Free tool · No signup

Keyboard accessibility test

A real Tab-key walk through your homepage: can a visitor without a mouse see where they are, skip the menu, reach your contact button and get back out of every widget?

No signup · Results in ~40 seconds · Nothing stored except the shareable report

What we check and why

What the test does

Automated rule engines read the page; this test uses it. We open your homepage in a real Chrome window at 1440 px and press the Tab key up to 60 times, recording every element that receives focus: what it is, its text, where it sits on the screen, whether a focus indicator is visible and whether a sticky header, chat widget or cookie banner covers it. The same run also tests the page at 320 CSS pixels wide (the equivalent of 400% zoom) and with the WCAG text-spacing override applied.

The report shows the tab stops as a numbered list with their position, so you can see the route a keyboard user takes through your page. We also run the multi-engine WCAG scan, because many keyboard problems — missing names on icon buttons, links that are not links — show up there too.

Visible focus indicator on each stop (WCAG 2.4.7) and focus not hidden behind fixed elements (2.4.11).
A skip link as the first stop that actually moves focus to the main content (2.4.1).
Keyboard traps: focus stuck on one element or cycling inside a widget (2.1.2).
Focus order that jumps against the visual reading order (2.4.3).
Clickable elements that cannot be reached with Tab at all (2.1.1).
Reflow at 320 px without sideways scrolling (1.4.10) and text spacing without clipped text (1.4.12).

Who relies on the keyboard

Keyboard access is not only for screen-reader users. People with tremors, repetitive strain injury or a broken wrist, people using switch devices or voice control, and power users who simply prefer the keyboard all move through a page with Tab, Shift+Tab, Enter and the arrow keys. If your contact form or booking button cannot be reached that way, those visitors cannot become clients.

Common problems and fixes

The most frequent failure is a theme that removes the browser's focus outline with outline: none and puts nothing in its place. Restore it with a clear :focus-visible style, at least two pixels thick and in a colour that contrasts with the background. Sticky headers that cover the focused element are fixed with scroll-padding-top on the html element. Menus and cookie banners built from div elements need real buttons, or tabindex and key handlers, and must let focus leave when they are closed.

Limits

The test covers the homepage and the first 60 tab stops. It cannot open menus that need arrow keys, fill in forms or judge whether the order makes sense for your content. Treat it as a fast first pass, then try your key journeys by hand with the keyboard. It is not a compliance certificate.

Frequently asked questions

How do I test keyboard access myself?

Click in the address bar, then press Tab repeatedly. You should always see where focus is, reach every link and button in a sensible order, open and close menus and dialogs, and never get stuck. Shift+Tab goes back.

What is a skip link?

A link at the very top of the page, usually visible only when focused, that jumps past the navigation to the main content. Without it a keyboard user has to tab through every menu item on every page.

Why does a cookie banner matter here?

Banners often appear on top of the page without taking focus, so keyboard users tab invisibly behind them, or they trap focus without a way to dismiss them by keyboard. Both are reported.

Is focus not obscured a new requirement?

Yes. Success criterion 2.4.11 Focus Not Obscured (Minimum) was added in WCAG 2.2: the focused element must not be entirely hidden by content you created, such as sticky headers or chat widgets. We flag elements at least half covered.

Does passing this test make my site accessible?

No. It measures specific keyboard behaviour on one page. Combine it with the accessibility checker, manual testing and, ideally, testing by people who use assistive technology.