Custom fitness studio software in Baltimore: when to build instead of renting by the member

The studios this is actually about

It is worth being precise about the size of the businesses in question, because almost every article written about studio software is written for a business that does not exist.

In the most recent County Business Patterns data — reference year 2023, released in 2025, which is genuinely the latest the Census Bureau has published — Baltimore City contained 54 establishments under NAICS 713940, fitness and recreational sports centers. Those 54 businesses employed 1,233 people between them and paid out $23.0 million in annual payroll. Twenty-four of the 54, a little over 44%, employed fewer than five people. Thirty-eight of them, just over 70%, employed fewer than twenty. Because the Census suppresses the largest size bands in a county this small, that 70% is a floor rather than a ceiling — the two establishments unaccounted for in the size table sit somewhere above 100 employees, and even assigning both to the top band barely moves the share.

The average annual wage across that payroll works out to about $18,700, which tells you more about how a studio runs than any other number in the dataset. This is not a workforce of salaried staff. It is a roster of part-time instructors teaching a handful of classes a week, and it is the reason instructor pay is one of the most persistently annoying pieces of studio administration — it is a lot of small, conditional calculations attached to people who are not in the building most of the time.

Widen out and the shape holds. The Baltimore–Columbia–Towson metro area has 362 such establishments and 7,462 employees; Maryland as a whole has 704 and 14,580. The proportions barely move across those three geographies, and they sit within a couple of points of the national distribution, so Baltimore is not unusual here. It is a completely ordinary instance of an industry made almost entirely of very small businesses.

One trend is worth pulling out, because we have not seen anyone else cite it and it is government data rather than a vendor survey. Between 2019 and 2023, employment in Baltimore City fitness centers fell from 1,750 to 1,233 — down 29.5% — while the establishment count only fell from 59 to 54. Roughly the same number of businesses, running with substantially fewer people. Over the same period, the adjacent code that partly captures recovery, wellness and personal care services grew from 16 establishments to 25 in the city, and statewide payroll in that code rose by two thirds. Traditional fitness contracted and has not gotten its headcount back; the wellness end of the market grew. Any studio owner who lived through those years already knows this in their body. It is useful to know the data agrees, and it explains why so many studios now sell something — recovery, sauna, cold plunge, small-group training — that their software was not designed to schedule.

What the platforms charge, and what they are charging you for

We went through the pricing pages in this category in July 2026 the same way we did for law firms and nonprofits, and the result was the same in structure and worse in kind. It is not that studio software is expensive. It is that no two vendors are selling the same unit, so there is no arithmetic that puts two quotes side by side.

Read down the middle column and the problem announces itself. Consider one real studio — two hundred people on the books, of whom about a hundred and twenty book something in a given month, one room, four instructors. TeamUp bills that studio on 120, because its help documentation defines an active client as someone who "either register[s] for any event... or make[s] a purchase" in the billing month. Zen Planner and Gymdesk bill it on 200. Vagaro bills it on four. Pike13 and Punchpass bill it on nothing — the same flat rate at 40 members or 4,000. Mindbody ignores the members entirely and bills it on one, the location. Momence bills it on the money.

Those are not five prices for one product. They are five different questions being asked about the same business, and the answer that makes you cheap under one vendor is the answer that makes you expensive under the next. Nothing published in this category normalizes them, which means the tier comparison every studio owner starts with — three tabs open, three ladders side by side — is comparing quantities that have no common unit.

Mindbody deserves its own section, because it is the market leader and because what it does with its pricing page is the most interesting thing we found in this research.

Start with the page itself. It presents three plans — Starter, Accelerate and Ultimate — and explains its own meter honestly: the base plan is "Monthly subscription price (per location)," and you get unlimited users per location, with the price rising only when you add locations. It carries one anchor price, "starting at," and then nine separate "Let's talk" buttons where the three tier prices should be. There is no published rate for Starter, none for Accelerate, none for Ultimate.

Now the part that took some work. That URL is geo-served. Fetch it from the United States and the anchor reads $79 a month per location. Fetch it from Europe and the same URL returns €99, from the United Kingdom £89. We know this because our first pass at the page came back in euros and we nearly published that as a localization bug on the American site, which would have been wrong — it was our own connection. The correct finding is better than the mistake would have been. At current rates €99 is about $107 and £89 about $113, so European and British studios are quoted roughly a third above the American floor for the same software. There is also no /en-us/ version of the page; that path returns a 404. The United States gets the bare URL and geo-detection, and nothing else.

And then there is Australia, which is where this becomes properly strange. On the Australian path, Mindbody publishes all three prices: Go at A$89 a month per location, Core at A$279, Growth at A$369. Read the feature lists beside them and they are the Starter, Accelerate and Ultimate bullets — the same tiers, renamed, with the numbers left in. So an American studio owner who wants to know the shape of the ladder their own sales rep will not show them can go and read the Australian page, where the same company publishes it without apparent difficulty.

