Paediatric bodies in most countries now recommend that swimming lessons can begin around age one, and drowning remains among the leading causes of death for children in the one to four age band. Which means a specific thing for your business: a parent of a one year old is looking for you, feels urgency about it, and increasingly does not start that search on Google.
She asks an assistant. She types something close to what she would say to a friend, which is usually a full question rather than three keywords. When should my baby start swimming lessons and where can I find them near me. And she gets back a paragraph with two or three schools named in it, or none.
Whether you are one of the named ones is now a distinct discipline from ranking, and almost no swim school is working on it. AEO for swim schools is mostly the unglamorous business of answering the questions parents actually ask, in writing, on your own site.
Why assistants ignore most swim school websites
Open the average swim school site and count the sentences that answer a question.
There is a homepage with a photograph and a tagline. A page listing class levels with internal names like Starfish and Dolphin that mean nothing to an outsider. A timetable, often a PDF. A prices page, sometimes. A contact form. Total prose that plainly answers a parent’s question: close to none.
An assistant assembling an answer has nothing to work with there. It cannot tell from Dolphin Level 3 what age that suits or whether it involves putting a face in the water. It cannot find your view on when to start, because you never wrote one. So it uses national guidance from a governing body, a listings directory, and whichever local school did write something. That last slot is the one available to you and it is usually empty in any given town.
The fix is not technical. AEO for swim schools begins with somebody sitting down and writing the answers out.
The Parent Question Map
Parent questions fall into five clusters, and a school that covers all five has more usable material than every competitor in its area combined.
The timing cluster. When to start. Whether six months is too early, whether five is too late, whether a nervous child should wait. Whether to start before or after a holiday. This cluster carries the most anxiety and the most searching.
The safety cluster. Whether lessons prevent drowning, which deserves an honest answer rather than a marketing one. Whether armbands help or hurt. What supervision still requires once a child can swim. What water temperature suits an infant. Whether a child with grommets or eczema can swim.
The logistics cluster. What to bring. Whether swim nappies are required and which kind. What happens if a child cries the whole lesson. Whether parents get in the water, and at what age that stops. Changing facilities, whether there is parking, how early to arrive. Whether you can watch. This is the least glamorous cluster and it converts hardest, because it is what a parent is actually nervous about the night before.
The progress cluster. What a child can do after ten lessons, honestly. What the levels mean in plain language rather than animal names. How long until a child swims a length unaided. Why progress stalls around age five, which it commonly does. When to stop lessons.
The money cluster. What it costs, whether you can pay monthly, what happens to a missed lesson, whether there is a sibling discount, whether you can pause over winter. Schools hide this and it is one of the most frequently asked things an assistant gets and cannot answer for you.
Write one page per question. Use the parent’s words in the heading. Answer in the first two sentences, then expand. That inverted order matters more for AEO than anything else on this list, because an assistant extracting an answer takes what comes first.
Translate your level names while you are at it. Internal names are a genuine barrier. A page titled Dolphin Level 2 is unmatched to any question a human being has ever asked. The same page titled what your child learns once she can swim five metres unaided, with the internal name mentioned in the text, becomes retrievable. Keep the names for your own operations and stop using them as the public entry point.
Do the same with your timetable. A PDF of a weekly grid is close to invisible. The same information as an HTML table, with ages and levels described in words, can be read and quoted. This single change is one of the most common reasons a school with good provision loses to a school with worse provision and a plainer website.
One caution on volume. Twenty five thorough pages beat two hundred thin ones. Pages generated to cover every permutation of a question tend to say nothing specific, and material that reads as mass-produced is not what gets cited. Write the ones you can answer properly and leave the rest.
Answer the hard one honestly
The question a school most wants to avoid is whether lessons prevent drowning, and the honest answer is what earns you citation.
Lessons substantially reduce risk and do not eliminate it. Research broadly supports that formal instruction lowers drowning risk in young children, and no competent instructor claims a four year old who has had a term of lessons is safe unsupervised near water. Say both halves. Explain that layered protection means barriers, supervision, lessons and knowing what drowning looks like, and that lessons are one layer.
