8 of 11 Fitness Sites Name a Coach, None Up Front

A buyer picking a Pilates studio or a personal trainer is not buying a room. They are buying a person, and they want to know who that person is before they pay. So on 26 September 2026 we took 12 Singapore fitness websites that appeared for real buyer searches and checked one thing above all others: can you find out who will actually train you, and can a machine find out too?

The people are on the website. They are just not where it counts. Of the 11 sites we could read, 8 named at least one coach or instructor on a team page. Exactly 1 declared a person in its homepage structured data. And not one of the 11 named a coach in its opening paragraph.

That gap is the whole finding. These studios have the strongest trust asset in their industry sitting two clicks deep, in a format nothing can quote.

The 12 sites, and how we picked them

We ran four buyer searches through SingRank's self-hosted SearXNG metasearch on 26 September 2026: "personal training studio singapore", "pilates studio singapore price", "muay thai gym singapore" and "strength and conditioning gym singapore". From the results we kept only Singapore fitness businesses' own websites. Directories, listicles, social profiles, news sites, overseas gyms and instructor-training schools were dropped.

That left 12 sites. We fetched each homepage without running JavaScript, because AI crawlers generally do not run it, then followed up to three internal links that looked like a team page and up to three that looked like a pricing page. We also fetched each robots.txt.

One site, reached at its homepage, returned an HTTP 521 error from its network provider instead of a page. We could measure nothing on it. That leaves 11 measurable sites, and every count below is out of 11 unless it says otherwise.

We checked the roster of businesses SingRank works for first. None of the 12 is a client, so no result had to be swapped out.

8 of 11 sites name a coach somewhere

According to our sample, 8 of the 11 readable sites named at least one coach, instructor or founder by name. The names sat on a team page, an instructor page, or a per-location staff profile.

Three sites named nobody. On one, the about page introduced the business as a group of coaches without a single name attached. On another, the pages our crawl found under the word "instructor" turned out to be instructor-training courses the studio sells, not the people who teach its classes; its sitemap of 26 URLs held no team page at all. The third had no staff page we could find.

The 8 that do name people often do it well. One site listed seven teachers with full names, each with a written background. Another split its instructor profiles by branch, so you could see who teaches at Bukit Timah and who teaches at Orchard.

That is a real asset, and most of these owners have already paid for it. Someone wrote those biographies.

Only 1 of 11 tells a machine who the coach is

Here is where it falls apart. On the 11 homepages we read, 1 declared a Person entity in structured data. That site also declared an ExerciseGym and a credential type, and it named three founders in its markup, including a doctor of medicine.

One other homepage did carry Person markup, and we nearly counted it. Reading the code, every one of those entities was the author of a testimonial, and each author was named "admin". The markup describes reviewers, not staff. So it does not tell a machine anything about who teaches there.

Schema.org defines Person plainly as "A person (alive, dead, undead, or fictional)", and gives it properties built for exactly this job: hasCredential is "A credential awarded to the Person or Organization", jobTitle is "The job title of the person", and worksFor is "Organizations that the person works for". Nine of the 11 sites used none of them.

We are not claiming markup buys you recommendations. It does not. But if you have already written a biography for seven teachers, leaving that out of the machine-readable layer is work you have done and then thrown away.

Not one opening paragraph named a person

Zero of 11. We read the first substantial paragraph of every homepage by hand, and none of them named a coach.

What they opened with instead: a slogan ("Strength training made easy"), a promotional banner, a count of studios worldwide, or a statement of the studio's philosophy. One opened with the words "Discipline creates champions" and did say "Coach-led every session" further down the same block, but named no coach.

This matters more than it sounds. The opening paragraph is the passage a search system is most likely to lift, and it is the first thing a buyer reads. Google's own guidance on creating helpful, reliable, people-first content asks site owners to self-assess with questions including "Is it self-evident to your visitors who authored your content?" and "Is this content written or reviewed by an expert or enthusiast who demonstrably knows the topic well?"

In a trade where the product is a named human being, an opening paragraph with no human in it is a strange choice. Our earlier study of 109 local pages across five regions found the same habit in other trades: 44 of 109 pages opened without saying plainly what the business does.

Where the credentials actually sit

Of the 8 sites that named someone, 6 showed a qualification next to a named person. We counted a named certification body, a graded level, a diploma or degree, or a stated clinical background.

The named systems we saw included STOTT, BASI, APPI, Peak Pilates, a CrossFit level, and physiotherapy training. Two of the 8 named their coaches but attached no qualification to any of them.

So 6 of 11 sites, just over half, let a buyer check a specific person against a specific standard. That is the single most persuasive thing a boutique studio owns, and it is invisible to anything that reads only the homepage.

What these sites did well

Two things, and they deserve credit, because both were better than we expected.

Price. We found a price on 9 of the 11 sites, either on the homepage or on a pricing page it linked to. Five showed one on the homepage itself. The figures were specific: monthly group memberships, per-session private rates, trial packs, unlimited plans. Two sites showed no price anywhere we looked. That is a stronger showing than the 99 of 211 across 20 trades we measured yesterday.

A way in for a first-timer. 9 of 11 named a trial, intro pack or first-session offer, usually with its own price.

Two more counts worth having. 8 of 11 used at least one question as a heading, which is the format an answer engine can quote cleanly. And 8 of 11 had a WhatsApp link on the homepage, against 3 with a tappable phone number.

Three sites carried no business entity at all

Of the 11 homepages, 5 declared a business-type entity: ExerciseGym, LocalBusiness or SportsActivityLocation. Three declared only a generic Organization. Three declared neither, carrying nothing beyond WebSite and page-level markup.

Six of 11 showed a Singapore postal address on the homepage. Four named their neighbourhood in the opening paragraph: the CBD, Orchard, Aljunied, and a station name. The other seven left a reader to hunt for where they are.

