Med Spa Rebooking Rates: Where the 1.85 Figure Comes From, and What Actually Moved Retention

The most quoted number in med spa retention is 1.85 treatments per patient per year. It is an estimate rather than a measurement, it is scoped to botulinum toxin retreatment rather than to all med spa visits, and it reaches the literature as a citation to a subscription dataset inside a three-page commentary funded by a company that sells a toxin marketed on lasting longer.

That does not make it useless. It makes it a directional figure that has to be introduced with its provenance attached. The gap it points at is real, because the other half of the comparison comes from a regulator rather than a vendor: the FDA-approved BOTOX Cosmetic label states that duration of effect for glabellar lines is approximately 3 to 4 months. What the label does not do is recommend four visits a year. It says the opposite kind of thing, that dosing more often than every three months has not been clinically evaluated.

There is one paper in this literature that measured a retention intervention instead of describing the problem, and almost nobody cites it. A 2006 chart review in Dermatologic Surgery reports retention of 55% before a policy change and 67% after it. The policy was a mandatory two-week post-treatment evaluation for new patients, which is a scheduling rule, not a marketing campaign.

This post traces the 1.85 figure to its origin, prints what the FDA label actually says, works through the 2006 study including the places it contradicts itself, and then gives you definitions for rebooking that you can run against your own books. Where a definition has no source outside our own operating practice, we say so.

Every source on this page was read on 19 August 2026. We make med spa software, so we have an obvious interest in you measuring rebooking and acting on it. The sourcing, the definitions and the arithmetic below hold regardless of whose software you run them in.

The gap, stated as carefully as the sources allow

Here is every figure in the rebooking argument with its actual source, what the source is, and how far the number can honestly be pushed.

Read the first two rows together and the shape of the problem is obvious. The number everyone repeats and the dataset behind it live in different places. The commentary is public and free to read on PubMed Central. The 690,513 patients are inside a paid panel whose access options, read on 19 August 2026, are a demo request and a client login.

Read rows three and four together and the cadence half changes character. A great deal of retention marketing turns "approximately 3 to 4 months of effect" into "four visits a year, therefore your patients owe you two more." The label supports the first clause and says nothing that supports the second as a clinical standard. We found no clinical guideline recommending a fixed three-month cadence, read 19 August 2026.

Row five is the useful one for an operator, with its caveats intact. If 57% of patients at a selected set of practices came back within six months in 2005, that is broadly consistent in direction with an estimate under two treatments a year. Two weak, old, commercially sponsored data points pointing the same way is not proof. It is enough to justify measuring your own number rather than assuming it.

Where the 1.85 number actually comes from

The sentence appears in the first paragraph of Dover, Solish, Gross, Gallagher and Brown, "Bridging the Gap: Sustained Treatment Effect of Glabellar Lines With Twice-A-Year Treatment With DaxibotulinumtoxinA," Dermatologic Surgery 2023;49(9):862-864, PMID 37384899. The wording carries its own hedge.

"However, in clinical practice, it is estimated that an average patient returns for 1.85 treatments per year."

The parenthetical that follows attributes the figure, and the reference line is printed in full in the paper: "Guidepoint. Analysis of QSIGHT Aesthetics Data 2019-2021. QSIGHT Website; 2021. Available from: https://qsight.guidepoint.com/. Accessed October 20, 2022." The authors were transparent about it. The distortion happens downstream, when the figure gets recycled as a research finding.

Three things follow, and each of them matters when you decide how much weight to put on the number. First, the article is a three-page commentary with six references. Its own methods are limited to comparing two previously published median durations of effect. It is not a cohort study of 690,513 people and it should not be described as one.

Second, the commercial interest is total. Gross, Gallagher and Brown are employees of and/or hold stock in Revance Therapeutics. Dover discloses Revance among his commercial relationships and Solish discloses research and/or honorarium from Revance Therapeutics. Medical writing for the paper was funded by Revance Therapeutics. All five names on the byline have a Revance tie, and the paper's argument is that a twice-yearly product suits the low visit frequency it cites. The low number is the premise of the sales case.

Third, the underlying data cannot be inspected. We found no published figure for average treatments per patient per year on the Guidepoint Qsight site, and no published methodology or dataset documentation there, read 19 August 2026. The site offers a demo request and a client login. That is not an accusation of anything. It is the ordinary condition of commercial panel data, and it is the reason the figure should be introduced as an estimate from a paid panel rather than as a research result.