It has not always been this way, and the archive says so. Mindbody published its full US ladder for years: $129 / $259 / $399 / $599 in April 2020, rising to $169 / $299 / $539 / $759 by April 2024 — a 31% increase on the entry tier over four years, and 27% at the top. Then, sometime between June 14 and July 12 of 2024, the per-tier prices came off the US page and were replaced with the anchor-plus-"Let's talk" arrangement that is there now. The fourth tier, Ultimate Plus, disappeared from the page altogether, though the company's own contracts explainer still names it. We saw exactly this pattern last year when we went through legal software, where Clio publishes its complete ladder in the UK and Australia while showing American firms a single entry price and a lead form. It appears to be a category habit rather than a company one.

The predictable downstream effect is that the vacuum gets filled by people with an interest in what goes into it. We found the Mindbody US entry price quoted at $79, $99, $129 and $159 across four different sites, several of them selling competing software. One more thing worth knowing before you weigh any review of this product: Mindbody's terms of service contain a non-disparagement clause under which the customer agrees not to communicate, "in any medium (including social media, online reviews, or other digital platforms)," remarks that are "false, misleading, or disparaging" about the company. Reasonable people can disagree about how much that clause chills, but you should factor it in when you read the review sites — including when you read the silence.

To be fair in the other direction, several vendors here are exemplary. Punchpass publishes three flat tiers and states plainly that the subscription is the only fee it charges — card processing goes straight to Stripe at Stripe's rates and Punchpass takes nothing off the top. Gymdesk publishes a complete active-member ladder, includes unlimited staff at every tier, and carries the line "No contracts, hidden costs or cancellation fees." It is also, remarkably, at the same price today as in January 2022 — $75, $100, $150, $200, four years without an increase, which in this market is close to eccentric. Zen Planner publishes its full band table and specifies that it counts active members "never staff." Pike13 charges the same whatever your headcount. If parts of this article read as hostile to software vendors, those four are the reason it should not read that way about all of them.

To be fair in the other direction, several vendors here are exemplary. Punchpass publishes three flat tiers and states plainly that the subscription is the only fee it charges — card processing goes straight to Stripe at Stripe's rates and Punchpass takes nothing off the top. Gymdesk publishes a complete active-member ladder, includes unlimited staff accounts at every tier, and carries the line "No contracts, hidden costs or cancellation fees." It is also, remarkably, the same price today as it was in January 2022 — $75, $100, $150, $200, four years without an increase, which in this market is close to eccentric. Zen Planner publishes its full band table and specifies that it counts active members "never staff." If parts of this article read as hostile to software vendors, those three are the reason it should not read that way about all of them.

The cheapest plan is often the most expensive one

Momence is the most interesting pricing page in the category and the one worth doing arithmetic on, because it is the clearest example of a pattern that runs through this whole market: the subscription is not where the money is.

Momence publishes three tiers — Basic at no charge, Pro at $60 a month, and Custom at $199 a month. Alongside each tier it publishes a platform fee taken on payments: 5% on the business (plus 4% on the client) at Basic, 2.5% on the business at Pro, and no platform fee at all at Custom. Standard card processing rates are listed separately and are the same whichever tier you are on, so we can leave them out entirely and still compare the three fairly. Here is what those three plans actually cost a studio, by the volume of payments running through the platform.

The free plan stops being the cheapest option at $28,800 of annual volume — that is $720 of subscription divided by the 2.5 percentage points of difference. The $60 plan stops being the cheapest at $66,720. Above that, the most expensive-looking plan on the page is the least expensive one you can buy, and the gap widens with every class you sell. At $150,000 a year through the platform — a single-room studio with a few hundred members, not a large business — choosing the free plan over the $199 plan costs $5,112 a year. That is not a rounding error for a business with $23 million of payroll spread across 54 establishments. That is an instructor.

We want to be careful not to make this sound like a trap, because Momence publishes every one of these numbers plainly on its pricing page, which is more than most of its competitors do. The trap is not the pricing. The trap is the word "free," and the fact that a studio owner choosing a plan on a Tuesday evening is comparing $0 against $199 rather than comparing 5% against nothing. Percentage meters are hard for people to price, which is exactly why they are popular.

The general lesson generalizes past Momence: in this category, any fee expressed as a percentage of your revenue will eventually dwarf any fee expressed in dollars per month, because your revenue grows and the subscription does not. When you evaluate a platform, convert every percentage into the dollars it will take at your actual volume before you compare anything.

The number your software calls revenue, and Maryland calls a debt

Here is the part of this category nobody writes about, and it is the reason we wanted to write about studios at all.

When a studio sells a ten-class pack for $180, almost every product in this market does the same thing with it. It records $180, dates it today, and drops it into this morning's sales total. The owner opens the dashboard on the way home, sees a good day, and gets on with running the studio. Nothing about that is dishonest and nothing about it is unusual — it is what a point-of-sale system does, because a point-of-sale system is built to answer the question "what came in today."