On AI crawler access, the answer was the same as every sample we have taken this month. None of the 11 robots.txt files blocked GPTBot, OAI-SearchBot, ChatGPT-User, PerplexityBot or ClaudeBot from their public pages. Two files named those agents explicitly and allowed them, excluding only admin and API paths. The door is open, which matches our census of 219 Singapore sites.

We also ran our live audit tool on 6 of the sites, over 8 attempts. It returned for 5 of them and timed out at 60 seconds on the sixth. Three of those runs included a mobile speed test: the PageSpeed scores were 45, 50 and 82, with largest-contentful-paint times of 18.5 seconds, 5.7 seconds and 3.0 seconds. The other two ran with the speed test switched off. One homepage held 183 words. One site's blog still carried three untouched template posts titled "Blog Post Title One", "Two" and "Three", each with the meta description "It all begins with an idea."

What I would fix first, in an afternoon

Three edits, in this order. None needs a developer.

• Put a name in the first paragraph. Name the studio, the discipline, the neighbourhood, and the person who leads the teaching. One sentence. If your head coach has 300 fights or a physiotherapy degree, that belongs in line two, not on page four.
• Move one credential up. Pick the coach a buyer is most likely to train with and put their certification on the homepage next to their name. Keep the full team page; just stop making it the only place.
• Mark up the people you already describe. Add Person for each coach on your team page, with jobTitle, hasCredential and worksFor pointing at your business entity. If your homepage has no LocalBusiness or ExerciseGym entity, add that first.

If your timetable is full and you are turning people away, do none of this. Fix it when you have empty slots.

What it costs

Honesty about the trade-offs, because there are some.

Naming people publicly has a cost. Coaches leave, and then your homepage is wrong until someone updates it. Naming a credential invites a buyer to compare it against a competitor's, and you may lose that comparison. Putting a lead coach front and centre makes your business look like one person, which can be a problem when you want to sell it or when that person is on leave.

And the honest limit on the upside: none of this makes an assistant recommend you. AI platforms choose their own answers. What these edits do is make sure that when something does describe your studio, it has your words and your coach's real credential to work from, instead of a listicle's summary.

Method and limits

Read these before quoting anything above.

• Metasearch is not Google. Results came from SingRank's self-hosted SearXNG on 26 September 2026. Six engines returned errors on each of the four searches, so the ordering is a blend of what answered, mostly Yandex and mwmbl. It is not Google's ranking and not the Maps pack.
• One day, four searches, 12 sites. A snapshot of one morning, and a small one. A single site changes any count here by roughly 9 percentage points.
• 11 measurable, not 12. One site returned HTTP 521 and was excluded from every count.
• Some calls were made by hand. Whether an opening paragraph names a person, and whether a qualification sits next to a named person, were read and judged by us against a written rule, not matched by a pattern. Another reader might score one or two sites differently.
• Raw HTML only. We did not run JavaScript, so anything injected by a script, a booking widget or a page builder after load was invisible to us. Content behind a firewall or on a client-rendered page was not measured.
• Homepage plus a few linked pages. We followed up to three team-type links and three pricing-type links per site. A price or a team page elsewhere on the site would have been missed.
• Our audit tool did not finish everywhere. We attempted 6 sites over 8 calls; 3 calls timed out at 60 seconds and one site never returned. For another site the tool could not retrieve a robots.txt or sitemap that our own direct fetch did reach, so treat its technical row with caution. Speed figures cover 3 sites only.
• Checks are signals, not proof. Passing them does not mean an assistant will cite you, and failing them does not mean your studio is badly run.
• This page names no business. It is not a rating of any company, and every count is reported against its base.

Want us to read your studio's page?

Send us the one page that should win your discipline plus your neighbourhood. We will tell you whether a stranger, and a machine, can work out who teaches there and what it costs. It starts as a free visibility check through our contact page, and our SEO, AEO and GEO programme starts from S$500/month if you want us to do the work.

If you run a gym or a studio and want the full picture first, our guide to SEO for gyms and fitness studios in Singapore is the longer read.

These findings describe 12 websites sampled on 26 September 2026 through metasearch. No business is named, and nothing here alleges wrongdoing by anyone.

FAQ

Does naming my coaches on my website help me get found?

It helps a reader decide, which is the part that earns money. For search and AI systems, a named person with a stated credential gives them something specific to repeat about you. Google's own content guidance asks whether it is self-evident who wrote your content. Neither guarantees a ranking or a citation.

What is Person schema, and do I need it?

Person is the schema.org type for a human being, with properties including jobTitle, hasCredential and worksFor. It makes your team page readable by machines rather than only by people. In our sample of 11 Singapore fitness homepages, exactly 1 used it for its own coaches. It is optional, not required.

Should a small studio show prices on its website?

In our view, yes, at least a range or a trial price. 9 of the 11 fitness sites we measured showed a price somewhere, which is well ahead of most Singapore service trades. Buyers search for price constantly, and a page with no number sends them to a third-party guide instead.

Is a team page enough on its own?

It is most of the work, and 8 of 11 sites had one. The gap we found is that nothing on the homepage points at it: no name in the opening, no credential, no Person markup. A buyer who lands and leaves in ten seconds never reaches it.

Why only 12 websites?

Because we read every one of them properly rather than scoring a large sample automatically. Four searches through our own metasearch returned only these 12 as Singapore fitness businesses' own sites; the rest of the results were listicles, social pages, directories and gyms in other countries. Small samples carry wide error, and we say so above.

Can I cite these figures?

Yes. Please cite "SingRank, Singapore fitness website study, 26 September 2026" with a link to this page, and quote each figure with its base, such as 8 of 11. Challenges to any number are welcome through our contact page.