How Much Do Online Reviews Actually Affect Your Website Traffic?

November 16, 2026 · 8 min read · The Crafty Meerkat Team

This is one of the most reasonable questions a business owner can ask, and it rarely gets a straight answer. You have spent an afternoon chasing a customer for a review, or an evening stewing over a three-star one, and you would quite like to know whether any of it touches the thing you paid someone to build.

The short version: reviews almost never send someone to your website by clicking a link. Hardly anybody reads a review and then taps through to your site from the review page itself. What reviews change is something slower and harder to see — whether a person who has heard your name decides to look you up at all, and what they do in the ninety seconds after they land. That effect is real, it is measurable in a rough way, and it shows up in two particular corners of your analytics that most people ignore.

Reviews do not send clicks. They change what people type.

Imagine someone needs a plumber on a Tuesday evening. They search for plumbers in their town, see four or five businesses on a map with star ratings underneath, and start comparing. At this point your reviews are doing almost all the work, and none of it involves your website. They are deciding which two names are worth a second look.

The ones that pass that filter get looked up properly. Sometimes that means tapping your website link from the map listing. Often it means a second, separate search a few minutes or a few days later — your business name, typed in on its own. Occasionally it means typing your web address straight into the browser bar because a friend mentioned you and the reviews confirmed it.

So the chain is: reviews change whether someone remembers and trusts your name, and remembering and trusting your name is what produces a visit. The review is two steps upstream from the traffic. That is why you will never see a neat line in your analytics saying "this visit came from a review", and why the effect is nonetheless one of the larger things happening to a small local business's numbers.

Branded searches and direct traffic, explained once

Two terms, and then I will stop using the jargon. A branded search is someone typing your business name into Google rather than typing what they need. "Mallow Street Bakery" is a branded search. "sourdough near me" is not. Direct traffic is a visit where the browser did not tell us where the person came from — they typed your address in, used a bookmark, tapped a link in a text message, or came from an app that strips that information out.

From here I will just call them name searches and type-ins. These two are where review effects land. Name searches go up because more people have encountered your name attached to something reassuring. Type-ins go up partly for the same reason and partly for a technical quirk: clicks from map listings, messaging apps and some email clients arrive at your site with no source attached, so they get filed as type-ins by default. If you want the detail on why that bucket is messier than it looks, we have written it up in what direct traffic actually means, and the companion piece on how referrers work covers what gets recorded and what does not.

The practical upshot: if reviews are working, the share of your traffic that arrives already knowing who you are should grow. The share arriving from generic searches might not move at all.

What the shift looks like in real numbers

Let me use round, invented figures so the shape is clear. Say a small trade business gets 300 visits a month. Sixty of those are type-ins, forty are name searches, the rest come from generic searches, a directory listing and a bit of social media. Nine people a month fill in the contact form.

Three months of asking every finished job for a review — let us say that takes you from eleven reviews to thirty-four — and the plausible picture is something like 340 visits, of which 105 are type-ins and 70 are name searches. Total traffic up by about thirteen per cent. The people-who-already-know-your-name bucket up by roughly three quarters.

Now the part that matters more. Those visitors convert at a far higher rate than strangers, because they arrived having already made a provisional decision. In that made-up example the contact form might go from nine a month to fifteen, which is a bigger proportional jump than the traffic. This is the honest reason to care about reviews: not the extra visits, but the quality of them. A visitor who searched your name by hand is a different animal from one who landed on you third in a list.

I am not claiming those are your numbers. They are arithmetic to show you what to look for. Your own ratio will depend entirely on your trade and your town.

The things reviews will not do

Reviews will not rescue a website that confuses people. If your site is slow on a phone, or your opening hours are wrong, or your contact form quietly fails, more reviews just mean more people discovering that. The extra trust gets spent in the first ten seconds and then the site has to do its job. This is genuinely the most common waste I see — effort poured into the top of the funnel while the bottom leaks. If you are not sure your site is in a fit state to receive the attention, it is worth running a report card on it before you start chasing reviews, because the fixes are usually small and the order matters.

Reviews also will not produce referral traffic worth counting. Review platforms send a trickle of clicks at best. If you are judging the effort by how many visits arrive from the review site itself, you will conclude reviews do nothing, and you will be measuring the wrong thing entirely.

And you cannot trace a single visit back to a single review. Nobody can, with any tool, at any price. What you can do is watch the aggregate over a quarter and see whether the shape of your traffic changed. That is a weaker claim than most marketers will make and it is the true one.

A routine for asking that does not annoy anyone

Almost all the awkwardness around asking comes from asking badly: too late, too vaguely, too often, and by email from an address nobody recognises. A routine fixes most of it.

The rule underneath all of this is one ask, one reminder, then silence. The people who were going to write one have written it. The people who were not will only resent you for the third message.

  • Ask within 24 hours of the job finishing, while the relief is still fresh.
  • Ask in person or in the same channel you have been using — if you have been texting, text.
  • The person who did the work does the asking. It reads completely differently from a company address.
  • Send a direct link that opens the review box, not the homepage of a review site.
  • Say what would help: "if you could mention the boiler and that we came out same day, that's the sort of thing people are looking for".
  • One reminder after four or five days, then stop.
  • Never offer anything in exchange. It breaks the rules on most platforms and it shows in the writing.
  • Reply to every review you get, including the unkind ones, briefly and without arguing.

Where to ask first, and where not to bother

For most local businesses the answer is Google first and everything else a distant second, because Google reviews are the ones that appear next to you in the map results at the moment someone is choosing. Spreading thirty reviews across six platforms gives you six listings that look neglected. Thirty in one place looks like a business people use.

The exception is trades and professions with a dominant industry directory — the sort of site customers in your field go to by habit. If yours has one, a steady handful there is worth having. A solicitor, for instance, may find a legal directory does more persuading than Google does. Check your own numbers rather than guessing: if a directory already sends you visits, it is worth feeding, and if it has sent you four visits in a year it is not.

How to tell whether it worked

Pick a start date and write it down in your analytics so the comparison is not a memory test. If you use annotations, that is exactly what a dated note on your traffic chart is for — the week you started asking, the week a bad review appeared, the week a local paper mentioned you.

Then wait ninety days and look at three things: the number of visits arriving as type-ins, the number of searches for your name, and your enquiries. Name searches come from Google's own search data rather than your analytics, which you can connect up via Search Console. Do not look weekly. Review effects accumulate at the speed of people needing you again, which for a plumber might be two years.

One newer wrinkle worth knowing: the AI assistants people now use to find local businesses lean heavily on review text when they summarise who is good at what. Specific reviews that mention the actual service tend to get quoted back. You can see whether any of your visits are coming from those tools in the AI traffic report, and it is a number worth watching now rather than in two years.

The sensible expectation

Reviews are not a traffic tactic. They are a trust tactic that produces some traffic as a side effect, and the traffic it produces is unusually good. A business going from almost no reviews to a respectable, recent, specific set of them will typically see a modest lift in total visits and a much larger lift in the people arriving already convinced. A business going from forty reviews to sixty will see much less, because it has already cleared the bar in the customer's head.

So do the asking, do it with a routine so it costs you ninety seconds a job rather than an afternoon a month, and judge it on enquiries over a quarter rather than visits over a week. And fix the website first, so the people your reviews send are not landing on something that undoes the good work.

Wondering how your website scores?

Get an instant, free report card — speed, SEO basics, security, and availability, graded A to F with plain-English fixes. No signup.

Get your free report card

Learn the details

More from the blog