But that $180 is not revenue. It is a promise to teach ten classes, and until those classes are taught it is money the studio is holding on behalf of somebody who has not yet received what they paid for. The accounting treatment is not ambiguous and not exotic: under ASC 606 the payment lands as a contract liability on the day of sale and is released into revenue class by class as the pack is redeemed. Any breakage — the classes the member is never going to take — is recognized in proportion to the redemption pattern as the rest of the pack is used, rather than in one lump when the pack expires. It is the same accounting that governs gift cards and airline miles, and it exists precisely because taking money up front makes a business look better than it is.

For most small businesses this stays an accounting nicety, something the bookkeeper reconciles at year end and the owner never thinks about. In Maryland it does not stay a nicety, because the state has written the same number into a bonding statute.

Section 14-12B-02 of the Commercial Law Article requires a seller of health club services agreements to file security with the Consumer Protection Division, in an amount not less than "the aggregate value of outstanding liabilities to members, including all prepaid fees, membership fees, dues, deposits, initiation fees, and fees for health club services." A surety bond is the usual route; an irrevocable letter of credit in a form the Division accepts, or plain cash, may be filed instead. There is a floor of $50,000 for a facility that has not opened yet, and a ceiling: no seller has to post more than $200,000 per facility.

Then the statute defines its own term, and this is the sentence that ought to be famous:

"'Liabilities' means the money actually received in advance from the members less the prorated value of services rendered by the health club facility."

Money received in advance, less the value of what you have delivered. That is deferred revenue. It is the ASC 606 contract liability, written into the Maryland Commercial Law Article and attached to a bonding requirement. The statute goes further and says how the figure must be produced: the bond amount "shall be based upon a report prepared by an independent certified public accountant describing the health club's outstanding liabilities to the members using accepted standard accounting principles." The Attorney General's Consumer Protection Division has a name for the resulting document — a Schedule of Outstanding Liabilities — and bonded registrants file one, CPA-certified, every year.

And it is not an annual exercise. The bond "shall be increased, or may be decreased," on whichever comes first of two triggers: when outstanding liabilities move by $10,000, or quarterly. If liabilities exceed the bond and the registrant has not topped it up, the statute is unambiguous about what happens next — the registrant "shall immediately stop selling health club services agreements."

Put those pieces together and you have something genuinely unusual. Maryland requires a studio to compute its deferred revenue balance at least four times a year, to have a CPA certify it, to re-post security every time it moves by ten thousand dollars, and to stop selling if it falls behind. Meanwhile the software that holds all the underlying data — every pack sold, every class redeemed, every membership day elapsed — mostly reports that money as revenue on the day it arrives.

So the obvious question is which platforms can actually produce the number. We went looking expecting the answer to be none of them. That is not what we found, and the real answer is more interesting.

Credit where it is due: the expensive incumbent does the hard thing

Mindbody — the vendor we have just spent several paragraphs criticizing for its pricing page — publishes an Earned Revenue report, and it does the calculation properly.

For limited pricing options, the ten-class cards and five-class packs, it computes unearned revenue from the number of visits remaining, which is the correct treatment. For unlimited memberships, where there are no visits to count, it prorates: the amount paid, divided by the duration of the membership, multiplied by the days remaining. It carries a separate section for expired series so you can see the packs that ran out with sessions unused, which is where breakage lives. And its own documentation is honest about the weakness in the method, noting that for unlimited options the value may change dramatically depending on when the report is generated — which is true, and is the kind of caveat a vendor only writes when it expects the number to be used for something.

That is a genuinely good piece of software, and it is not what we expected to be writing. The most costly and least transparent platform in the category is the one that gives a Baltimore studio the closest thing to the figure Maryland's bonding statute asks for. If you are on Mindbody and you have never opened that report, open it before you do anything else in this article — you may already own the answer.

There is one catch, and it is a real one: the Earned Revenue report is, by Mindbody's own documentation, "only available with certain software packages." So the number Maryland requires a bonded studio to certify every quarter sits behind a tier gate, on the product that will not tell you what its tiers cost.

WellnessLiving deserves nearly equal credit and produces the strangest artifact we found anywhere in this research. Its Visits Remaining Report has columns literally headed Deferred Revenue — "the value of visits that haven't been booked or attended with this Purchase Option" — and Unearned Revenue. That is the right calculation with the right label, which is more than most accounting software manages. Two things then happen to it. It is filed under Attendance rather than Sales, where no bookkeeper would think to look. And it carries this note: "For data security purposes, the Visits Remaining Report doesn't have an export option. If you need to export this report, please contact our Support Team." Every other report in the product exports freely. The one carrying the liability figure does not. Meanwhile WellnessLiving's own FAQ concedes the underlying problem in a sentence worth reading twice: when a client buys a purchase option, "the sale appears on reports filtered for accrual-based accounting with its full revenue on the day it was sold."

Below those two it thins out quickly. Walla genuinely amortizes prepaid time-based memberships across the period they cover — but its own documentation marks Credit Bundles, which are its class packs, as not amortized, so the exact instrument this article is about is carved out. Mariana Tek treats credits as deferred until used, expired or forfeited, which is real revenue-recognition thinking. Punchpass is admirably honest that it cannot do it: "until a pass is actually full and/or expired we can only estimate the per-visit revenue." Zen Planner has no such report at all — a full-text search of its help documentation returns nothing for deferred, unearned, accrual or revenue recognition, and the closest available field returns remaining class counts rather than dollars, for one membership type only. Glofox has nothing either, which is the sharpest irony in the set: its parent company ABC Fitness ships a "PIF Cash Liability" report in its other products, computing value remaining as the down payment divided by term length times days remaining. The company knows exactly how to build this. Glofox customers do not get it.

