Pilates grew faster than any other fitness activity in America between 2019 and 2024. Yoga grew more — in absolute numbers it added the most people of anything — and is still the larger of the two. Both of those are true, and the gap between how they feel and what they measure is what this page is about.
Every discipline gets a decade. What decides the shape of it is not a business model and certainly not the law, because the courts have already ruled that nobody can own a method of exercise — they ruled it for Pilates in 2000 and for Bikram’s yoga sequence in 2015. What decides it is demand, and demand moves for reasons that have very little to do with the practice itself.
Reformer Pilates is the loudest thing in fitness right now, and the numbers behind the noise are real. They are also smaller than the noise suggests, which is the usual shape of these things.
| Studios | Note | |
|---|---|---|
| Club Pilates, 2015 | <50 | the year Allison Beardsley sold it |
| Club Pilates, end 2024 | 1,100+ | |
| Club Pilates, end 2025 | ~1,414 | $1.14bn in system-wide sales, about 65% of all Xponential revenue |
| Solidcore, 2026 | 110 | targeting 250 US studios by 2028; L Catterton took a majority stake in September 2024 |
| JETSET Pilates, 2026 | ~60 | and 350+ territories sold |
| BODYBAR Pilates, 2026 | 73 | and 190 signed development agreements |
| Pilates Addiction, 2026 | — | 300+ territories sold across 27 states in its first year; plans 100 openings in 2026 |
| CorePower Yoga, 2024 | 220 | still the largest single yoga studio brand in the country |
| Every US Pilates and yoga studio | 37,317 | $19.2bn of revenue, and no company above 5% share |
Pilates grew at the highest rate of any fitness activity between 2019 and 2024. Yoga added the most people over the same period, and remains the bigger activity in absolute participation. Both facts get reported, rarely in the same sentence, and the second one is the one that gets dropped — a fast-growing small thing makes better copy than a large thing growing steadily.
One franchise passing 1,400 studios is a genuinely large business and a small share of how Americans actually move. The most common forms of exercise in the country remain treadmills, free weights and jogging, and it is not close. Planet Fitness on its own finished 2025 with 20.8 million members across 2,896 clubs, having opened 181 of them that year. The boutique category is the one worth writing about. It is not the one most people are in.
Club Pilates is the largest Pilates brand in the world and it is about four percent of the storefronts — of an industry the census counts as Pilates and yoga studios together, 37,317 rooms and $19.2bn, in which IBISWorld reports no company anywhere above five percent share. Most Pilates in America is one person who owns one room, and the saturation everybody can feel is coming from them rather than from the franchises: the share of studios offering Pilates at all went from 17% in 2021 to 45% in 2025.
The newer franchise brands are worth reading by a different number. JETSET has awarded 350+ territories and opened about 60 studios. Pilates Addiction sold 300+ territories inside its first year against a plan to open 100. BODYBAR has 73 rooms open and 190 signed development agreements. Selling a territory books revenue now; opening the studio is somebody else’s problem, and section VII has what that gap looks like three years on.
Growth has already cooled. Xponential’s North American same-store sales ran at 16% in 2023, 7% in 2024 and 0.5% in 2025. The category is still opening studios; the easy years are behind it. That is the top of a cycle, not a plateau.
None of the reasons are about Pilates. They are about what happened to the people doing it, which is generally how these cycles work.
The first three are well documented. The GLP-1 link is real as a marketing strategy and plausible as a driver, but nobody has sized it, so treat it as a mechanism rather than a measured cause.
It is tempting to explain the divergence between these disciplines by how tightly each one is owned. That explanation does not survive contact with the case law, because the courts have said the same thing to both.
In 1996 Sean Gallagher, who held registered trademarks on the name, sued Balanced Body for using it. On 19 October 2000 Judge Miriam Cedarbaum issued a 93-page opinion invalidating the marks: consumers already understood Pilates to mean a kind of exercise rather than one company’s brand of it, so “plaintiff cannot monopolize” it. The man who sued to own the name lost it for everybody, himself included. Joseph Pilates’ own word for the method — Contrology — had already vanished.
Bikram Choudhury tried the same thing from the other direction, claiming copyright in his twenty-six-posture sequence and licensing it to studios. A district court rejected it in December 2012, and the Ninth Circuit affirmed on 8 October 2015: the sequence is “an idea, process, or system designed to improve health,” and copyright protects only the expression of an idea — the words and photographs in his 1979 book — not the idea itself. Not a compilation, not a choreographic work. He lost, and lost again.
