SEO and AI marketing specialist for fitness studiosDaniyal Sherafat

Guide

AI search visibility for boutique fitness studios, explained

What AI search visibility means for a pilates, yoga or CrossFit studio, how ChatGPT and Claude pick which studio to name, and what to fix first on your site.

Someone new to your city opens ChatGPT and types “which pilates studios are there in Rotterdam”. The assistant searches the web, reads a handful of pages, and answers with a short list of names. Your studio is on that list or it is not. That is AI search visibility for a boutique fitness studio. This page explains how the assistants pick their names, what the check on this site measures, and what to fix first.

What AI search visibility is for a studio

Classic search visibility is a position in a list of links. AI search visibility is a sentence. The assistant either names your studio in its answer or it does not, and it either points to your website as its source or it points somewhere else.

For a boutique fitness studio, four questions matter. Which studios of your kind are there in your city. Where a beginner can start. Where your studio is and what classes it runs. What a first class costs and how to book it. The first two are what a stranger asks. The last two are what someone asks once they have heard your name. Being visible means your studio is in all four answers, with your own pages as the source.

How assistants choose which studio to name

Some of this is documented by the companies that run the assistants. Some of it is inference. It helps to keep the two apart.

What is documented. When ChatGPT or Claude answers with web search switched on, the assistant runs searches, reads the results, and writes an answer with citations back to the pages it used. OpenAI’s API documentation says the response “will include inline citations for URLs found in the web search results”, and Anthropic’s says “the response includes citations for sources drawn from search results”. Both let the caller set an approximate location, such as a city and country, so that local questions get local results.

Both companies also run separate crawlers for search and for training. OpenAI states that OAI-SearchBot “is used to surface websites in search results in ChatGPT’s search features” and that sites which opt out of it “will not be shown in ChatGPT search answers”. Anthropic states that blocking Claude-SearchBot “may reduce your site’s visibility and accuracy in user search results”, while blocking ClaudeBot only affects training data. So a robots.txt rule written to keep your content out of model training can, if it names the wrong crawler, keep your studio out of the answers too.

Google says something similar about its own AI features: there are “no additional requirements to appear in AI Overviews or AI Mode” beyond being indexed and eligible for a snippet.

What is inference. Nobody outside those companies knows the exact rules that turn a page found into a name in the answer. What I can observe is that the assistants cite pages, and that a page can only be cited for a fact it contains. If your first-class price exists only inside a booking widget, or your class list only in an image, there is no sentence to quote. The plain-text items further down follow from that observation. They are the best-supported starting point, not a guarantee.

One more documented fact matters. Answers change between runs. A 2026 paper on measuring visibility in AI search puts it plainly: “answers can vary across runs, prompts, and time, making one-off observations unreliable.” One answer is an anecdote. Several are a measurement.

The difference from classic SEO, and where they overlap

Google describes SEO as “helping search engines understand your content, and helping users find your site”. That covers most of the AI side too. An assistant that searches the web depends on the same things a search engine does, because it starts from a search: crawlable pages, clear titles, text that says what the business is and where it is.

The differences sit at the edges. Classic SEO is measured in rankings, impressions and clicks. AI visibility is measured in whether you are named and whether you are cited, and the answer is one paragraph, so there is no page two. Classic SEO has years of tools and data. AI visibility has few of either, and the answers vary, so any honest measurement repeats the question.

The overlap is the useful part. Almost everything that helps an assistant read your studio also helps Google. There is no separate “AI version” of your website to build. There is one website, and the question is whether it states the facts as text.

What the SEO & AI visibility checker measures, and what it cannot

The SEO & AI visibility checker on this site does three things. It reads your public pages the way a crawler does, without running JavaScript and without logging in. It asks ChatGPT and Claude the four questions above, through their APIs with web search on and your city as the location (the two without your name three times each, the two by name once), and records how often your studio is named, how often your website is cited, and which other names appear. And it runs Google’s PageSpeed test on your homepage. Every finding carries the excerpt that supports it. There is no score.

It cannot see your Google rankings, impressions or clicks, or your Google Business Profile and reviews. It does not measure Google AI Overviews or AI Mode, Gemini or Perplexity. It does not know what a person sees inside the ChatGPT or Claude apps, which add their own memory and personalisation. It does not measure traffic, bookings or revenue. The full list is on the method page, with a longer piece on what any check of a public URL can and cannot know.