If you are evaluating a platform and this matters to you — in Maryland it is not optional — the question to ask a salesperson is not "do you have reporting." It is: show me the report that gives the dollar value of classes I have sold and not yet taught, as of today; show me how it handles an unlimited membership; and show me the export button.

The reason we care so much about this one report is not really compliance, incidentally. It is that the same number is the best retention instrument a studio has, and almost nobody uses it that way. A member holding seven unused classes with nothing on the calendar is not a line on your balance sheet. That is somebody who is drifting away three weeks before they will admit it to themselves, and you have a list of them.

What Baltimore studios actually run on

We wanted to know what this market looks like on the ground rather than in a vendor's case study, so we checked. The method was crude but reproducible: fetch each studio's public pages, read the raw HTML, and look for the signature of a booking platform — a widget script, a tenant name, a members subdomain. It only catches platforms that leave a trace in server-rendered markup, so treat it as a floor rather than a census, and we would not publish a market share off it. What it does show is a pattern strong enough to survive the method's limits.

The Baltimore market is segmented by discipline, not by size. At the boutique end — reformer pilates, barre, spin, yoga — Mariana Tek turns up more than anything else, with Momence taking the small independents and Mindbody rather less visible than its reputation suggests. Two studios were caught mid-migration, which is the sort of detail you only get by reading the markup: one still carries a stale meta description telling clients to book through Mindbody while its live widget points at Momence, and another runs Mariana Tek while still loading Mindbody's widget library.

Then go and look at the CrossFit boxes and the martial arts schools, and Mindbody essentially vanishes. Baltimore City has seven CrossFit affiliates by the registry's own API — and across the functional-fitness gyms we checked, the platforms were PushPress, Wodify, Zen Planner and Kilo, with no Mindbody anywhere. Martial arts was starker still. Of seventeen schools checked, the software found was Gymdesk, MyStudio, Spark and ABC — and nine of the seventeen ran no booking software at all. A contact form on a Squarespace site, and a person who knows everyone's name.

That last group is worth sitting with, because it is a useful corrective to an article like this one. A meaningful share of this market has not been captured by any platform, is not paying a per-member meter to anybody, and is doing fine. Software is not oxygen. The businesses that need what we build are the ones that have outgrown the contact form and then outgrown the platform they bought to replace it — and that is a narrower set than any vendor, ourselves included, would like you to believe.

One more local note, since it bears on everything above. Merritt Clubs, the largest homegrown operator in the region at eight locations, publishes no membership price anywhere on its site; we checked five separate pages. Brick Bodies, its nearest local peer, publishes a complete rate table down to the bi-weekly billing amount and the annual enhancement fee. Same city, same category, opposite instincts — and if you have been reading this far you can probably guess which one we think is playing the stronger hand.

What else Maryland puts in a membership agreement

The bond is the part that surprises people, but it is not the only place where this statute and your software disagree about how a studio should work.

Start with the threshold question, because it decides whether any of this applies to you, and because the version of it circulating on the internet is wrong.

Maryland does not regulate studios; it regulates agreements. Section 14-12B-01(d)(1) defines a "health club services agreement" as one under which the buyer purchases services "to be rendered over a period longer than 3 months," and the seller "collects more than 3 months' payment in advance," and the service is for personal, family or household use. That conjunction matters enormously and it is routinely misread — several secondary summaries state the test disjunctively, as though either limb alone were enough. It is not. All three elements have to be present. The consequence is that a membership sold on a rolling monthly basis and billed monthly is not a health club services agreement at all, however many years the member stays, because nothing is collected more than three months ahead. The subtitle names the businesses it has in mind — a health spa, a figure salon, a weight reduction center, a self defense school, "or other physical culture service enterprises" — and carves out nonprofits, schools and government bodies. A martial arts school is explicitly in scope.

So the line falls on prepayment, not on longevity, and that is the interesting part. The studio selling month-to-month is outside. The same studio, on the day it offers twelve months paid up front at a discount — which every retention consultant and every platform's own best-practice guide will tell you to do, because prepayment is the most reliable predictor of whether a member stays — has crossed into a different legal regime. The trigger is not the size of the business, the number of members, or the revenue. It is the term and the prepayment window, both of which are fields on a product setup screen. The most consequential compliance decision a Baltimore studio makes is made in a dropdown, by whoever was building out the pricing options that afternoon.

And there is a trapdoor underneath even the month-to-month studio. Section 14-12B-02(i) provides that a fee over $200 which is not identified as payment for specific future services is deemed to be payment for services to be delivered during the initial two years. Read that against the three-month test and the effect is clean: a studio charging a $250 joining fee on an otherwise monthly membership has, by operation of that deeming provision, collected far more than three months in advance. A number typed into an initiation-fee box has moved the business into the bonded tier. Nothing in any software we looked at mentions this, and there is no reason it would.