So the legal position is identical across both traditions: you cannot own a method of moving a body. What can be owned is a name attached to a business, a manual, a book, a teacher-training certificate. Baron Baptiste holds fourteen trademarks; the Baptiste affiliate network still went from about 200 studios to 29. CorePower holds no ancestral claim at all and has 220. Club Pilates could not own the word if it wanted to, and has 1,414.
The three outcomes have no relationship to the strength of the claim. I had this page arranged the other way round when I first wrote it, with the trademarks doing the explaining, and it was too simple to be true.
The courts closed the front door. They did not close the other two, and almost all of the money in Pilates has walked through them.
You cannot trademark the name of a method once people use it as a common noun, and you cannot copyright a sequence of postures. You can patent a machine. Joseph Pilates did — his equipment patents are the part of his estate that behaved like property, while Contrology, the word he actually wanted, is gone. The ruling that freed the name in 2000 left the apparatus and the credential exactly where they were.
In early 2025 Balanced Body acquired Merrithew. Balanced Body, in Sacramento since 1976, and Merrithew, in Toronto since 1988 and the owner of STOTT Pilates, were the two dominant reformer manufacturers and the two most widely adopted contemporary certification systems. They are now one company, serving more than 150,000 certified instructors, and professional studios went from three major suppliers to two.
Balanced Body is the defendant that won in 2000 — the company Sean Gallagher sued, whose victory made the word Pilates free for everybody to use.
Twenty-five years later that same company owns the machine you do it on and the certificate that says you may teach it. This is not a betrayal of anything; it is what the ruling made available. Once the word is free, the value moves to whatever is not.
Lagree is the live version of the argument. Sebastien Lagree holds well over a hundred patents on the Megaformer and enforces them. He sued Solidcore’s founder in 2015 over a $65,000 licensing agreement, alleging she had stripped his name off his machines and rebranded them as her own; she countersued, built her own machine, and the two settled. More recently he has sued a Florida studio for infringing 23 patents by swapping licensed Megaformers for overseas copies while still selling the classes as Lagree. He also insists, firmly and often, that Lagree is not Pilates — which is a sound methodological position and an even sounder trademark strategy, given what happened to the last person who let the word go generic.
So the honest statement is narrower than the heading above it. Nobody owns the movement. Somebody owns the apparatus and somebody owns the credential, and both are consolidating while the method itself stays free. Any yoga teacher should recognize the arrangement immediately: the asana is free and always has been; the two hundred hours are not.
The last fifteen years produced six ways of organizing the same impulse, and they are now far enough apart to compare. Overlapping decades, largely the same customers — and the difference between them is not the workout. It is who, if anyone, holds the thing.
| Rooms | Structure, and how it is going | |
|---|---|---|
| Barry’s from 1998 | 89 | Invented the group HIIT class in a red room in West Hollywood and then largely refused to scale: 89 studios in fifteen countries after twenty-eight years. Company-operated rather than franchised, premium-priced, took outside money from Princeton Equity in January 2025. Small on purpose. |
| Orangetheory from 2010 | 1,500+ | Franchised hard on a heart-rate-monitor gimmick that made the workout legible, and was then absorbed: merged with Anytime Fitness’s parent in April 2024 into Purpose Brands, 7,000+ locations and $3.5bn in combined sales, Roark Capital behind it. The founder’s method became a line item in a holding company. |
| CrossFit from 2000 | ~9,900 | Licensed affiliates paying an annual fee, near-total autonomy inside the box. Peaked around 15,000 in 2018. Below 12,500 by 2023, below 10,000 by Christmas 2024. Berkshire Partners bought it from the founder in 2020 for a reported $200m and is now looking to sell. |
| Hyrox from 2017 | 5,000+ | Owns no rooms at all. It owns a race — eight 1km runs alternating with eight workout stations, identical in every city, so a time in Hamburg means the same thing as a time in Dallas. Gyms affiliate in order to train people for it, about 5,000 of them by the end of 2024, and the revenue is in entries. First race, Hamburg, April 2018: 650 people. The 2025/26 season: over 1.5 million, across more than 200 days of racing. The 2026/27 calendar targets two million over 100+ race weekends on six continents. It has moved into the cultural position CrossFit vacated without taking on what made that position fragile. |
| Calisthenics no founding date | — | No brand, no franchise, no fee, no equipment. Grew on public infrastructure — cities have spent the last decade building calisthenics parks — and on video. The world championships have run since 2011 and draw athletes from sixty countries. Nobody owns it and nobody can sell it, which is also why nobody reports its size. |
| Running no founding date | — | The biggest of the six and the only one with no company in it anywhere. Run clubs are the fastest-growing social format in fitness: club creations on Strava rose 3.5× year over year through late 2025 and club-organized events 1.5×, on a platform that added 35 million registered users in twelve months. The 2026 London Marathon ballot took 1,133,813 applications — a world record, and 36% above the record set the year before. Free to join, nothing to buy, nobody to pay. |
In June 2020 the founder posted about the killing of George Floyd. Within weeks more than 1,350 gyms announced they were de-affiliating — not because the training changed, but because the name on their door had become a liability. Glassman sold and left.
