A button visibly says “Search,” but assistive technology receives the name “Find products.” Both phrases sound reasonable in isolation. Together, they can create a mismatch for someone trying to activate the control by saying the words they see.
WCAG’s Label in Name guidance explains why the accessible name should contain the visible text label. This supports people using speech input as well as consistency between the visual interface and its programmatic description.
Inspect what the browser exposes
Check the accessible name in browser accessibility tools rather than guessing from the HTML text. An added ARIA attribute can override a visible label, so a well-intentioned attempt to be more descriptive may replace the words users rely on.
Prefer straightforward native labels and button text where they fit the task. If additional wording is needed, keep the visible phrase intact within the accessible name and avoid unnecessary repetition. An icon-only button presents a different naming question because it has no ordinary visible text label.
Test a concrete instruction
Ask what would happen if someone said “click Search” while looking at the page. Then compare the name announced by a screen reader. Both should lead to the intended control without requiring knowledge of hidden wording.
Repeat the check after translations, loading states and responsive layout changes. A mobile variant may show a shortened label while retaining an unrelated accessible name from the desktop component.
Do not use this check as the only accessibility test. Keyboard operation, focus, role, state and useful error handling still matter. Its scope is specific: make the words people see agree with the name software receives. When those two layers stay connected, the interface is easier to describe, operate and support across different ways of interacting.

