Research · run by us, measured by us
Clinic website benchmark: 272 US clinic sites, measured in August 2026
Most clinic websites look booked and are not. We measured 272 US clinic websites, med spa and dental, against what a patient and an AI assistant actually need: a booking that books, a price a patient can find, and four facts stated as text. The numbers below are ours, with the method and the limits on the same page.
- Name
- HeadPills clinic website benchmark
- Size
- 272 clinic websites
- Where
- United States, medical aesthetics and dental
- How
- checked by script against fixed rules; one month; frozen before the numbers were read
The numbers
- 92.4%show a booking button
A “Book” or “Request an appointment” call to action on the site.
- 36.7%connect it to a booking system
The button leads to a scheduler that can take a booking, not to a form or a mailbox.
- 86.9%publish no price list
Among dental practices in the sample, 95%. Of the pages named “Pricing”, about a third showed prices.
- 6.3%publish the four facts an assistant needs
What the clinic does, where, for whom and at what price, stated as text a machine can quote.
What we found
The button is there; the booking is not
Two gaps decide whether a site works: between showing a booking button and having a booking, and between having a “Pricing” page and showing a price.
Method
How it was measured
- 01A fixed list of sites
272 clinic websites in the United States, medical aesthetics and dental, collected in August 2026 and frozen before any number was read.
- 02Fixed rules, checked by script
Each site was checked for a booking call to action, whether it leads to a scheduling system, a price list, prices on any page named “Pricing”, the four facts as text, and robots rules that block AI crawlers.
- 03The four facts
What the clinic does, where it is, for whom, and at what price, each counted only when stated as text on the site, not inferred from a map, an image or a PDF.
- 04Counted, not interpreted
A site either passed a rule or did not. No scoring, no weighting, no opinion about design.
What this does not show
Limits, stated before anyone quotes it
- One country, one month, two kinds of clinic: the numbers describe this sample, not every clinic website.
- Script-based detection can miss a scheduler embedded in an unusual way, so the share with a real booking may be slightly higher than 36.7%.
- A price list was counted when present, not when complete; a page with one price passed.
- The four facts were counted as present or absent; whether an assistant actually quotes a site was not measured here.
- The full table is available on request; this page shows the rules and the totals.
What we do with it
From a number to a service
The benchmark is the reason our clinic pages start with the booking and the price instead of the design. For a clinic we connect the booking to the calendar, publish prices from, and state the four facts as text, then measure whether assistants quote the site. The same rules apply to any business whose customers book.
Questions
Questions about this study
How many clinic websites did HeadPills measure, and where?
272, in the United States, medical aesthetics and dental, in August 2026. The list was frozen before the numbers were read.
What does “connected to a booking system” mean in the clinic benchmark?
That the booking call to action leads to a scheduler that can take a booking, rather than to a form that lands in a mailbox. 92.4% of sites had the button; 36.7% had the booking.
Can I get the data behind the clinic website benchmark?
The rules and the totals are on this page; the full table is available on request by email, with the site list.
Want your own site measured against these rules?
Send the address; we reply with the same checks and what they point at.