everything in between is invisible.

That matters more than it sounds, because the number your customers judge you on moves in steps. Sit on 40 reviews at 4.44 and a single 5-star review takes your public rating from 4.4 to 4.5. Most calculators aim at the raw average and miss this entirely. The one above gives you both, because the cheap win is usually the visible one.

What you can ask for

Worth being clear, because the rules changed. Since 6 April 2025 the DMCC Act has made it a banned practice in the UK to write or commission fake reviews, or to offer an incentive for a positive review without disclosing it. The CMA can now fine businesses directly, up to £300,000 or 10% of global turnover, whichever is higher.

Still completely fine: asking every customer, every time, for an honest review. Putting the link on the invoice, the receipt, the thank-you email and the van. Replying to the ones you get.

Not fine: asking only the customers you think are happy, offering a discount for five stars, or writing them yourself. The line is not about asking. It is about filtering and paying. This is general information rather than legal advice, and the CMA publishes its own guidance if you want the detail.

Getting the number

Nearly every business that asks us about reviews is not short of happy customers. It is short of a habit. The ones who fix it do the same three things: they ask at the moment the job is finished rather than a week later, they make the link one tap rather than a search, and they ask every customer instead of the ones they feel confident about.

If the number above looks impossible, the target is usually wrong rather than the business. Going from 3.8 to 4.5 on six reviews takes nine. Going from 4.3 to 4.9 on 112 takes over a thousand. Pick the target you can reach this year.

Common questions

It depends entirely on how many reviews you already have. The number you need is your current review count multiplied by the gap you want to close, divided by how far a 5-star review sits above your target. A business on 23 reviews at 4.3 needs 10 more five-star reviews to reach 4.5. A business on 112 reviews at 4.3 needs 45 to make the same move, because every existing review is holding the average in place.

No. Once you have a single review below 5 stars, a true 5.0 average is mathematically unreachable, no matter how many 5-star reviews you add. You can get arbitrarily close, and 4.9 displays on Google as 4.9, but the average itself can never return to 5.0. Any calculator that tells you otherwise is wrong.

Google displays one decimal place and rounds to it. An average of 4.44 displays as 4.4, and an average of 4.45 displays as 4.5. That means the displayed rating moves in steps, and you are sometimes a single review away from a visible jump that looks far bigger than the maths behind it. The calculator shows you that threshold as well as the raw average.

The damage is to the rating people see, and it depends almost entirely on how many reviews you hold. On 6 reviews at 4.8, one 1-star takes the public rating from 4.8 down to 4.3, a fall of five whole steps. On 112 reviews at 4.8 the same review moves the average by 0.03 and the public rating does not change at all. Recovering the true average takes a similar number of 5-star reviews in both cases, so the real difference is how exposed you look while you do it.

In the UK you can ask any customer for an honest review, and you can ask all of them. What you cannot do, since 6 April 2025 under the DMCC Act, is write or commission fake reviews, or offer an incentive for a positive review without disclosing that incentive. Asking everyone and accepting whatever they write is safe. Asking only the happy ones, or paying for stars, is not. This is general information rather than legal advice.

You Have The Number.
Now The Harder Part.

Getting reviews is a habit, not a campaign, and it works best wired into the website and the follow-up rather than bolted on afterwards.