Registration and bonding are also not the same trigger, which the statute is loose about. The registration duty in § 14-12B-02(a) falls on "each person who sells health club services" — a phrase the subtitle never defines, and which is plainly broader than the defined term. The bond attaches to sellers of health club services agreements. That the two differ is confirmed by the fee schedule itself, which includes a $75 tier for a registrant who does not collect any payment in advance — a tier that would be meaningless if only prepaying businesses had to register. So: a pay-as-you-go studio still registers. It simply does not bond.

Registration goes to the Consumer Protection Division before you sell, and renews every year on September 1. The Division runs a dedicated Health Club Unit for exactly this, out of 200 St. Paul Place, which is a short walk from a good number of the studios this article is about. The published fees are $1,200 for a bonded business, $300 for a bond-exempt one, and $75 for pay-per-day. The security described above attaches if you take more than three months in advance, trip the $200 deeming provision, or collect any money before the doors open.

Then the agreement itself changes shape. Section 14-12B-06 says a health club services agreement "may not contain an automatic renewal clause, unless the agreement provides for a renewal option for continued membership which must be accepted by the buyer." Sit with that for a second in the context of software whose central abstraction is a recurring subscription that runs until somebody cancels it. The statute is not banning recurring billing, and an ongoing month-to-month membership is not the target. What it forbids is the specific pattern where a fixed-term agreement silently becomes another fixed term without the member agreeing to it — which is exactly the auto-renew checkbox sitting in the contract setup screen of every platform we looked at, because in most states that checkbox is perfectly lawful.

The buyer also gets three business days from receiving a copy of the agreement to cancel it, in writing, delivered in person or by certified mail with return receipt requested, and if it is mailed it only has to be postmarked by midnight of the third business day. On a valid cancellation the studio refunds everything — the statute lists "any deposit, down payment, or payment on the agreement including any initiation, deposit, membership, or other fees." Not a pro-rated share. All of it.

And the agreement has to carry a section headed "Notice of Consumer Rights" that discloses the club's registration number, its bonding status and the bond amount or the basis on which it is exempt, the buyer's cancellation rights, and the buyer's rights if the facility temporarily closes or the member becomes disabled. Some registered facilities additionally have to post a conspicuous sign stating that they do not charge advance fees or collect payment before services are provided.

There is also a record-keeping duty most operators have never read. Section 14-12B-02(g) requires a registrant to maintain accurate, updated records of the name, address, contract terms and payments of every buyer, available for inspection by the Division — which is, if you look at it squarely, a database schema written into a consumer protection statute. And § 14-12B-07 says that on closure a studio must produce, within fifteen business days, a member list with addresses, copies of all membership agreements, and a record of all payments. Hold that thought; it becomes relevant in a moment.

The newest law, seven weeks old

Something changed on June 1, 2026 that nobody in this category has written about yet, and if you sell memberships online in Maryland it applies to you.

Maryland's general automatic renewal statute, enacted as Chapters 204 and 205 of 2025 and codified at Commercial Law § 14-1329, took effect at the start of June. It is a different instrument from the health club rule above and it stacks on top of it. Where § 14-12B-06 governs only health club services agreements, this one reaches ordinary consumer subscriptions — and the legislature struck the draft language that would have limited it to terms longer than a month, so a plain month-to-month membership is inside it. The core requirement is cancellation parity: the mechanism to cancel must be at least as easy to use as the one the consumer used to sign up, and available through the same medium. If somebody joined in three taps on your website, they must be able to leave in about three taps on your website. Terms of a year or more also need a renewal reminder in a defined window ahead of the renewal date. Health clubs get no exemption.

It is worth understanding why this now matters more than it would have two years ago. The Federal Trade Commission finalized a national "click to cancel" rule in November 2024 covering exactly this ground — and the Eighth Circuit vacated it in its entirety on July 8, 2025, in Custom Communications, Inc. v. FTC, on the grounds that the Commission had skipped a required preliminary regulatory analysis. The rule was to have taken effect six days later. A new federal rulemaking has since started over. The federal backstops survive — ROSCA still governs anyone selling a subscription online, and Section 5 still prohibits unfair or deceptive practices — but the detailed, prescriptive rulebook is now state law, and in Maryland it is seven weeks old. It is worth noting that a good deal of the compliance content published by studio software vendors still describes the vacated federal rule as though it were in force, with its 2025 deadlines presented as live.

Leaving, and the thing you cannot take with you

Which brings us back to § 14-12B-07 and the fifteen business days.

Every discussion of switching platforms in this category treats data export as an inconvenience. In Maryland, for a bonded studio, producing a member list with addresses, every membership agreement and a complete payment record is a statutory obligation with a deadline attached. That reframes the question. It is not "how annoying will it be to leave." It is "can I produce, on demand, the three artifacts the state will ask me for."