What the FDA label says about cadence, and what it does not

If you want the cadence side of this argument to rest on something other than a manufacturer's commentary, use the label. The BOTOX Cosmetic prescribing information from Allergan, BLA 103000, as posted on DailyMed with a label revision date of 18 October 2024, contains two sentences that do all the work.

Section 2.4 states that the duration of effect of BOTOX Cosmetic for glabellar lines is approximately 3 to 4 months. Section 2.2 states that the safety and effectiveness of dosing more frequently than every 3 months have not been clinically evaluated. The first sentence gives you the biological clock your appointment book is competing with. The second tells you the regulator has not blessed anything tighter than a quarterly interval.

The honest version of the thesis is therefore narrower than the version in circulation, and it is still strong. Effect fades over roughly three to four months. The best available estimate of actual toxin retreatment frequency sits below two a year. Somewhere between the two lies a set of appointments that were clinically reasonable, that the patient would likely have accepted, and that were never put on a calendar. Your job is to find out how many of those are yours.

One more note on product claims, since they show up in the same conversations. The DAXXIFY prescribing information from Revance, BLA 761127, as posted on DailyMed with a November 2023 label revision, states in section 2.1 that the product should be administered no more frequently than every three months for any indication, and states in section 14.1 only that subjects were followed through at least Week 24. The 168-day median figure is the manufacturer's own reading of pooled trial data in its commentary, not something the label establishes.

The 2006 chart review is the most useful paper here, and it is barely cited

White L, Tanzi EL, Alster TS, "Improving patient retention after botulinum toxin type A treatment," Dermatologic Surgery 2006 Feb;32(2):212-215, PMID 16442041, from the Washington Institute of Dermatologic Laser Surgery in Washington DC, did something the rest of the literature mostly skips. It changed a policy and then measured what happened.

The design was a retrospective chart review of all patients who received botulinum toxin injections in one private cosmetic dermatology practice over a two-year period, plus a survey of patients who had discontinued. The abstract reports an initial patient retention rate of 55%, and a 67% patient retention rate after initiation of the mandatory two-week post-treatment office evaluation for new patients.

The table below sets out what that source states and what it does not, because the second column is as load-bearing as the first.

Take the contradiction in row four seriously rather than resolving it in your own favour. The Objective describes a single intervention intended to improve patient retention. The Conclusion states that managing patient expectations of botulinum toxin treatments and mandatory post-treatment appointments for evaluation of the initial procedure increase the patient retention rate. Two mechanisms, one result, no split. Anyone who tells you the appointment alone bought twelve points has read past the conclusion.

Take the design seriously too. One practice, retrospective, before and after, no control arm, no p-value in the abstract. That supports an association at one site in the mid-2000s. It does not support "do this and your rebooking rate goes up twelve points," and we are not going to write that sentence.

What survives is still worth more than most of what gets published about retention. A clinic changed one operational rule about when the next appointment happens, and the retention number moved in the right direction by a margin large enough to notice. That is a testable idea you can run yourself, which is more than the 1.85 figure offers.

Two of the four stated reasons for stopping are scheduling problems

The same 2006 paper surveyed patients who had discontinued. The abstract lists the most common reasons cited: procedural cost, patient failure to re-schedule, perceived lack of product longevity, and clinical effect falling short of expectations. The paper gives no ranking, no percentage split and no survey response rate, so treat these as four named reasons rather than a distribution. They are also described as the most common reasons, not the only ones.

Sort them by what a clinic controls at the front desk and the list splits cleanly in half. Procedural cost and perceived lack of product longevity are pricing and product questions. Patient failure to re-schedule and clinical effect falling short of expectations are process questions, and they are the two that a checkout-time booking and a scheduled two-week review touch directly.

That mapping is the strongest structural argument in this whole literature and it is sitting in a twenty-year-old abstract. A patient who never re-scheduled did not decide against you. A patient whose result fell short of what she pictured had nobody look at her face at the two-week mark and explain what she was seeing. Both are addressable without discounting, which matters if you have read our comparison of retention spend against acquisition spend and concluded that discounting your way to loyalty is the expensive path.

Rebooking and reactivation are different events, and the difference is a timestamp

Before you can act on any of this you need one number that means the same thing every month. That is harder than it sounds, because there is no external standard to adopt. We searched for one and found no standards body, journal or government source that defines rebooking rate, a rebooking window, or a reactivation threshold for medical aesthetics, read 19 August 2026. The only parties publishing formulas are software vendors and consultancies, ourselves included.

