Med Spa Membership Revenue: What You Can and Can't Count

A membership and a prepaid package feel like the same sale. Money arrives, the client commits, the month looks good. But they are two different revenue models, and most med spas run them through one line called "memberships." One is recognized evenly whether she ever walks in; the other is drawn down as she redeems, and leaves behind sessions she'll never use — money you eventually keep, but not on the date you'd assume. Getting these two apart is the difference between recurring revenue you can trust and a number that flatters you until the packages come due.

Two products wearing one name

Walk into most med spas and you'll find both of these on the price list:

• An unlimited or access membership — a monthly fee that buys member pricing and unlimited use of something. You are selling availability.
• A prepaid package — six laser sessions for $3,600, nonrefundable, banked and redeemed over time. You are selling a countable number of treatments.

Commercially they're cousins. Accounting-wise they are not related at all, and the distinction isn't a technicality — it changes what revenue looks like every single month, and whether the phrase "unused sessions" even means anything.

First: the money isn't revenue yet

Before either model, the same rule applies to both. When a client prepays, you have cash and an obligation — not revenue. The standard is unusually blunt about it (ASC 606-10-55-46): on receipt of a prepayment an entity recognizes a contract liability "for its performance obligation to transfer, or to stand ready to transfer, goods or services in the future," and derecognizes it — recognizing revenue — when it transfers those goods or services.

So a $3,600 six-session package sold this morning is $3,600 of obligation and $0 of revenue. It's the mechanism behind the most common way a med spa misreads itself: a strong selling month gets read as a strong earning month, margins look better than they are, and the practice spends against cash it still owes in treatments.

Model one: the unlimited membership recognizes evenly — used or not

For an access membership, ASC 606 has an illustration that could have been written for you. Example 18 concerns a health club selling one year of unlimited access at $100 a month. The standard's reasoning: the promise is "a service of making the health clubs available for the customer to use as and when the customer wishes," because how much she uses it "does not affect the amount of the remaining goods and services to which the customer is entitled."

And then the line that matters most for a med spa (606-10-55-186): the customer benefits from availability "regardless of whether the customer uses it or not" — so revenue is recognized straight-line, $100 a month.

One consequence owners find counterintuitive: a genuinely unlimited membership generally has no breakage. KPMG makes the point in passing, while distinguishing take-or-pay contracts from stand-ready ones — where an entity provides a stand-ready service over distinct time periods, the customer's right "does not expire unexercised; therefore, a modification or breakage approach would not apply." Keep the condition attached to it, though: that reasoning holds where "the number of distinct goods or services does not change." Add a countable included treatment that banks, and the condition is no longer met. If a report shows breakage on your unlimited tier, the likeliest explanation is that the tier is mapped as a package.

Model two: the package draws down — and leaves breakage

Now the six-session package. Here there are countable units, they can be forfeited, and the standard gives them a name. ASC 606-10-55-47: a client's nonrefundable prepayment gives her a right to a future service, "however, customers may not exercise all of their contractual rights. Those unexercised rights are often referred to as breakage."

Breakage is genuine economics — it's the part of the package that becomes margin without costing you product or chair time. The trap is when you're allowed to count it, and both instincts are wrong.

It can't be booked at the sale

Even with years of history showing that a fixed share of sessions never gets redeemed, that slice can't be recognized up front. EY's guidance is direct: entities cannot recognize estimated breakage immediately on receipt of prepayment — and that holds "even if an entity has historical evidence to support that no further performance will be required" for some portion of contracts. (EY and KPMG are interpretive handbooks, not the codified standard — auditors follow them, but the authority is Topic 606 itself.)

And you can't elect to wait for the expiry date

The other instinct — adopt a policy of leaving it alone and booking the balance at expiry — isn't available as a policy. Asked whether an entity may recognize breakage only at redemption, expiration, or when redemption becomes remote, KPMG answers no: the estimate has to be attempted first. That is not the same as saying expiry is never the answer. Where there's genuinely no basis to estimate, the standard routes you to recognition when exercise becomes remote — and KPMG notes that point may well occur on expiration. What's prohibited is skipping the estimate, not landing on expiry after making one.