Most of the platforms will give you the client list and the payment history. The place it gets interesting is the card vault, and here the terms are unusually explicit. Mindbody's terms of service commit it, for thirty days after a termination notice, to make your data available — "except Cardholder Data." Its support documentation is blunter still: an export of billing information is possible only if you are cancelling the subscription outright, not if you merely want to move to a different payment processor while staying a customer. The export is a paid service, it requires the account owner personally with identity verification, it needs a PGP key from your incoming processor, and it must be requested inside a narrow window after cancellation. Mindbody does not publish what it costs; you are told to ask the agent handling your cancellation.

We want to be precise about the reason, because the industry is not. A widespread claim — including from vendors — is that card data cannot move because PCI forbids it. That is not true. PCI DSS defines a compliant path, and every major processor documents it: the old processor de-tokenizes to real card numbers, encrypts the file to the receiving processor's published PGP key, and ships it directly. Stripe, Adyen, Braintree and Square all do this, mostly free, typically in ten business days. What they will not do is hand that file to you, because a studio holding raw card numbers would pull itself into the heaviest tier of PCI validation overnight. So the transfer is routine between two audited processors and impossible for a merchant to perform alone — which means when it does not happen, that is a vendor decision, not a regulatory one. Some vendors are straightforwardly on the right side of this: TeamUp, Punchpass and Gymdesk have you hold your own processor account, so the vault was never theirs to withhold.

None of which is an argument for building your own booking platform. It is an argument for knowing, before you sign, which of your assets you are renting and which you own — and for keeping the customer relationships that matter most to you somewhere you control.

We want to be careful and fair here, because it would be easy to turn this into a complaint about vendors. Every platform we looked at lets you paste your own legal text into the agreement attached to a contract. Nothing stops a Baltimore studio from putting a correct Notice of Consumer Rights in that box. The problem is narrower and more practical than "the software will not let you." It is that the field is empty by default, the registration number and the bond amount are two values the software does not hold and cannot fill in, the bond amount is derived from a liability figure most of these products do not compute, and there is no point in the flow — not when you create a twelve-month membership, not when you enable auto-renew, not when you set an initiation fee at $250 — where anything tells a Maryland business that it has just changed its legal position. A national platform serving fifty states with fifty different health club regimes is never going to build that, and it would be slightly unreasonable to expect it to. That is precisely why the layer belongs to you.

And now the software itself is taxed

One more line that moved recently, and that almost nobody has repriced for.

Maryland's Budget Reconciliation and Financing Act of 2025, House Bill 352, was signed on May 20, 2025 and took effect on July 1 of that year. It applies a 3% sales and use tax to data and information technology services — the NAICS 518, 519 and 5415 families — along with system and application software publishing under 5132. The rate is deliberately set below the general 6% and applies only where a higher rate does not already. The practically important part for a studio is that business-to-business software, which had been sitting outside Maryland's sales tax base, moved inside it. Your studio management subscription now carries 3% that it did not carry in June 2025, and so does a good deal of the rest of the stack — the scheduling add-on, the email tool, the analytics.

On a $400 monthly platform bill that is $144 a year, which is not going to change anybody's life. It is worth stating anyway, for two reasons. The first is that it lands on every tool separately and compounds quietly with each one you add, which is a real argument for running fewer of them. The second is that we should be straight about which side of this we are on: NAICS 5415 includes 541511, custom computer programming services, so the 3% applies to our invoices too. We are not selling you a way out of a tax. We are telling you it exists because you should be able to see the whole bill, including ours.

What custom actually costs

We publish our prices, which in this market feels like a small act of provocation. Here is what the four things we sell cost when they are pointed at a studio.

Each of those is a fixed price against a fixed timeline, half up front and the balance when it ships, and you own the code, the keys and the accounts at the end. The full detail of what is and is not included lives on the pricing page. For a studio, the first purchase is almost never the Operations System, and we will usually talk you out of it on the call. It is either the Prototype Sprint, because a week of work makes the argument better than a proposal does, or it is the Online Store — the same engine as the store we built in a weekend, pointed at a different job.

That second one deserves a paragraph, because it is where the money is and it is the thing studios are least likely to think of as software. Selling class packs, memberships, workshops and teacher training on your own domain through a processor you chose means the platform percentage on that revenue goes to zero — go back and look at what the 5% column did in the table above — and the card rate becomes whatever your processor quotes you rather than whatever your software vendor marked it up to.

A word of caution on the merchandise end of that, because we would rather correct an expectation than sell into it. There is a genre of advice telling studios that retail is an underexploited revenue stream worth fifteen percent of the top line. It is not. Franchise disclosure documents are the one place in this industry where studios are legally obliged to publish their revenue mix, and across more than two thousand studios in six national brands, products and merchandise run between two and nine percent of revenue, clustering around four. Memberships are north of eighty percent. Barre studios sit at the top of the retail range for the unromantic reason that grip socks are effectively compulsory. So build the store for the memberships and the packages, which is where the volume and the platform percentage both are, and treat the merchandise as a pleasant sideline rather than a strategy.

