At the end of the month a revenue figure lands on the table, and nobody in the team can explain why it came out the way it did. That is the normal state of affairs in most aesthetic clinics, and the answers are already sitting in the appointment book and the till. Six metrics are enough to understand where your revenue comes from and where it leaks away. Here is how to calculate each one from data you already hold, and how often each is worth looking at.
Why revenue alone explains nothing
Revenue is an outcome, not a lever. It rises or falls, but it never tells you what to change. It becomes useful only once you break it into its parts:
Revenue = active patients × treatments per patient × average treatment value.
Three levers, three completely different sets of actions. If revenue is flat because too few new patients come through the door, a better treatment protocol will not fix it. If it is flat because every patient comes once and never returns, more ad spend buys a short and expensive reprieve.
Add two metrics for the losses, utilisation and no-show rate, and one for predictability, the share of recurring revenue. Six numbers in total. Nearly everything else that fills a clinic dashboard belongs in the second tier.
That includes the most popular category of all: Instagram followers, website sessions, newsletter open rates. A clinic can double its reach and earn less in the same quarter. Apply one test to every metric you are tempted to add: can you name a specific action it would trigger? If not, it is decoration.
Return rate
The single most important number in an aesthetic clinic. It describes what proportion of your patients actually come back, and it decides whether your marketing budget is an investment or a leaking bucket.
How to calculate it: take a cohort, meaning everyone treated for the first time in a given month. Count how many of them had at least one further paid appointment within the following twelve months. That share is the return rate for that cohort.
The usual mistake is calculating without cohorts. A flat “our return rate is 60%” means nothing without a stated period and window, because the figure moves on how many new patients you have recently won. Growth pushes the headline number down even when nothing has got worse.
In practice, run two versions: the twelve-month cohort for the strategic view, and the second-visit rate at 90 days as an early indicator. The second is more useful day to day, because it tells you within three months whether new patients are staying rather than making you wait a year. The step from first to second appointment is where most patients are lost.
You will not recognise a healthy return rate from an absolute value, but from newer cohorts outperforming older ones. Compare this March against last March. External benchmarks help little here, because treatment mix, price positioning and catchment area move the number more than anything you control.
Treatments per patient per year
This one measures the depth of the relationship. Calculation: all treatments delivered in the period divided by active patients in the same period, where active means anyone who attended at least once.
Two decisions determine whether the number is usable. First, count redeemed treatments, not sold ones: someone who buys a six-session course and attends three has had three treatments. Second, calculate it by treatment category as well as in total. A patient who comes three times a year for injectables and one working through a laser course look identical in the headline figure and mean very different things commercially.
The value of this metric is diagnostic. If your return rate rises while treatments per patient stay flat, you are winning back people who only ever buy one thing from you. That is the prompt to look hard at the end of an appointment: is the next interval discussed and booked there and then, or does the patient leave without a follow-up?
Average treatment value
Calculation: treatment revenue divided by the number of attended treatment appointments. Not by the number of invoices, because one invoice can cover several appointments or a whole course.
Keep retail product sales separate, as their own metric. Blend the two and a good month in skincare reads as a good month in the clinic, which it may not be.
Here too, the overall average is the least useful cut. It gets interesting per category and per practitioner. When two practitioners deliver the same treatment and their averages diverge noticeably, the cause is rarely clinical skill. It is the consultation, the course structure, or whether add-ons are offered at all.
Before trying to lift this number, check the foundation, namely whether your prices and packages are costed properly in the first place. We worked through that in a separate piece on costing treatment prices and packages. A higher average value on a thin margin improves nothing.
Utilisation
Utilisation is the metric clinics most often fool themselves with, because the denominator gets quietly dropped. Three figures need to stay separate:
- Available hours: the hours when a practitioner and a room exist at all.
- Bookable hours: the share of those that are genuinely open for appointments, after documentation, preparation, cleaning and internal blocks.
- Booked and attended hours: what actually happened.
The meaningful ratio is attended hours divided by bookable hours. Calculate it once per practitioner and once per treatment room. The two often diverge, and that gap tells you whether you are short of people or of space.
Add a split between prime time and off-peak. If Tuesday evenings have been full for months while Monday mornings sit empty, your problem is distribution, not demand, and it is solved with scheduling and waiting lists rather than more advertising.
One warning: high utilisation with a low average treatment value is not a good position. You are fully booked with appointments that tie up your most expensive resources for little return.
No-show rate
A missed appointment costs twice: the revenue that did not happen, and the blocked time nobody else could use. The important thing is to separate two cases that most clinics lump together.