What the standard actually requires

ASC 606-10-55-48 sets two paths, and which one applies turns on whether you expect to be entitled to the breakage — a test the standard runs through the constraint on variable consideration:

In plain terms: if you expect that one session in six goes unused, you recognize that sliver gradually as she redeems the other five — not in a lump at the end. If you don't expect to be entitled to it, you recognize it only when the likelihood of her using the remaining rights "becomes remote."

The math the rule actually produces

Here's the arithmetic on a $3,600 package of six sessions, where experience suggests five get redeemed and one doesn't. The denominator is expected redemptions, not sessions sold — that's the mechanism that pulls breakage into revenue proportionally rather than in a lump:

Each redemption carries $600 of service plus a $120 slice of the session you expect she'll never take. If she surprises you and books a sixth, you revise the estimate and true up in the current period rather than restating. This is illustrative arithmetic to show the mechanic — it assumes you have a defensible basis for the one-in-six, which is the whole question.

If your practice is new, you may not have that basis. The standard routes this through the constraint (606-10-32-12(c)), which counts against you where "the entity's experience (or other evidence) with similar types of contracts is limited, or that experience (or other evidence) has limited predictive value." Note or other evidence — your own history isn't the only permissible input, which is why this is a judgement rather than a rule about your age. KPMG works one case where a new program with neither entity-specific data nor knowledge of comparable market experience is fully constrained, with a cumulative catch-up once real redemption history exists. Whether yours is that case depends on facts you and your CPA have and I don't. And it isn't all-or-nothing: 606-10-32-11 allows "some or all," so a smaller estimate you can stand behind is recognized proportionally rather than deferred wholesale.

So how much breakage should you expect?

There is no benchmark to give you. Published breakage rates exist almost entirely for retail gift cards — restaurant and retail chains that disclose them in SEC filings. I didn't survey them systematically, but the ones I looked at were far apart from each other, which alone should discourage picking a number off the shelf. And they don't transfer: a gift card is stored value, while a prepaid treatment package is an obligation to deliver a specific service. Reported breakage is also a revised estimate, disclosed net of amounts remitted to states — not a count of what went unused.

The closest published analogue I could find to what a med spa actually sells — a gym operator recognizing breakage on prepaid personal-training sessions — discloses the policy and the method, and no dollar figure at all. So there is no benchmark here to borrow. There is only your own redemption curve, which is a thing you can measure and nobody else can hand you.

Why prepaid sessions go unused

Breakage gets treated as luck. There is measurement suggesting it's partly structural — bundling itself suppresses redemption — though less of that research transfers to a med spa than I'd like, so let me be precise about which part does.

The closer analogue is theatre, because the shape matches: one payment, several countable admissions, each of which can be forfeited. Across 6,070 tickets at a 1997 Shakespeare festival, a ticket bought on its own was used 99.4% of the time; one bought inside a four-play bundle, 78.5%. That raw gap flatters the effect, though — a bundle buyer's third and fourth plays also carry plain fatigue with the series. The authors ran a stricter test comparing only each buyer's first play, and the gap narrows to 99.4% versus 84.2%. Bundle tickets weren't priced identically either, and price was the strongest predictor in their model. So: a real effect, smaller than the headline, pointing the direction you'd expect — the more someone buys at once, the less likely any single unit gets used.

The health club study usually cited alongside this one is about something different, and conflating them would undercut everything above. It measured when members consumed, not whether they forfeited. Members paying in two semiannual installments concentrated their visits around the payment: about 35% of a member's six-month attendance fell in the month he paid, decaying to roughly 6% five months later, with the pattern resetting on the next payment cycle. But that facility sold unlimited access — the article's Model One. Nothing there was countable and nothing was forfeited. It tells you the felt cost of a visit fades as the payment recedes; it does not tell you sessions get lost.

