Growth leak calculator / Free, and it stays in your browser

Put a number
on the leak.

Answer eight questions about your own booking path, add the two numbers you already know, and this page models what those leaks plausibly cost you in a month. Every assumption behind the figure is printed on the page and every one of them is yours to change. It is a model, not a measurement.

The dataset behind the questions ↗ Nothing is submitted, nothing is stored
01 / What you already know

Two numbers,
then eight questions.

How this runs

It never leaves the page.Everything below is calculated in your browser as you type. No value is sent anywhere, nothing is saved, and there is no form at the end of it.

The eight questions are the leak types that came up most often across the 513 practices in our booking path index. Each one takes a minute to check on your own site.

See the index ↗
What one booked treatment is typically worth to the practice.
Inquiries, form fills, calls and booking starts together. A rough count is fine.
02 / Your booking path

Check every one that is true of your site.

Open your own site on your phone and walk it the way a new patient would. Leave anything unchecked if you are not sure. The model only counts what you tell it.

The full figure ↗
03 / The assumptions

Our modelling choices,
not measured facts.

Every number in this panel is an assumption IteraUX picked. None of it is a published benchmark, a survey result, or something we measured. What we measured is how often each leak turns up, which is the percentage printed beside each question above. What a leak costs is modelled, and the model is here in full so you can disagree with it. Change any figure and the number below changes with it.

The model, open for editing

Assumed share of monthly booking attempts lost when this leak is present, as a percentage.

Per leak / share of booking attempts lost
Across the model
How much weight each further leak keeps, next to the one above it.
How far either side of the point estimate we report, because a single figure would be a lie of precision.

Leaks overlap. The same patient often meets two of them and only leaves once, so the model does not add the percentages up. The leaks you checked are sorted largest first. The largest applies in full, each one after it applies at 70% of the weight the one before it carried, and what is left is combined so that every leak acts on the attempts the earlier ones did not already take.

04 / Modelled monthly exposure
$0 to $0

Point estimate $0 a month, about $0 a year at the same rate.

Check the statements above that are true of your site. The figure appears here, and moves as you go.

Your path checks out clean on these eight. The leaks we usually find from here are the ones a checklist cannot see: what the treatment page fails to answer, what the form asks for too early, and what happens to an inquiry nobody replies to before Monday.

A modelled number,
not a measured one.

This is arithmetic run on two figures you supplied and eight assumptions we chose. It is not a reading taken off your analytics, and nobody has looked at your site. Treat it as the shape of the problem, worth what its weakest assumption is worth.

Leaks you checked0 of 8
Modelled share of attempts lost0%
Monthly demand in play$28,000
05 / How the arithmetic works

What the calculator does

Your average treatment value multiplied by your monthly booking attempts gives the demand in play for a month. The leaks you checked produce a single modelled share of those attempts, and that share of the demand is the point estimate. The range either side of it is the point estimate plus and minus the range width in the panel above, set at 35% to start with. The yearly line is the monthly point estimate multiplied by twelve, which assumes nothing changes, and something always changes.

The stacking rule

Percentages of the same population cannot be added together. A patient who meets a broken booking button and a missing price only leaves once, and she leaves at whichever she meets first. So the model sorts the leaks you checked by their assumed share, largest first. The largest applies in full. Each one after it applies at 70% of the weight the leak before it carried, compounding down the list. Those discounted shares are then combined multiplicatively, which is to say each one acts on the attempts the earlier ones left behind rather than on the original count. Checking all eight therefore lands well under the sum of the eight numbers in the panel, which is the point.

Why this is a model and not a measurement

The per-leak percentages are IteraUX's judgement, informed by walking several hundred booking paths. They are not a published statistic, an industry average, or a figure we can source to a study, and we are not going to dress them up as one. Different practices, treatments, price points and traffic sources would move every one of them. That is exactly why they sit in an open panel with editable fields instead of buried in a script. If you think a leak costs half what we assumed, halve it. The arithmetic will hold either way.

The other honest limit is the input. Everything here rests on two numbers you estimated and eight boxes you ticked from memory. Nobody opened your site.

Where the percentages beside the questions come from

Those are measurements, and they are the one part of this page that is not modelled. IteraUX has walked the booking path of several hundred med spa and aesthetic practices on a phone, from the first click toward a real time on a real calendar, and written down every leak found. 513 of those audits are published, anonymised, in the booking path index, along with how the leaks were grouped and what is deliberately left out of it. The percentage beside each question is how many of those 513 practices had that leak documented. It says nothing about how much any of them cost.

What we do not collect

Nothing. The calculator is a single file of JavaScript that runs in your browser. There is no form, no submission, no analytics event tied to your figures, and no request of any kind leaves the page while you use it. Close the tab and the numbers are gone.

Remote / working with practices across the US

The model guesses.
The scan looks.

06 / Fifteen minutes

We open your site together and walk the path a patient walks, on a phone, while you watch. You get the real count instead of the checklist version of it. No deck and no pitch.