How many people actually ask an AI for a local recommendation?
More than you would guess from your own analytics, and far more than a year ago.
Ofcom's Adults' Media Use and Attitudes report, published in April 2026 and based on a survey of 7,533 UK adults, found that 54% of UK adults now use AI tools such as ChatGPT, Copilot or Gemini. Among 16 to 24 year olds that rises to 79%, and among 25 to 34 year olds to 74%.1
General usage is not the same as local intent, though. For that, the most direct evidence comes from BrightLocal's Local Consumer Review Survey, published in March 2026. It found that 45% of consumers have used an AI tool to find a local business recommendation, up from 6% the year before, with 31% doing so at least monthly. Only 19% said they would not consider it at all.3
One caveat that most write-ups of this survey leave out: it polled 1,002 US consumers, not UK ones. The direction of travel is almost certainly the same here, given Ofcom's UK adoption figures, but the exact number should be treated as an indication rather than a UK measurement.
How much traffic do AI assistants actually send?
Very little, and you should be suspicious of anyone who tells you otherwise.
SparkToro's 2026 analysis of Similarweb and Datos clickstream data found that AI tools send less than 1% of all traffic out to the open web. The same study found 68.01% of Google searches in the US ended without a click at all between January and April 2026, up from 60.45% in 2024.4
Sector matters too. Similarweb's July 2026 analysis of ChatGPT referrals found citation rates of around 23% for travel and hospitality and 20% for automotive, but under 4% for professional services.6 If you sell a service rather than a product or a booking, you are in the category that AI assistants currently link out to least.
If the traffic is that small, why does any of this matter?
Because the AI is not the destination. It is the step before the destination.
The BrightLocal survey found that 88% of people who use AI for local recommendations then fact-check what they were told, most commonly by checking whether a review is legitimate or by looking at the source. Only 18% said they felt ready to contact a business straight off the back of an AI recommendation.3
Read that carefully, because it is the most useful sentence in any of this research. It means an AI recommendation is not a booking. It is a shortlist. The customer then goes looking for you by name, and what they find at that moment decides whether you get the call.
of people who get a local recommendation from an AI go on to verify it themselves, usually by checking reviews or the original source. Being named is the start of the process, not the end of it.
They ask
"Who is a good physio near Chesterfield?" No search results page, just a question.
They get a shortlist
Two or three names, drawn from whatever the engine can read and corroborate about local businesses.
They verify
88% check it themselves. Your name goes into Google. They look at reviews, your site and your profile.
They choose
Only 18% feel ready to make contact off the AI answer alone. The verification step is where you win or lose.
Meanwhile the older route to your website is narrowing. Pew Research tracked the actual browsing of 900 US adults in March 2025 and found that when an AI summary appeared on a Google results page, people clicked a normal search result in 8% of visits, against 15% when no summary appeared. Only 1% clicked a link inside the summary itself.5 Ofcom reports that around 30% of UK searches now show an AI overview, and 53% of UK adults say they see these summaries often.2
So the picture is not "AI replaces Google and sends you traffic instead". It is that fewer searches end in a click at all, and a growing share of the ones that do are people who already have your name.
Do visitors who arrive from AI actually convert?
When they do arrive, the evidence is genuinely encouraging, with one large caveat.
Adobe Digital Insights, working from more than a trillion visits to US sites, reported that in May 2026 AI-referred visitors to retail sites converted 54% better than non-AI traffic. On travel sites the same visitors were 21% more engaged, spent 70% longer per visit and bounced 41% less often, and yet still converted 28% worse than non-AI traffic.7
Those two numbers sitting side by side are the reason to be careful with the headline "AI traffic converts better" claim you will see quoted everywhere. It is true in retail and not true in travel, in the same dataset, in the same month.
| Sector | Conversion vs non-AI | Engagement | Traffic growth, year on year |
|---|---|---|---|
| Retail | 54% better | Higher | Up 138% |
| Travel | 28% worse | 21% more engaged, 70% longer visits, 41% lower bounce | Up 194% |
| UK service businesses | No published data | No published data | No published data |
What this looks like on a real UK site
Everything above is someone else's data. Here is ours, from a site we build and manage, because we could not find a single published example of this pattern on a small UK service business.
