Your real enemy isn't the algorithm
It's inconsistency. And it hides in three places most school owners never think to look.
Ask a room of martial arts school owners why enrolments are down and you'll get the same answers. Reach is worse than it used to be. The algorithm changed. Facebook is dead. Everyone's on TikTok now and I haven't got time to learn it.
I've said versions of all of it myself. It's a comfortable answer, because it puts the problem somewhere you can't be blamed for it.
But it doesn't survive contact with the numbers.
You've already tried the usual three
Nearly every owner I speak to has been through the same sequence, in roughly the same order:
- Boosting random posts. A tenner here, twenty quid there. Likes go up. Trial bookings don't.
- Ads with no system behind them. The leads actually arrive — and then sit in an inbox. No follow-up process, no booking system, no idea what's working.
- Hiring an agency. Big promises, monthly retainer, lock-in contract, and leads who never show up.
Here's the part worth sitting with: none of those three failed because of reach. The boosted posts were seen. The ads delivered leads — sometimes plenty of them. The agency did put your name in front of people.
They failed because nothing underneath them was repeatable.
Where inconsistency actually lives
It's almost never in the marketing. It's in the three joints either side of it.
How enquiries are handled. A lead comes in at 8pm on a Tuesday, halfway through a teen class. Sometimes it gets answered in ten minutes. Sometimes it gets answered on Thursday. Sometimes an instructor replies, sometimes you do, and the two of you say completely different things. That isn't a marketing problem — the marketing worked. It's a handling problem, and it stays invisible because nobody logs the ones that went cold.
How students are kept. Most schools are pouring water into a leaky bucket and calling the leak normal. Lose twenty over a summer, sign twenty in September, and you've run very hard to stand still — then concluded you need more marketing. One school in our network lost five students over summer 2026, against twenty the summer before. Same school, same town, same weather. A retention system in between.
How sales conversations happen. A trial turns up and what happens next depends entirely on who's on the mats. You can close, because it's your school and you believe in it. Your newest instructor freezes and says "have a think about it". Neither of you is wrong. There just isn't an agreed way to do it.
The test that tells you the truth
Here's the one I'd use, because it can't be argued with and you can run it this month.
Can your team run the school for three consecutive days without you — enquiries answered, trials booked and taught, and nobody phoning you for a decision? Not "would it survive". Would it run.
If the answer is no, more marketing will not fix your school. It will expose it. Turning up the top of the funnel while the middle leaks just means you pay to lose people faster.
That's the uncomfortable bit. Most owners' instinct when enrolments dip is to spend more on attention. It's the one move that reliably makes a leaky school worse.
What to do instead — and the order matters
The fix isn't more hustle. It's a growth system, installed in the right order.
Order is the whole game. Fix retention before you scale marketing, or you fill the bucket faster than it drains. Fix the sales conversation before you buy more leads, or you pay for trials that were always going to walk. Build the team before you try to buy your time back, because automation applied to an undefined process just gets you the wrong answer faster.
That's why the system runs as three pillars — members, then money, then time — rather than nine things to have a go at. The nine steps are laid out on the homepage if you want to see where your own school actually sits.
And if you take nothing else from this: stop asking why reach is down. Start asking what happens to an enquiry that arrives at 8pm on a Tuesday while you're on the mats. That answer is worth more than any algorithm update.