That is the failure mode a licensed-affiliate network has and the others do not. A franchise transmits a brand; an affiliate network transmits a reputation, into every room, instantly, with no way to opt out except to leave. The decline since has other causes — a 50% affiliate fee increase, a death at the 2024 Games, Open registrations down 32% to their lowest since 2014 — but the shape was set in one week by something that happened on a social media account.
This is the honest version of the ownership argument, and it is narrower than the one this page originally made. Ownership does not determine whether a discipline grows. It determines what failure looks like when it comes. Nobody owns calisthenics, so calisthenics cannot have a bad week. Nobody owns Pilates, so no Pilates studio has ever had to change its sign because of something its founder said.
It cuts the other way too. The fastest-growing row on that table is Hyrox, which is owned, so being unowned plainly does not cause growth. What the two unowned rows have instead is that nothing can be taken away from them: calisthenics and running spread at whatever rate the interest allows, with no franchise agreement setting the pace and no founder capable of costing them a decade. Hyrox is the instructive case in between. It gave the training away to five thousand gyms it does not own and kept only the finish line, which is the one part people will queue to pay for.
Everything above is counted in storefronts. The largest single thing that happened to American exercise in the last ten years did not have any.
Peloton went public in September 2019 and peaked on 14 January 2021 at a market capitalization of nearly $50bn. It is worth about $2.25bn now. Paid connected-fitness subscriptions stood at 2.553 million at the end of June 2026, down 8.8% on the year, with another ~10% decline guided for the quarter after. Fiscal 2026 was the first profitable year in the company’s history — $63m of net income — earned by raising prices and cutting costs while the subscriber base shrank.
That is the whole cycle in six years, faster than Pilates and faster than CrossFit, and it is the only case on this page where the boom and the bust had the same cause. Hardware demand does not just rise, it pulls forward: everyone who was going to buy a bike bought one in eighteen months, and there was no second wave behind them.
Lululemon paid $500m for Mirror in June 2020 and took a $442.7m post-tax impairment on it in March 2023, then shut the hardware down and signed a five-year deal for Peloton to supply the content instead. An apparel company bought a screen at the top of the cycle and wrote it off inside three years, and the company it had been trying to compete with now makes its classes.
None of it shows up in a studio count, and studio counts are what the rest of this page is built on. CorePower is the largest yoga studio brand in the country with 220 rooms. Peloton has 2.55 million paying subscribers, has carried yoga since 2018 with its own roster of instructors on salary, and it is routinely described as one of the platform’s largest non-cycling categories.
The company does not break out participation by category, so that is where the evidence stops and I am not going to pretend otherwise. But the shape of the problem is clear enough: a frame that counts rooms will always report the studio brand as the larger thing, and on the only measure that matters — how many people take a yoga class this week — it may have that backwards.
It changes the teaching job too. A Peloton instructor is a salaried employee of a media company with a production schedule, which is a different occupation from the one the rest of this page describes, and it pays differently in both directions.
The interesting half of a cycle is the other half. Several formats that looked permanent fifteen years ago are now gone or badly reduced, and none of them lost a lawsuit.
Yoga itself did not fall. Studios and brands fell; participation kept climbing, and over the last five years yoga added more people than any other fitness activity in the country. That distinction is the one worth holding on to: the discipline and the businesses built on it are different things, and they do not rise and fall together.
Three things follow from all of this, and only the third one is uncomfortable.
The format you trained in is not the format you will finish in. Spin instructors in 2015 were not wrong about spin; the demand moved. Anyone teaching one branded method exclusively is making a bet on a cycle, and the cycle is shorter than a career.
What transfers is not the brand. It is whether you can see a body, sequence for the room in front of you, and say a useful thing at the right moment. A reformer studio hiring in 2026 wants the same underlying competence a vinyasa room wanted in 2014, which is the argument the practice page makes from the other end.
And the thing you paid to be certified in cannot be owned by the people who certified you. That is what both courts said. The sequence is an idea; the certificate is a business arrangement. Knowing the difference is worth something, particularly when the annual dues come round.
Which is where this connects back to the interpretations: a tradition survives by being exported, not by being held. Pilates lost its name and went everywhere. Bikram’s sequence outlived his brand precisely because he could not stop anyone teaching it. The methods that vanish are usually the ones nobody was allowed to copy.