What a med spa website actually costs

The short answer

A med spa website from us costs $2,500, $4,500 or $6,500, one time, depending on the tier. Repairing the booking path on a site you already like costs from $750, flat, paid on completion. The website assistant is $199 a month on its own, with no setup fee. Every one of those figures is on the pricing page, and none of them is a starting point for a quote. The number you read is the number you pay.

That is the answer to the question in the title. The rest of this article is about a quieter question we think matters more: which of those figures a practice actually needs. Having walked a few hundred med spa booking paths, our honest answer is that most practices asking what a website costs do not need a website. They need the $750 repair, and some of them need nothing at all.

Why the cheap answer comes first

Between 26 and 27 August 2026 we walked the booking path on 221 med spa domains across North and South Carolina, once each, with an automated walker that takes the route a new patient takes, from the homepage to the vendor’s first screen and onward until it sees selectable appointment times or something stands in the way. The walk and its limits are described on the methodology page, and the first article holds every audit figure this one borrows.

36 of the 140 paths we could walk had at least one finding, or 25.7%, roughly one walked path in four. The denominator is the walked population: every domain where the walker reached a booking surface and something happened there. It excludes the 19 domains a vendor’s robots file kept us out of and the 62 where no booking surface was reached, because nothing was walked in those cases and counting them either way would be inventing a result.

Here is what the 38 findings on those 36 domains were. Thirteen were a phone-verification wall standing before any appointment time. Eleven were no online booking mechanism at all. Four were a booking link that resolved to an error page. Three were a page rendering wider than a 390px viewport. Two were a login form with no visible new-patient route, two were a booking widget that rendered nothing after a 22-second settle, and one each were an account wall, a form error, and a booking URL caught in a redirect loop.

Notice what is not on that list. Not one finding says the site looked dated, the photographs were wrong, or the copy was clumsy. The walker measures none of that. What it measures is whether the path works, and that is why this article puts the repair before the build.

The $750 repair, and what it is for

The repair is for a practice that likes its site and does not like what happens when somebody tries to book on it. The design stays. The photography and copy stay. The booking system stays, whichever one it is, because the repair works on everything in front of the calendar and never inside it.

It starts with the audit, which is free. We walk every place a patient can try to book on your site, the menu, the homepage, the service pages, the footer, the tap-to-call links, the forms and the calendar they all lead to, on a phone and on a desktop, and write up what we find with screenshots. You keep that report whether or not you buy anything. If it finds nothing broken, there is nothing to fix and nothing to pay, and we would rather tell you your path is clean than invent work.

If it finds something, the quote is the price: one flat figure, from $750, agreed before work starts and paid on completion. Every defect the report names is repaired on the pages it names, then clicked through on real phones before we call it done. If the defects in your report are not repaired and verified working, you owe nothing. It takes about five business days, and your site stays up throughout.

Set that against the finding classes above and the fit is not accidental. A booking link that lands on an error page is a repair. A booking URL in a redirect loop is a repair. A widget that rendered nothing is, in most cases, a repair. A path that resolves to a staff login because the button points at the wrong place is usually a repair, and a page that renders wider than a phone screen often is too, depending on how the page was built. A form that throws an error while it is being filled is a repair. Those are wiring defects, and wiring is what $750 buys.

What it does not buy is a new site around the wiring. If you have been putting off sending people the link because of how the site looks, the repair will make the booking button work and will not change how you feel about the page it sits on. We will say so in the report, because selling a repair to a practice that needs a build is getting paid for the wrong thing.

What no invoice from us can fix

The largest class in the audit is the one we want to be clearest about, because it is not something we sell.

All thirteen phone-verification walls were one vendor’s default online-booking flow, AestheticRecord’s, shipped thirteen times on sites whose owners in all likelihood never chose it. A wall like that is not broken wiring. It is a setting, and a practice can change it inside its own vendor account without paying anybody, including us. SMS verification before booking is a defensible feature that cuts no-shows, and a practice that looks at the setting and keeps it is making a reasonable choice. The audit records only that the wall stood before any appointment times were shown. It does not measure what that costs, and neither do we.

The second largest class is no online booking at all, eleven of the 36. That is a business model, not a defect, and the audit counts it only because the question it asks is whether somebody can book online. A phone-only practice that wants to stay phone-only needs nothing from this price sheet. One that wants to add online booking needs a booking vendor first, and possibly nothing else.