What to fix first

These are the facts a prospective member needs, and an assistant can only cite them if they are written as text on your own pages. Each is a sentence, not a project.

  1. City next to the studio name. On the homepage, in the title, in the footer. “Studio Noord” tells an assistant nothing. “Studio Noord, reformer pilates in Rotterdam” answers the discovery question.
  2. Street address. As text, not only in an embedded map.
  3. Class types, named. Reformer pilates, vinyasa yoga, barre, CrossFit, small-group training, spinning. If the list lives only in a timetable widget or an image, it is not on your website in any sense a machine understands.
  4. Beginner information. One paragraph: whether a complete beginner can join, which class to start with, what to bring.
  5. First-class price. The intro offer and its price, as a sentence near the booking button and on the pricing page. There is a separate article on it.
  6. A booking path. A link or button whose text says “book a trial class”, pointing to where the booking happens.
  7. Crawler access. Check robots.txt. Google describes it as a file that “tells search engine crawlers which URLs the crawler can access on your site”. If it blocks OAI-SearchBot or Claude-SearchBot, the search side of the assistants cannot read you. Blocking GPTBot or ClaudeBot is a separate decision about training. The checker’s own crawler is documented on the crawler page.

If your studio is already missing from the answers, read why doesn’t ChatGPT recommend my gym for the short version of the causes.

What happens after the report

The report is the end of the check. Many of the fixes above take an afternoon with your website builder, and the report lists them in order with the evidence.

If you would rather have the work done, ClickZero AI, which I run from Rotterdam, offers a visibility programme for boutique fitness studios in the Netherlands. The plain description, without prices or promises, is on what happens after the report. Talking to ClickZero is optional. The check is free, needs no account, and stands on its own.

Sources

  1. OpenAI, Overview of OpenAI crawlers (OAI-SearchBot, GPTBot, ChatGPT-User) (checked 2026-09-18)
  2. OpenAI, Web search tool (citations, user_location) (checked 2026-09-18)
  3. Anthropic, How site owners can block Anthropic's crawlers (checked 2026-09-18)
  4. Anthropic, Web search tool (citations, user_location) (checked 2026-09-18)
  5. Google Search Central, AI features and your website (checked 2026-09-18)
  6. Google Search Central, SEO Starter Guide (checked 2026-09-18)
  7. Google Search Central, Introduction to robots.txt (checked 2026-09-18)
  8. Google Search Central, Understand JavaScript SEO basics (checked 2026-09-18)
  9. Schulte, Bleeker and Kaufmann, Don't Measure Once: Measuring Visibility in AI Search (GEO), arXiv 2026 (checked 2026-09-18)

FAQ

Questions studio owners ask

Is AI search visibility only about ChatGPT?

No. ChatGPT, Claude, Perplexity, Gemini and Google's AI features all answer studio questions. The check on this site asks ChatGPT and Claude, because both offer an API with web search and citations that can be measured honestly. The plain-text fixes above are not specific to any one assistant.

Does blocking GPTBot hurt my visibility in ChatGPT?

According to OpenAI's crawler documentation, no. GPTBot is the training crawler; OAI-SearchBot is the one that surfaces sites in ChatGPT search. Blocking the second removes you from search answers. Blocking the first is a training decision. Check which one your robots.txt names.

My studio is on Google Maps. Is that not enough?

The check cannot see your Google Business Profile, so it cannot say. What it can say is what ChatGPT and Claude answered when asked about studios in your city, and whether your own website was among the pages they cited. Both are worth having.

Will fixing these things get my studio recommended?

Nothing guarantees it. Answers vary between runs and between assistants, and the selection rules are not public. What the fixes do is make sure that when an assistant reads your pages, the facts it needs are there to cite. Without that, there is nothing to work with.

How is this different from an SEO audit?

An SEO audit is about ranking on Google and needs your Search Console data to say anything about traffic. This check is about what two AI assistants say when asked, and whether the facts they need are readable on your site. It records the Google-facing basics it can see, such as titles, canonical tags, robots.txt and page speed, but it does not measure rankings.