Practice & Presence
People, practice & public presencePeople · Websites · Communication
Practice tools

Test a practice booking journey using only a keyboard

A practical walkthrough for finding lost focus, inaccessible appointment controls and confusing handoffs—without mistaking a quick check for a full accessibility audit.

Hands typing on a backlit computer keyboard on a marble surface.
Keyboard use photographed by Kamil Switalski, published in March 2026. Unsplash License.Photography by Kamil Switalski / Unsplash · Original collection ↗

Choose one journey and a safe way to test it

Begin with a specific task: find an initial appointment, select an available time, complete the required fields and understand the confirmation. Test the route visitors actually use, starting on the practice’s service page rather than opening the booking provider in isolation.

Use a test environment or a procedure agreed with the practice so that you do not occupy real appointment capacity, submit personal information or trigger messages to an unsuspecting recipient. If a safe end-to-end test is unavailable, stop before the irreversible action and record exactly how far you checked.

W3C’s keyboard guidance explains why keyboard operation matters for people who cannot use a mouse and for other input methods that rely on equivalent interactions. It is a basic part of usable design, not a special feature reserved for an optional version of the site.

Start at the service page, with the mouse aside

Use Tab to move forward through focusable controls and Shift+Tab to move backwards. Look for a visible indication of the current focus. Follow the sequence to the booking link and check that the label identifies the action accurately.

Write down the last control you could identify if the focus disappears. Record the page address, browser and the sequence of actions. “Keyboard broken” is difficult to investigate; “after the service selector, the next Tab press moves somewhere with no visible indication” gives the developer a reproducible starting point.

Pay attention to overlays such as a cookie notice or menu. Can you reach their controls, close them when appropriate, and continue to the intended page? A polished appointment screen is of limited help if someone cannot get past the first overlay.

Give appointment widgets their own check

A calendar, service selector or practitioner menu may use more than the Tab key. Native controls and accessible widgets can have expected arrow-key, Enter, Space or Escape behaviour, depending on the control. Follow any instructions supplied and record what the interface actually permits.

Check whether you can identify the selected date, distinguish available choices and revise a selection. Do not treat a visible calendar grid as proof that its dates are operable. Equally, do not assume every component must use the same keystrokes: the important question is whether its purpose and operation are understandable and usable.

Check an error and the final handoff

With safe test data, examine what happens when a required field is left blank or a permitted test error is triggered. Can you find the explanation, reach the relevant field and understand how to correct it? Make sure the correction process does not silently erase other useful information.

If booking moves to an external provider, continue the check there. Note the point where control passes between systems. A problem in the provider’s interface still affects the visitor even if the practice cannot directly edit that code. It needs a documented owner and an appropriate route for correction.

Report findings without overstating the result

Record each issue as a task, the steps that produced it, the observed behaviour and the expected result. Retest the same steps after a change. Preserve any limits on the test, including screens you could not safely reach.

W3C’s Easy Checks describes an initial review, not a comprehensive conformance assessment. A successful keyboard walkthrough therefore supports a narrower conclusion: this particular journey worked under the conditions tested. It does not establish that the whole website is accessible.

Use the Cliniko booking-journey guide when mapping the handoff, and the guide to accessible enquiry forms for the labels and errors around it. Keeping the task specific makes both the findings and the improvements easier to verify.

Project & further reading

W3C WAI: Keyboard compatibility ↗W3C WAI: Easy checks ↗

More stories

The journal ↗