And if the audit finds that the calendar itself is failing, rather than the path that leads to it, the report says so and a repair is the wrong purchase. We work in front of the booking system, never inside it, so what we can do is show you what we found so you can take it to your vendor.

Between those three cases, a large share of what the audit flagged costs nothing to put right, or is not wrong at all. That is the part of the answer a pricing page has the least incentive to say, which is why it is here.

When it is a build

A build is right when the site itself is the problem rather than the path through it. The signs are ordinary. The site looks dated next to the work you actually do. You hesitate before sending someone the link. Patients still call to ask things the site should answer. Nobody has touched it since whoever built it left.

It is wrong in some equally ordinary cases. If you like your site and only the booking is broken, that is a $750 fix, not a build. If you need it live next week, a build is three weeks, and three weeks is three weeks. If you want someone to run your ads and post on Instagram, we do not do that; we build the systems that book patients. And if your booking path is already clean and you are happy with the site, you do not need us at all.

What moves the price from $2,500 to $6,500

Three tiers, each a fixed quote agreed before we start and never moved, each live in about three weeks with your current site running until you have clicked through the new one and told us to switch it. Everything is yours on day one, in your name: the web address, the hosting, the code and the visitor statistics.

The Foundation, $2,500, is a clean, fast site where every page ends in a booking button. Custom design for phone and desktop, loading in under two seconds. Your services in plain patient language with a question section for each, your Google reviews on the site, a before-and-after gallery done properly, and financing options displayed.

The Found Practice, $4,500, is the Foundation plus the work of being found. Services written the way patients actually search, so a page answers the question a person typed rather than the phrase a brochure would use. Your Google listing connected, a page built for each core service, and a site structured so search engines and AI assistants can cite the practice when somebody asks who is good. Ninety days of tweaks after launch. Most practices pick this one.

The Booking Machine, $6,500, is the Found Practice plus the site working the phones. Claire, the assistant described below, answers questions and books patients at two in the morning, with her first ninety days included. Missed calls get a text back. A quiz matches patients to treatments.

The build section of this article is shorter than the repair section. That is deliberate, and it is the recommendation.

The assistant on its own

Claire costs $199 a month, no setup fee, month to month, cancel any month. If she came bundled with the Booking Machine, the first ninety days are already covered. She answers what your front desk answers, what you offer, what it costs, what to do before and after a treatment, hours and parking, and moves the conversation toward booking, mostly at night and on weekends. She does not give medical advice, and anything clinical is handed to you.

She is last on this sheet for the same reason the repair is first. An assistant in front of a broken booking path does not fix the path. It makes the failure conversational: the patient gets every question answered at eleven at night and is then walked to the same error page. If your path has a finding on it, we will tell you that on the call before we sell you an assistant. The assistant page already says as much, and it is true whether or not it costs us the sale.

What the audit does not measure about your site

The 25.7% is a figure about 221 domains we chose for other reasons. It is not a figure about yours.

The sample is not a survey. The domains came off an outbound prospecting queue, found through metro text searches for med-spa-type businesses and filtered by a predicate built to find practices worth writing to. That predicate excludes practices we had already written a friction brief about, so some of the sites most likely to have something wrong were never in the set, and whatever the rate measures, it biases the finding rate down. The findings are machine-flagged, by a deterministic walker on one day from one place, and over-reporting is its expected failure mode, which is why the operator review that produced the 36 exists. The right sentence is “the audit flagged”, never “the site is broken”.

Nothing in the audit measures a consequence. We did not count visitors, bookings or revenue, and we are not going to tell you what a broken link cost you or that a repair pays for itself. We do not know that, and a price sheet that claims to is one you should read more carefully. What we can say is narrower: on the day we walked, this was on screen, and this is what it costs to change.

Every audit figure here is pinned to two dates: measured 27 August 2026, reviewed 7 September 2026. Sites get fixed. Prices change less often, and when one does, this article is updated and says so.

How to decide in ten minutes

None of this needs us. Take a phone that has never logged into anything of yours, open your site from a search result, tap the booking button, and go all the way to the last screen before an appointment would be confirmed. Count the screens between the button and the first real appointment time, and look at where the button actually lands.

If the path is clean and you are happy to send people the link, you need nothing from this page, and we mean that. If the path is clean and you flinch at the link, you are shopping for a build. If the path stops and you like the site, it is the repair, and it starts with a free audit whose report you keep either way. Add the assistant last, once the path she would walk people down is one you have walked yourself.