28% of people have tried a local business they'd never heard of because AI recommended it.

Yext surveyed 3,848 people to get that. These are people who actually went to a business they'd never heard of, purely because an AI suggested it.

More than four in ten now use an AI tool to search for a local business in any given month, and 71% of those say they're doing it more than they were a year ago.

So what's actually going on when someone types "best gym near me"?

You get two components, not one. A map, and some text. They come from different pipes to form the answer.

Which means there are two ways for a brand to be invisible, and fixing one doesn't fix the other.

Being named in the answer is the recommendation. That's what a local business is actually after. That is the objective.

I thought I'd take a look at my gym - Rise. I've been a member there since 2023, and I absolutely love it. It's gone through an incredible transformation over the past six months and I'd be there every day if I could.

It offers so many classes, so many machines, and there's a really good vibe that I love. I wish I could go more, but as a 44-year-old man with two boys, you take what you can.

The setup is simple. They have two established locations in Welwyn Garden City and Stevenage. They have a new gym opening in Brentwood, Essex, early 2027. And I've also included a very affluent city, St Albans, in the mix where they do not have a facility.

Same prompts, same week, every major AI platform.

The four results tell you almost everything about how local AI search works.

In Welwyn Garden City: 90% visibility for "best gym near me for spin classes", ranking 1.3. 88% for one-to-one personal training. 87% for wellness and recovery facilities.

In Stevenage: 88% for "where can I train for HYROX near me". 85% for gyms with sleds and rowers. 80% for weightlifting.

An independent with two sites, beating PureGym and The Gym Group in their own towns. And look at what they win on - spin, recovery, HYROX, personal training. All the stuff I mentioned above. The classes and the kit they've put the work into are exactly what the models pick up.

So, how should a business handle a location that is still being developed? Brentwood is a new location for Rise that is due to launch in early 2027. So should you just wait until you launch or start seeding your brand early doors?

47% for wellness and recovery facilities. 46% for ice baths, ranking 1.5. 42% for reformer Pilates with a full membership.

A gym you can't join yet, appearing in nearly half the answers for the facilities it will have. The page exists, the town is named, the offer is described. That's enough to start.

If you've got a site opening, that's worth sitting with. The work starts before the doors do.

St Albans is the only location where they do not have a presence. Unsurprisingly, for all of the tracked prompts, every one returns zero. Not for ice baths, not for personal trainers, not for "how do I join a gym near me".

St Albans is eleven miles away from Welwyn Garden City and 15 miles from Stevenage, affluent, and full of exactly the people who buy premium gym memberships.

But there's no local premises, no address, no Google Business Profile, no location page. The nearest sites are too far away to count, and there are plenty of other gyms that meet the intent.

The AI isn't overlooking them. It has nothing to work with. Fitness First and Nuffield have sites there, and they own those answers.

Presence is the price of entry. A real address in the town with everything attached to it. Get that and you can beat the chains, like WGC and Stevenage do. Skip it and nothing else saves you, like St Albans.

But you can start early, like Brentwood.

One last thing. The national score for this brand is 25. It describes none of those four towns, and it tells you to do nothing.

Taking a look at ChatGPT, the map is a separate component, drawn by Mapbox and filled with structured business records. Name, address, hours, rating, phone.

The text underneath is different. That's built from web pages, and it shows its working in the links below the answer.

So you can be a pin on the map and absent from the text. Or cited five times in the text and missing from the map entirely. Different components to optimise for.

So I took a look under the hood using Chrome Dev Tools. Don't worry, I won't bore you with talk of the Performance tab and all that. I've always wanted to build Obsero and this newsletter for marketers, not technologists. But when you do look under the hood, you get an understanding of how the map pack is actually constructed.

In the UK, across a couple of logged-out prompts, it was all driven by Google My Business.

The key takeaway is to get your Google My Business listing in proper shape.

Google My Business is still crucial to local search across the board, and in ChatGPT, which is what I tested this on.

When we ran the same logged-out test in the US, it was more of a mix. Most results linked out to the business's own website, but some routed through Yelp instead. Not the majority, but enough to make an impact. So if you’re a US business, it's worth claiming your Yelp listing and keeping the details accurate alongside everything else.

Here's the thing that I find most interesting.

On Google's own surfaces, the most cited source isn't a website at all. It's the Google Business Profile. 93% of AI Mode's top citations, 65% on AI Overviews.

Everywhere else it's gym websites. Not directories, not review sites, not clever content marketing. The businesses' own sites.

But which ones, and how much, changes completely depending on where you ask.

On Gemini, Rise's own website is the single most cited domain, 1,717 appearances, nearly double the next one. On ChatGPT it's fifth, behind Hussle, PureGym and Better. On Perplexity it doesn't appear in the top five at all - that's Gym Group, Nuffield, Hussle, PureGym, Better.

Same brand. Same month. Same prompts.

Two other things worth knowing. On AI Mode and AI Overviews, Facebook and Instagram rank above most gym websites, so your social profiles are doing more work than you'd think. And on ChatGPT, the most cited domain in the whole category is Hussle, an aggregator, ahead of every gym including the national chains.

Then this made me take a double look. Across the 3,445 cited pages we track, the only ones that mention Rise by name are Rise's own pages. Every competitor page cited: no mention. So the brand gets talked about when it publishes, and essentially nowhere else.

Which means their own site is doing all the work. But it also tells you something about local. In a national category you'd need third-party coverage to compete. Here, with fewer players fighting over the same town, your own site and your own profiles can carry it.

And within that source, the location pages matter more than the homepage. Rise's Welwyn Garden City page appears across 45 prompts, Stevenage 40, Brentwood 36.

The chains play the same game - Gym Group's WGC page across 44, Fitness First St Albans across 44.

No presence in the town, no page. And without a page, no presence in the answer.

AI is moving from experiment… to essential.

Every major industry is integrating it.
Every major company is investing in it.

By late 2025, AI was already an $800B market — growing at a pace that could push it well beyond $1 trillion in the years ahead.

Cloud infrastructure is scaling fast.
AI-enabled devices are multiplying.
Automation is becoming standard.

But here’s the real question…

When trillions flow into this transformation — which stocks stand to benefit most?

Our new report reveals 10 AI stocks positioned across the backbone of this shift — from the companies powering the infrastructure… to those embedding intelligence into everyday systems.

If you want exposure to one of the defining growth trends of this decade, start here.

Download the Report Now

Doing all of that by hand for one town takes an hour. For eight towns across six platforms every week, it's a full time job.

So we built it.

Obsero Local shows you exactly where you stand in every town you trade in, against the competitors who actually operate there. Which locations are winning, which are invisible, and what's driving the difference.

We knew early on that local was going to be a big part of our proposition, given how many businesses depend on local discovery. As that discovery moved into AI platforms, following it was the obvious next step.

I’m so pleased with our local product and know it will help thousands of local businesses get ahead in AI search.

That is it for this week - have a good one.

Andy