So the table below is Velarya's own operating definition, offered as a starting point rather than an industry standard. Pick one row, write it down, and use the same one every month.

The distinction that matters is which timestamp you count. A rebooking rate should be measured on the date the next appointment was created, not the date it happened. If you count on the date it happened, a patient who booked at checkout and a patient you chased for five weeks land in the same bucket, and you lose the ability to tell whether your checkout is working.

Set your lapse threshold from the clinical clock rather than from habit. For toxin, the label puts duration of effect at approximately 3 to 4 months, so a patient at 150 days is late and a patient at 300 days is a different problem requiring a different message. A single 12-month lapse threshold applied across every service line will classify a toxin patient as active for most of the period during which she has already gone somewhere else.

One caution on comparing eras. The 55% and 67% from 2006 cannot be lined up against any of the four formulas above, because the abstract never states what its retention rate counted or over what window, and the full text is paywalled. Use it as evidence that a scheduling policy moved a retention number at one practice. Do not use it as a benchmark for yours.

The denominator that hides a checkout failure

Most rebooking rates are calculated on a denominator that quietly excludes the patients you most need to see. Run through what typically gets dropped: cancellations, no-shows, consultations that did not convert to treatment, and in many setups anything that was not a paid service line. Each exclusion has a defensible reason and the combined effect is a number that only counts the visits that already went well.

The version worth watching is the ugly one. Put every completed visit in the denominator, including the ones with a zero-dollar line, and count in the numerator only appointments created before the patient left the building. That single ratio will tell you within one month whether your checkout is a scheduling step or a payment step.

Then split it by staff member and by service line before you draw any conclusion. Rebooking is a behaviour performed by a specific person at a specific counter at a specific hour, and a clinic-wide average is usually two or three very different numbers averaged into something nobody can act on. If one injector pre-books at double the rate of the rest, you do not have a software problem, you have a script that needs copying. Our longer piece on med spa client retention works through the reporting cuts that surface this.

What your software has to do here is narrow and specific. It has to store the creation timestamp of an appointment separately from the appointment date, attribute the creation to the user or the patient who made it, keep completed visits with a zero total in the denominator, and let you cut all of that by provider, location and service line without exporting to a spreadsheet. If you cannot answer "who booked this, and when did they book it" from your reports, every rebooking number you produce is an estimate.

Building the second appointment into the first, by service line

There is multicenter evidence that treatment mix affects whether patients stay. Humphrey S, Trindade de Almeida A, Safa M, Heydenrych I, Roberts S, Chantrey J, Ogilvie P, "Enhanced patient retention after combination vs single modality treatment using hyaluronic acid filler and neuromodulator: A multicenter, retrospective review by The Flame Group," Journal of Cosmetic Dermatology 2021;20(5):1495-1498, PMID 33217111, reviewed more than 2,600 patients across seven aesthetic practices on five continents and compared retention at one, three and five years. Combination therapy significantly increased the probability of retention at each time point against neuromodulator alone and against filler alone, both at P below .0001.

The abstract reports significance and no absolute retention percentages, and the full text returned HTTP 403 at Wiley when we attempted it on 19 August 2026, so there is no percentage here to quote and we are not going to invent one. The paper also treats retention as a surrogate marker for patient satisfaction, which is an assumption rather than a measurement. Read as direction rather than magnitude, it says that the patient with more than one reason to come back comes back more often.

Operationally that turns into a per-service-line rule about what the next appointment is, decided before the patient is at the counter. Toxin has an obvious answer given a 3 to 4 month duration of effect on the label. A first-time injectable patient has a second obvious answer in the 2006 paper's two-week review. Laser and energy-based series have their own protocol intervals. Skincare and facials have a product runway you can count in weeks.

The failure mode is leaving the decision to the front desk in the moment. If the next appointment for a given service is a policy rather than a judgement call, the checkout conversation stops being a sales attempt and becomes an instruction, which is exactly what the 2006 practice appears to have done when it made the two-week evaluation mandatory rather than offered. When the preferred slot is full, a structured waitlist keeps the patient inside the booking flow instead of sending her away to think about it.

How to check your own clinic in an afternoon

None of the numbers above are benchmarks. There is no primary-source figure for a good med spa rebooking rate, and we found none, read 19 August 2026. The only comparison worth making is against your own trailing baseline. Here is the sequence.

