Why a clinic should audit this itself
Accessibility is frequently treated as a specialist procurement: a supplier is engaged, a report arrives, a list of technical findings is filed. That approach produces documents rather than change, partly because the findings are expressed in terms nobody at the clinic can act on and partly because they arrive detached from any sense of which failures matter.
A self-audit is different in a useful way. When the practice manager discovers that the booking form cannot be completed with a keyboard, that finding has a name, a consequence and an owner within about four minutes. Nothing in the first six criteria of this instrument requires specialist knowledge; they require unplugging a mouse and looking.
There is also a specific reason clinics should care beyond obligation. Aesthetic clinics serve a population that includes people with visual impairment, motor conditions, cognitive differences and limited English, and the material this sector publishes is unusually image-heavy. Information carried only inside a graphic, which is common on clinic social accounts and price lists, is information some patients simply do not receive.
The criteria here are drawn from the Web Content Accessibility Guidelines, which are the published reference. This instrument is a subset chosen for what a clinic can assess without tooling, and it does not substitute for a formal evaluation.
How to run the audit
Start with the keyboard. Unplug the mouse or put it out of reach. Using only the tab, arrow, space and return keys, complete three journeys: find a treatment page, submit an enquiry, and open and read the price information. Score criterion one on whether all three completed and whether you could always see where you were.
For contrast, check body text, link text and button text against the published ratios using any contrast checking tool. Light grey text on white is the usual failure and it is usually a deliberate design decision.
For alternative text, view the page with images disabled. What remains should still convey the page's information. Auto-generated alternatives, and alternatives reading "image", score 0.
For heading structure, extract the headings alone and read them as a list. If they read as a sensible outline of the page, score well. If they read as a list of design decisions, score 0.
For documents, take a leaflet you issue and attempt to select text in it. A scanned image of a printed page cannot be read aloud by anything.
For reading level, apply the same test as the communication clarity rubric: mark the terms a lay reader would not know.
Common scoring errors
Trusting an automated checker. Automated tools detect a minority of issues and cannot judge whether an alternative text is meaningful or whether a heading describes its section. Use them for contrast and markup, then look.
Scoring the home page only. The failures that matter are usually in forms, booking journeys and documents, which are the pages nobody audits.
Treating a plugin overlay as remediation. Overlays that promise conformance do not fix the underlying content, and disabled users frequently report that they add obstacles. Score the underlying content.
Claiming conformance in a statement. A statement that names known failures is more useful and more defensible than one that claims a level the clinic has not evaluated.
Excluding social media. Text baked into images with no caption is the sector's most common accessibility failure and it happens almost entirely on social accounts.
Priorities when everything cannot be fixed at once
Order the work by consequence rather than by effort. A booking journey that cannot be completed without a mouse excludes people entirely; low contrast on a footer makes something harder to read. Both are findings and they are not equivalent.
The practical order for most clinics is: keyboard operation of transactional journeys, form labelling and error messages, document formats for anything a patient must act on, contrast, alternative text, then the rest. Publishing the accessibility statement can happen at any point and is worth doing early, because it creates a route for people to tell you what you missed.
Involve a real user as soon as the obvious failures are fixed. Simulation by a sighted person with a keyboard finds a great deal and misses the things that make a site tiring rather than impossible, which is what drives people away.
Re-audit after every redesign without exception. Accessibility is lost in redesigns more reliably than it is lost anywhere else, because visual decisions are made before content decisions and contrast is the first casualty. Pair this with the technical foundations audit, which shares several underlying causes.