Chequered Flag Driving School is a driving instructor covering Caerphilly and North Cardiff. We launched their site at the end of June 2026. On 24 July 2026 a search for "best driving instructors in Caerphilly" returned their name first in Google's AI Overview, alongside their 5.0 star rating, and they were the top citation underneath it.
The interesting part is not that. It is where the visits actually came from in that same period.
This is also why we tell people not to judge this work by a line in their analytics labelled "ChatGPT". On that account there barely is one. The effect showed up as direct visits and branded searches, which is exactly the traffic most owners assume was coming anyway.
What this means if you run a one-van or one-clinic business
It means the goal is not traffic from ChatGPT. The goal is to be the name that comes up, and to hold up when someone checks.
Those two jobs turn out to need the same things: a site an engine can read without guessing, facts that agree with each other everywhere they appear, and enough corroboration elsewhere that a machine is willing to repeat your name to a stranger.
Google is unusually blunt about this. Its own documentation on AI features states that there are "no additional requirements to appear in AI Overviews or AI Mode, nor other special optimizations necessary", and that you do not need to create new machine readable files or markup. A page needs to be indexed, crawlable and eligible to be shown with a snippet.8
In practice, for a small service business, that comes down to a short list:
- Your services and areas need to exist as text. If your service list is an image, a slider or a script that runs after load, it may as well not be there.
- Your name, address and phone number must match everywhere. Website, footer, Google Business Profile, directories. A machine that finds two versions of your phone number has a reason to trust neither.
- Answer the questions people actually ask. Prices, coverage, availability, qualifications. Answer engines lift self-contained passages, so a paragraph that only makes sense in context will not get used.
- Keep reviews recent and specific. The verification step almost always runs through reviews, and 88% of AI users go looking for them.
- Let the crawlers in. If your hosting or security settings block AI crawlers, none of the above matters. This is the single most common thing we find.
None of that list is proprietary. An owner with a free afternoon can do most of it, and we would rather say so than pretend otherwise.
Frontpage AI is a UK studio that builds and manages websites for independent service businesses, structured so that Google and AI answer engines can read, trust and name them. We are based in the UK, work only with UK service businesses such as trades, clinics, salons and instructors, and charge from £148 per month on a 30 day rolling contract with no setup fee. Hosting, structured data, Google Business Profile upkeep, monthly reporting and ongoing changes are included. The driving school above is one of our partner sites.
We do this because for most owners the constraint is time rather than knowledge, and because the checking step never really ends. Listings drift, prices change, an engine re-reads your site and finds last year's information.
Who are the best driving instructors in Caerphilly?
Based on recent reviews and local coverage, a few names come up consistently:
Chequered Flag Driving School is frequently recommended for learners in Caerphilly and north Cardiff, with a high pass rate and strong recent reviews.
How to check whether AI engines can read your site today
You can do most of this yourself in about ten minutes, without any tools.
- View the page source. Right click your homepage, choose "View page source", and use Ctrl+F or Cmd+F to look for your main service and your town. If they are not in that raw text, an engine may not see them either.
- Open yoursite.co.uk/robots.txt. Check that it does not disallow crawlers you want, and that it points to a sitemap. If the file does not exist, that is worth fixing.
- Search your own business name in an AI assistant. Ask it what your business does, where it works and what it charges. The mistakes it makes tell you exactly which facts are missing or contradictory online.
- Compare your Google Business Profile with your website, line by line, for name, address, phone, hours and services. Fix anything that disagrees.
- Ask for a local recommendation without naming yourself. "Who are the best electricians in Rotherham?" See who gets named, and go and look at what those businesses publish that you do not.
None of this requires a rebuild, and none of it is a trick. It is the unglamorous work of making a business legible to a machine that is trying to be helpful to someone in your town.