- No-show: the patient does not turn up and does not cancel.
- Late cancellation: the cancellation arrives too late for the slot to be realistically refilled, usually inside 24 or 48 hours.
Calculate both against the same base, all confirmed appointments in the period. The split is worth the effort because the causes differ: no-shows are usually a reminder problem, late cancellations more often a commitment problem.
Break both rates down by weekday, time of day, treatment type and whether the patient is new or returning. The problem almost always concentrates in a handful of situations, and a vague number turns into a specific fix. We covered which levers are worth pulling, from reminder timing to deposits, in our piece on reducing your no-show rate.
Share of recurring revenue
This one answers the question every owner asks on the first working day of the month: how much of this month is already settled before the phone rings?
Calculation: revenue from memberships, subscriptions, redeemed course sessions and agreed repeat intervals, divided by total revenue for the month.
The decisive detail is recognition. Book course revenue in the month the session takes place, not the month the course was sold. Otherwise you flatter one month, starve the next, and make your own time series unreadable. It is also the correct accounting view: a sold but unredeemed course is a liability, not earned revenue.
When this share rises quarter on quarter, your dependence on how marketing happened to perform that month falls. How to build it without quietly discounting your way there is the subject of our piece on memberships and predictable revenue.
The right cadence: weekly, monthly, quarterly
The most common reason metrics quietly die in a clinic is the wrong review interval. Looking at the return rate weekly means looking at noise. Looking at the no-show rate quarterly means hearing about a problem once it is over.
| Cadence | Metric | Why at this interval |
|---|---|---|
| Weekly | Last week’s no-show and late cancellation rates | You can still act on it in the same week |
| Weekly | Open slots in the next two weeks | Gaps can be filled from the waiting list or a recall |
| Monthly | Average treatment value, utilisation by practitioner and room | Enough volume to mean something, early enough to correct |
| Monthly | Second-visit rate at 90 days, share of recurring revenue | Leading indicators that already show a pattern month to month |
| Quarterly | Twelve-month return rate, treatments per patient | These need a long window; monthly readings are mostly chance |
| Quarterly | Revenue and margin by treatment category | The basis for pricing, staffing and equipment decisions |
Keep the weekly slot short: fifteen minutes, the same two numbers, one decision. The monthly review can take an hour. The quarterly one deserves a quiet session with the phones off.
Where the numbers come from and what corrupts them
All six metrics come from two sources every clinic already has: the appointment system and the billing system. The work is not in the arithmetic, it is in data hygiene. Five things reliably distort the results.
- Duplicate patient records. The same person under two spellings counts twice as new and never as returning, dragging your return rate down for no real reason.
- Deleted rather than cancelled appointments. If a no-show is removed from the diary instead of being marked as one, the rate can never be calculated. The status has to survive.
- Test bookings and internal blocks counted as real appointments, which inflates utilisation.
- Discounts and vouchers that reduce treatment revenue in one entry and retail revenue in the next. Pick one rule and hold to it.
- Treatments recorded without a category. Without clean categories, every breakdown by treatment type is worthless.
So write a one-page definition for each of the six: formula, numerator, denominator, period, and who produces it. It sounds bureaucratic, and it ends the recurring argument about why two people looking at the same data arrived at different numbers.
On tooling: a spreadsheet and a monthly export are enough to start, and consistency matters more than software. If you are considering a system that brings appointments, packages and patient communication together, test whether it produces these six without manual rework: cohort-based return rates rather than one blended figure, no-shows kept separate from late cancellations, course revenue recognised on redemption, and utilisation with a denominator you define. Ask to see that on your own data during the demo.
Common questions
How big does a clinic need to be for this to be worth it?
One practitioner is enough. With small case numbers the monthly figures will be jumpy, so use rolling three-month windows instead of judging each month on its own.
How many metrics are too many?
The moment a number is read out in the monthly review and nobody can name an action attached to it, you have one too many. Six to eight is comfortable. Anything beyond that belongs in an ad hoc analysis, not in the standing report.
What about patient acquisition cost?
A sensible seventh metric as soon as you spend regularly on advertising, but it only becomes interpretable once you know your return rate and treatments per patient. Only then do you know what an acquired patient is worth over time.
Should we compare ourselves with other clinics?
Carefully. Treatment mix, price positioning, location and staffing differ so much that figures from industry surveys rarely map onto your clinic. Your own time series is the more reliable reference.
Who should own the numbers?
A named person, not “reception”. Once set up, producing them should cost no more than an hour a month. If it takes longer, the problem is in your data sources, not in the metrics.