Scope, before the numbers: the gym figures are 33 screened members at one facility, 1997 data — screened meaning they had to have belonged for over a year and attended at least twelve times, which filters out precisely the disengaged client a med spa worries about; 16 of the 33 also paid in January. Those percentages are shares of each member's own six-month window, so they sum to 100% by construction and can't be read as an absolute drop in visits. Neither study tested a service package, a four-figure price, or a practice that sends reminders — and the authors themselves flag price and involvement as untested. Treat all of this as a hypothesis worth checking in your own data, not a rate to plan against.

The operational reading still holds, and it's the useful one. Breakage is margin, but it's margin made of clients drifting away — and a drifting client doesn't renew. If you want the recurring revenue without the churn, the lever is getting people to use what they bought, which makes the reminder cadence after a package sale a revenue-retention activity rather than admin.

"Members spend 3× more" — where that number comes from

Every membership pitch leans on some version of this. I went looking for the source of the 3× figure specifically and couldn't trace one: no dataset, no sample, no scope — it circulates between vendor and consultant pages without an origin. So treat the multiple itself as unsourced. In most practices' own reporting some gap does show up; I have no cross-practice dataset telling you how big it usually is, and I'd distrust anyone who quotes one.

The deeper problem is that even your own real gap doesn't measure what the program did. Clients choose to join, and the ones who choose are disproportionately the people already coming often and spending well — that's precisely why a membership appeals to them. Comparing members to non-members compares two groups that differed before the program existed. Some of the gap is the program working; some is sorting that already happened. The comparison can't separate them, and it will always flatter the program.

A better version is to watch whether the same clients changed after joining — her visit frequency against her own prior twelve months. That removes the fixed differences between joiners and non-joiners. It does not remove everything, and the article would be doing exactly what it criticizes if it stopped there:

• People join at a peak. Signing up is usually triggered by a burst of engagement, so the prior-12-month baseline sits above her long-run average and some of the fall-off afterwards is just reversion, not the program failing.
• If the plan includes a monthly treatment, part of any rise is arithmetic. A product that hands her visits raises visit count by definition.
• No control group means every trend gets credited to the membership — a new injector, a marketing push, a good year.
• Only clients with twelve months of history have a "before," so the measure quietly excludes newer joiners.

Strengthen it by running the same before-and-after over the same calendar window for comparable clients who didn't join, and by counting only visits the membership doesn't hand her automatically. That still isn't proof of causation — but it's honest about which direction it's wrong in.

What to actually track

If memberships are more than a rounding error in your practice, these are the numbers worth separating:

Two of these connect to numbers you're likely already watching: unredeemed sessions are a leading indicator of retention falling, and a member who stops booking shows up in your no-show and cancellation pattern before she formally cancels. For where memberships sit inside the wider financial picture, the med spa business plan numbers put them next to margin and break-even.

Frequently asked questions

Written by

Olha · clinic data analyst

I build the reporting our managers open every morning at a multi-branch medical clinic — and package it so other practices and med spas don't have to start from scratch.

I am a data analyst, not a CPA. I have read the standard; I have never applied it in a filing. This is an explanation of how the model works, not accounting advice, and the estimates involved — breakage above all — are judgement calls for you and your accountant. The accounting paragraphs are quoted from FASB's published text of ASU 2014-09 as originally issued, because the codification site is registration-walled; I did not check each paragraph against the amendments issued since 2014, and used KPMG's and EY's 2025 handbooks as the test of whether the guidance still reads this way. Those handbooks are interpretive, not codified GAAP. On the behavioral research: one study is small (33 members, one facility, 1997 data) and screened in a way that excludes disengaged members; the other is large (6,070 tickets) but from theatre rather than aesthetics. I've used them for direction, and the article says where the transfer to a med spa is weak. Two figures are named only to be flagged as untraceable, never used as data: the share of med spas said to run membership programs, and "members spend 3× more." One deliberate omission: prepaid balances can interact with state unclaimed-property rules, and ASC 606-10-55-49 says anything you must remit to a state stays a liability rather than becoming revenue. Official statute text was unreachable for most states, verified for only four, and those four are structured so differently that a summary would mislead — and no authority I found settles whether a prepaid treatment package is treated like a gift certificate at all. So this article does not summarize state law; ask a local advisor.