The store also means the customer record, the email address and the purchase history sit in a database you can query, which matters more than it sounds like it does. The single most valuable marketing asset a studio has is the list of people who used to come and stopped, and that list is remarkably hard to get out of most of these platforms in a shape you can use.

What we would actually build, and the case for not replacing anything

Now the part that runs against our own commercial interest, which is the part worth reading.

Do not replace your booking platform. For a studio with one location and a few hundred active members, the scheduling engine, the mobile app, the waitlist logic, the instructor substitution flow and the payment plumbing are collectively worth far more than what you are paying for them, and no fixed-price build competes with that on economics. A class schedule is a genuinely hard piece of software with twenty years of edge cases baked into it, and rebuilding it is the single most reliable way we know for a studio to spend $30,000 and end up with something worse. If somebody tells you otherwise, look closely at whose product the break-even calculation was written to sell.

What pays is the layer around it, because that layer replaces hours and judgment rather than replacing features.

The first build, nearly always, is the package liability ledger — the thing that turns the money you have collected into the two numbers you actually need. What have I sold and not yet delivered, and what is it worth today? That single object gives you the deferred revenue figure your accountant wants at year end, the outstanding-liability figure Maryland sizes your bond on, and the retention signal described earlier, all from the same data. If you are on a platform that already computes it, this build is an integration rather than a system, which makes it cheaper.

Second is the agreement layer. One template, generated rather than pasted, that puts the right Notice of Consumer Rights on every Maryland agreement with your live registration number and current bond amount in it, refuses to create a twelve-month auto-renewing contract without the acceptance step the statute requires, records the date the member received their copy so the three-business-day clock is evidenced rather than assumed, and files the signed copy somewhere you can find it in four years. This is unglamorous and it is the highest-return software a regulated small business can own, because its value is not efficiency. Its value is that the thing you were going to get wrong at some point now cannot go wrong.

Third is whatever your studio does that the platform has no concept of. This is different at every studio, which is precisely why it is never in the product. Teacher training cohorts with their own payment plans and hour tracking. Reformer equipment with per-piece capacity that is not the same as room capacity. A recovery suite booked in overlapping fifteen-minute slots. Instructor pay that is base plus head count above a threshold, with a different threshold for the 6am. Multi-site memberships where the home studio takes a share. Every one of those currently lives in a spreadsheet that one person maintains, and that person is the actual risk.

Fourth is the store, for the reasons above. And around all of it, the integrations that keep it honest: the booking platform's API so the schedule and the attendance stay in one place, your accounting system, your processor, and exports a bookkeeper can open without calling you.

When you should not call us

Being specific about this is more useful than another paragraph about our process. Do not build if any of the following is true:

• You have one location, under roughly 150 active members, and your real problem is that not enough people are walking in. Software is not the constraint, and a build will absorb attention you need elsewhere.
• You have not yet outgrown your platform's own tools. A surprising number of the reports studios ask us for already exist under a menu they have never opened — the Mindbody report above is a good example — and we would rather tell you that on a free call than sell you something.
• Nobody on your team owns operations. Custom software needs somebody who decides how the studio actually works; without that, you get an expensive rendering of a disagreement.
• You are inside twelve months of a planned sale. A buyer wants clean books and boring systems, not a bespoke stack they will have to take on.
• What you want is a mobile app so members can book on their phone. Your platform already does this, it does it well, and paying to rebuild it is the clearest waste of money in this category.

The honest summary is that a studio under about 150 members should keep everything and change nothing; a studio between roughly 150 and 600 members should keep the booking platform and own the layer above it; and only a multi-site operator with genuinely unusual programming should be having a conversation about replacing anything. We will tell you which of those you are on the call, including when the answer is the first one.

How we work

We are a two-person studio in Baltimore. You talk to the people writing the code, every time, and there is no account manager in the middle translating what you said into a ticket.

The first call is thirty minutes and free, and the useful version of it is not a demo. Bring the spreadsheet — there is always a spreadsheet — and the report you rebuild by hand every month, and the thing your platform will not do that you have stopped complaining about because you assumed it was impossible. We will tell you what we would build, what you should keep paying for, roughly how fast it ships and what it costs. If the answer is that you should change nothing, we will say that, and it costs you half an hour.

We use AI heavily in the build, which is why a fixed price of $12,000 buys what an agency quotes $50,000 for and takes two to four weeks rather than two to four months. We are specific about how in the piece on our build stack, because "AI-accelerated" is a claim worth being suspicious of when a vendor makes it. What it does not change is that somebody experienced still has to decide what the software should do, and that is the part we are actually selling.

Everything ships deployed, on your infrastructure, with the source in a repository you own. Half the price to start, half when it goes live, and the number does not move unless you ask for something that was not in the scope we agreed. We wrote about what custom software really costs and why hourly billing produces the numbers it does, if you want the longer argument.

Questions we get from studio owners

How much does custom fitness studio software cost in Baltimore?