A school that writes that paragraph becomes a more trustworthy source than one that implies a certificate makes a child waterproof, and assistants preferring cautious well-sourced material is a real pattern rather than a hopeful one. It also happens to be true, which is the better reason.
Make the boring facts machine readable
The single most common reason a school gets left out of a local recommendation is that its practical details are not stated as text.
Put your location, service area, opening hours, age ranges, price range and contact details in plain HTML on the page. Not in an image. Not in a PDF timetable. Not exclusively in a booking widget that loads after the page does, because whatever cannot be read without executing a script frequently is not read at all.
Name the places you serve the way a parent names them. Suburbs, neighbourhoods, nearby towns, the school catchments people actually use. A parent asks for lessons near a place name, and if that name appears nowhere on your site you are not a candidate.
Keep your Google Business Profile accurate and consistent with the site, because that profile still feeds a great deal of what assistants return about local businesses. Same name, same hours, same phone number. Inconsistency between the two is a common and quiet cause of being skipped.
Add FAQ structured data to the question pages. It is not magic and it is cheap, and it makes the question-and-answer relationship explicit rather than inferred.
Check what your booking system is doing to you. Many swim schools run enrolment through a third party platform, and in the worst arrangements the class information, prices and availability live entirely on that platform’s domain rather than yours. Every piece of specific, useful, citable text about your school then accrues to somebody else’s website. If that describes you, duplicate the substance as readable pages on your own domain and let the booking link sit at the end of them.
Say your own name and location together in plain sentences, more than once, across the site. It sounds trivial. A surprising number of school websites never state the city they operate in as text, because the address lives in a footer image or the assumption is that everybody already knows. An assistant answering a question about a specific town needs that town written down.
Collect the corroboration that makes you a safe recommendation
A site full of good answers still loses to a school that appears consistently in several places, because an assistant recommending a business to a parent is making a small bet and prefers claims it can see more than once.
Start with the records that verify rather than merely list. Your national or regional swimming governing body’s register of approved providers. Your instructors’ qualifications with the awarding body named. Any local authority or leisure trust partnership. These carry disproportionate weight because a third party checked something.
Then the ordinary ones, done consistently. Google Business Profile completed rather than claimed and abandoned, with hours, photographs, service area and a real description. Local directories and parenting sites for your region. The leisure centre’s own website if you operate from one, with your school named as text rather than as a logo.
Consistency across all of them is the part that matters and the part that gets neglected. One identical business name, one phone number, one address format, one set of opening hours. A school whose details disagree across four listings has made every listing less trustworthy, which is worse than having fewer.
Reviews contribute here too, and specific ones contribute more. A review mentioning that an instructor handled a terrified four year old patiently is evidence for a claim on your site. Fifteen reviews saying excellent lessons are not evidence of anything in particular.
Check whether it worked, because nothing will tell you
There is no analytics panel for this yet, so testing is manual and worth scheduling.
Once a quarter, sit down with ChatGPT, Perplexity, Gemini and Google’s AI results and ask the twenty questions from the Parent Question Map as a parent in your town would phrase them. Write down which schools get named, what gets said about yours, and anything stated about you that is wrong. Wrong information is the most actionable output, because it usually traces to a stale directory listing or an old page you forgot existed.
Watch your own enquiries for a second signal, because it arrives before anything else changes. When parents start phoning and repeating a phrase from one of your pages, or asking a question in the exact framing you used, the material is being read and passed along. Ask new enrolments how they found you and record the answer, including when the answer is that an assistant suggested you. That is currently the only attribution data available in this channel and it costs one question at the point of booking.
Then look at what the assistants cite. If they are pulling from a national governing body and a listings site, you have a clear gap to fill. If they start pulling from your question pages, the work is landing.
None of the work in AEO for swim schools is difficult. Most swim schools will not do any of it, which is the opportunity. The parent of a one year old is asking a machine where to go, the machine is answering from whatever text it can find, and in most towns nobody has bothered to write that text yet.