Pull twelve months of completed visits with four fields: patient, service line, visit date, and the creation timestamp of the patient's next appointment if one exists. Calculate pre-book at checkout, defined as next appointment created on or before the visit date, on a denominator of all completed visits. Do it for the clinic, then per provider, then per service line. That is your baseline and it took an afternoon.

Now pick one service line, most usefully first-time injectables, and change one rule for ninety days. Make the two-week post-treatment evaluation a standard part of the first treatment rather than an option offered at the end. Book it while the patient is still in the chair. Say out loud what she should expect to see at day three, day seven and day fourteen, because the 2006 conclusion credits expectation management alongside the appointment and you may as well run both arms of what that practice actually did.

Measure the same four cuts at the end of the ninety days and again at six months, since a toxin retention effect cannot show up in a window shorter than the duration of effect. Compare against your own baseline only. If the number moves, you have one practice's evidence, which is the same class of evidence the 2006 paper produced, and it is about your practice. If you want to see how this is instrumented end to end, our retention tooling is built around the creation timestamp for exactly this reason, and you can book a walkthrough if you would rather see the reports than build them.

Frequently asked questions

How many times a year does the average aesthetic patient come back?

We found no primary source measuring visit frequency across all med spa services. The figure usually quoted, an estimated 1.85 treatments per patient per year, is specific to botulinum toxin retreatment and appears in a 2023 Dermatologic Surgery commentary by Dover and colleagues, which cites Guidepoint's analysis of the proprietary QSIGHT Aesthetics dataset for 2019 to 2021. The commentary hedges it as an estimate and collected no patient data itself. Treat it as directional, not as a description of your clinic.

What is a good rebooking rate for a med spa?

No standards body, journal or government source publishes a benchmark, and we found none when we searched on 19 August 2026. Any percentage presented as an industry average traces back to a vendor or a consultancy rather than to a primary source. The practical answer is that your baseline is your benchmark. Calculate pre-book at checkout for the last twelve months, split it by provider and service line, and measure every change against that.

Is a rebook 60 days later still a rebook?

That depends entirely on which timestamp you count, and there is no external standard to settle it. Our own definition counts the date the next appointment was created, not the date it takes place. Under that rule, an appointment created 60 days after the visit is not a checkout rebook, whatever date it lands on. Keeping the two apart is what lets you tell whether your front desk is booking patients or whether your recall campaigns are cleaning up afterwards.

Does a two-week follow-up visit actually improve retention?

One paper reports that it moved with one. White, Tanzi and Alster, Dermatologic Surgery 2006;32(2):212-215, report retention of 55% at one private cosmetic dermatology practice, and 67% after a mandatory two-week post-treatment evaluation for new patients was introduced. The conclusion credits managing patient expectations alongside the appointment, and the design is a single-site retrospective before-and-after with no control group and no p-value in the abstract. It is an association at one practice, worth testing at yours.

Why do injectable patients stop coming back?

The 2006 Dermatologic Surgery chart review surveyed discontinued patients and lists the most common reasons cited as procedural cost, patient failure to re-schedule, perceived lack of product longevity, and clinical effect falling short of expectations. The abstract gives no ranking, no percentage breakdown and no response rate, so these are four named reasons rather than a distribution. Two of the four, failing to re-schedule and results falling short of expectations, are process problems a clinic controls without changing price.

Should I pre-book at checkout or send a reminder later?

We found no published head-to-head comparison of the two. What can be said from sources is that patient failure to re-schedule was one of the reasons cited for discontinuation in the 2006 Dermatologic Surgery review, and that the retention gain reported there followed a policy of booking a specific follow-up rather than leaving it to the patient. Checkout is also the only moment when the patient, the schedule and a staff member are in the same place, which is an argument on its own.

How long a window should I use to measure retention in aesthetics?

Anchor the window to the clinical clock for the service. The FDA-approved BOTOX Cosmetic label, revised 18 October 2024, states in section 2.4 that duration of effect for glabellar lines is approximately 3 to 4 months, so a toxin retention window shorter than about 180 days will report noise. Laser and energy series follow their own protocol intervals. Applying one 12-month window across every service line will mark lapsed toxin patients as active for most of the year.

What is the difference between rebooking and reactivation?

Rebooking is securing the next appointment while the relationship is still current, measured from the creation date of that appointment relative to the visit that preceded it. Reactivation is bringing back a patient who has already passed your lapse threshold. We located no external standard defining either term for medical aesthetics when we searched on 19 August 2026, so these are our operating definitions. The reason to separate them is that reactivation success can improve while the rebooking process quietly gets worse, and a combined number hides that.