Automated accessibility testing has a few blind-spots which require manual testing. These include checking whether:
- The page has a logical tab order
Ensure the order in which elements are focused generally follow reading order when tabbing through your page.
- Interactive controls are keyboard focusable
Allow focus for each custom control and display an indicator when in focus.
- Interactive elements indicate their purpose and state
Distinguish links, buttons and other interactive elements from non-interactive elements. In addition, you should include an indication of their state.
- The user’s focus is directed to new content added to the page
Draw the user’s attention to page updates (ex. triggered by AJAX) with an appropriate heading and focus on the element.
- User focus is not accidentally trapped in a region
Let users navigate back and forth between each page element using only the keyboard.
- Custom controls have associated labels
Develop interactive controls so that they are keyboard focusable.
- Custom controls have ARIA roles
Assign an appropriate role attribute and any associated ARIA attributes. You should also indicate the property and state for custom controls.
- Visual order on the page follows DOM order
Assess whether: The DOM elements are arranged in a logical order?Offscreen content is indeed hidden from navigation.
- Offscreen content is hidden from assistive technology
Only expose visible parts of the page to screen readers.
- HTML5 landmark elements are used to improve navigation
Use HTML5 landmark elements such as
footer, which put the page into context for screen readers allowing them to jump to those regions. By doing this, you will significantly improve the user experience of your site for those using assistive technology.
If you’re a web developer, make sure to learn about how to apply these here so that you are developing accessible websites from the outset.
Connect with us if you have any questions or wish to discuss how we can help your team develop accessible digital tools and websites.