Our prices are fixed and published. A one-week Prototype Sprint — one core flow, such as a package liability ledger or a Maryland-compliant agreement generator, clickable and deployed — is $3,500. A branded booking and store front on your own domain and your own payment processor starts at $6,000. A custom internal tool, or a full operations system covering schedule, memberships, instructor pay, liability and the compliance calendar, starts at $12,000. Most studio builds land between $6,000 and $25,000. The price is agreed before any code is written, and you own the code, the keys and the accounts at the end. Note that Maryland's 3% tax on information technology services applies to custom development invoices as well as to your software subscriptions.

Does a yoga or pilates studio count as a "health club" under Maryland law?

Maryland does not regulate studios by type; it regulates agreements. Under Commercial Law § 14-12B-01, you are selling a "health club services agreement" if the member is buying services over a period longer than three months, or if you collect more than three months' payment in advance. The subtitle names health spas, figure salons, weight reduction centers and self defense schools among the businesses it covers, and exempts nonprofits, schools and government bodies. So a studio selling month-to-month memberships and small class packs is likely outside it, and the same studio is inside it on the day it starts selling annual memberships or six-month prepaid packages. This is a question for your own attorney rather than for a blog post — but it is worth knowing that the trigger is a setting in your booking software, not the size of your business.

Does my Baltimore studio need to register with the state, and does it need a bond?

If you sell health club services agreements as Maryland defines them, you register with the Consumer Protection Division before you sell, and renew every September 1. The Division's published registration fees are $1,200 for a bonded business, $300 for a bond-exempt business and $75 for a pay-per-day operation. Security — a surety bond, an irrevocable letter of credit acceptable to the Division, or cash — is required if you take more than three months of payment in advance, charge an initiation or other up-front fee above $200, or collect money before the facility opens. Section 14-12B-02 sizes that security on the aggregate value of outstanding liabilities to members, including prepaid fees, dues, deposits and initiation fees. There is a $50,000 floor for a facility that has not opened and a $200,000 per-facility ceiling.

Can I auto-renew memberships in Maryland?

Not in the way most software does it by default. Section 14-12B-06 says a health club services agreement may not contain an automatic renewal clause unless the agreement provides a renewal option that the buyer must accept. That is aimed at the pattern where a fixed-term contract silently rolls into another fixed term, which is a standard configuration option in every platform we looked at because it is lawful in most states. Ongoing month-to-month billing is a different thing and is not the target. The same section gives the buyer three business days from receiving their copy of the agreement to cancel in writing, with a full refund of everything paid including initiation fees, and requires the agreement to carry a "Notice of Consumer Rights" section disclosing your registration number, your bonding status and amount, and those cancellation rights.

Should I replace Mindbody or WellnessLiving with a custom system?

Almost certainly not, and we will say so on the call. The scheduling engine, waitlist logic, instructor substitution, mobile app and payment plumbing in a mature booking platform represent twenty years of accumulated edge cases, and for a single-location studio they are worth considerably more than they cost. Rebuilding that is the most reliable way we know for a studio to spend $30,000 and end up with something worse. What is worth owning is the layer the platform was never designed to hold: package and membership liability, a compliant agreement generator, your unusual programming, and a store on your own processor. That is a much smaller piece of software and a much larger saving.

Why does unredeemed package liability matter if the cash is already in the bank?

Three reasons, in increasing order of how much they will cost you. First, it is the correct accounting — under ASC 606 a prepaid pack is a contract liability released into revenue as classes are redeemed, not revenue on the day of sale, and a studio that spends against the gross number is spending money it still owes in classes. Second, in Maryland it is the figure your required bond is sized on, so it is not merely an accounting preference. Third, and most usefully, it is the best retention signal a studio has: a member sitting on seven unused classes with nothing booked is a member you are about to lose, and almost nothing in the standard reporting surfaces that in time to do anything about it.

Do I pay Maryland sales tax on my studio software?

Yes, as of July 1, 2025. House Bill 352, the Budget Reconciliation and Financing Act of 2025, applies a 3% sales and use tax to data and information technology services in NAICS 518, 519 and 5415, and to software publishing in 5132. Business-to-business software that previously sat outside Maryland's sales tax base is now inside it at 3%. This applies across the stack — your booking platform, your email tool, your analytics — and, to be straight about it, to our development invoices too. A multiple points of use certificate is available where a purchase is used both inside and outside Maryland.

How long does a build take, and who owns it at the end?

A Prototype Sprint is about a week. A store or booking portal is one to two weeks. A custom tool or a full operations system is two to five weeks. Those are calendar timelines rather than estimates that drift, because the price is fixed and the scope is agreed before we start — that is the whole point of working this way, and we wrote about why we do it separately. You own everything at the end: the source code, the repository, the hosting account, the domain, the payment processor account and the database. We do not hold the keys, and there is no license that expires. If you want to hire somebody else to extend it next year, that is a normal thing to do and we will hand it over cleanly.

Figures in this article were verified in July 2026 from the vendors' own published pricing pages, the Maryland Code, and the Census Bureau's County Business Patterns for reference year 2023. Vendor pricing changes without notice, and nothing here is legal or tax advice — the Maryland provisions described are summarized, not reproduced in full, and your own attorney should tell you how they apply to your business.