How it works
Anyone can tell a gym owner to “post more Reels.” The part that is hard, and the part I built, is the layer underneath: the sourcing, the competitor audit, and the capacity maths that make a recommendation checkable rather than plausible.
Start here
Nothing is bought from a page. What you’re weighing up before you pay is a fifteen-minute conversation, and what follows it is one engagement in two parts.
Free, about fifteen minutes, and it happens before you pay anything. By the time we speak, I’ll already have looked at your catchment: who’s nearby, what they charge, who’s running ads. You get a straight read on whether there’s a case worth pursuing, and the pricing, live, for your gym. If your catchment’s already taken, or too thin to source properly, you’ll hear that instead.
One-off. Every competitor in your catchment priced and profiled, who’s running Meta ads right now, your district’s demographics and income data, and your capacity gap sized in pounds against your own membership fee, with a ranked, three-action plan. It comes bundled with the execution half: a 48-hour competitor brief, a setup session with ads live before it ends, drafted creative, a month of content, and a costed intro offer.
Ongoing, month to month, no minimum term. The research stays current instead of frozen on the day you signed: competitor alerts when someone nearby changes price or opens, creative rebuilt quarterly, research refreshed quarterly, and a standing line to think things through.
Covered by the same 90-day, full-refund guarantee as everything else.
The method
None of it is a black box, and that is deliberate. Every step below is one you could check yourself, given the time. The value is that it has been done properly, priced against your own numbers, and written down with the sources attached.
01
Every gym and studio within reach of yours, looked at the way a prospect would: current prices from their own site, their class model, and their reviews where a real review base exists.
02
Meta’s Ad Library shows every ad a page is running, right now. I pull who’s advertising near you, what they’re saying, and who’s gone quiet, which is often just as useful.
03
ONS Census, income and activity data for your district, not the national average: age bands, household composition, what people near you actually earn.
04
The space between the members you have and the members you could hold, priced at your own monthly fee. Your number on your fee, never an industry estimate.
Your current member count, against the capacity you set. The spare places, multiplied by your own monthly fee, over a year. That’s the whole calculation: no modelling, no assumptions borrowed from another gym, and every input is one you can read straight off your own books.
Why it matters that the maths is boring.
A number you can’t interrogate is a number you can’t act on with any confidence. Every figure in your report comes with the working attached, so when something looks off to you (and you know things the data does not), you can argue with it, and win.
Sourcing
No number reaches your report without a source behind it. If a figure can’t be confirmed, it is left out rather than estimated, and the report says so.
Before it reaches you
The last step isn’t writing. It’s three separate reviews, each looking for a different way the report could be wrong.
And when something still gets through, it goes in a correction log: what was wrong, what it’s been corrected to, and what changed so it doesn’t happen the same way again. Showing you how I check my own work beats claiming it never needs checking.
01
The load-bearing claims are traced back to the original source. Special attention goes to hedged findings that quietly lost their caveat, and to numbers that drifted during editing.
02
Read the way a sceptical gym owner would read it. Does it sound specific and checkable, or does it sound like AI marketing copy? Generic paragraphs get cut, not softened.
03
Can each insight actually be found and understood on a phone, in one sitting? A true finding buried in a wall of text has failed just as badly as a wrong one.
The method matters less than the output. Read the sample report and judge it on whether it would tell you something you didn’t already know.