Standard accessibility acceptance criteria 

This is based on a document I wrote several years ago, a proposal to standardise test acceptance criteria for our quality assurance process.

I deliberately avoided any reference to WCAG guidelines or success criteria – instead, I chose to keep the focus on the user experience, what they should be able to perceive, understand, and do, in that scenario.

It didn’t really get anywhere, but I came across it the other day and thought I may as well publish it; perhaps someone can make use of it.

It’s written in gherkin format, because that was what we were using at the time. But it could be adapted to any format you need

This is far from perfect. I welcome contributions from others more knowledgeable than me. Especially people with lived experience.

Scenario: User cannot use a mouse or other pointing device 

• Given: I am using a device with a keyboard 
• And: I cannot use a mouse or other pointing device 
• When: the page loads 
• And: the page has focusable elements 
• Then: I can give each focusable element focus using the keyboard alone 
• And: There is a clear and obvious visual cue indicating the element has focus 

Scenario: User with screen reading software

• Given: I have no visual representation of the page available 
• When: The page loads 
• And: the page has focusable elements 
• Then: I can give each focusable element focus using the keyboard alone 
• And: My software reads out a clear description of the element’s function and purpose 
• And: I can interact with all interactive elements using the arrow keys 

Scenario: User with screen reading software (2)

• Given: I have no visual representation of the page available 
• When: The page loads 
• Then: I can use appropriate keyboard commands to have the software read aloud all the content on the page 
• And: the reading order matches the structure of the page 
• And: an alternative description is provided for all images that are not purely decorative 
• And: the target of any links is clearly described 
• And: the function and purpose of any interactive elements is clearly described 
• And: other structure (e.g. ordered/unordered lists, headings, blockquotes) is announced 

Scenario: User with screen reading software (3)

• Given: I have no visual representation of the page available 
• When: The page loads 
• Then: I can use appropriate keyboard commands to have the software read aloud all the headings on the page 

Scenario: User has low vision 

• Given: I must increase text size to make it legible 
• When: I increase text zoom to 200% 
• Then: I can read all text without scrolling horizontally 
• And: I can interact with all interactive elements 

Scenario: User has low vision (2)

• Given: I have difficulty reading text with insufficient contrast between foreground and background colours 
• When: the page loads 
• Then: the colour contrast for standard text is at least 4.5:1 
• And: the colour contrast for bold or large (>17point or 13 point bold) text is at least 3.5:1 

Scenario: User is colour-blind 

• Given: I cannot perceive difference in certain colour combinations 
• When: the page loads 
• Then: wherever colour is used to convey information a redundant perceptible means is also employed to convey the same information 

Scenario: user has poor fine motor control 

• Given: I have difficulty positioning a mouse or pointing device on small or proximate targets 
• When: the page loads 
• Then: links and other interactive elements are at least 44px in each dimension 
• And: they do not overlap 
• And: I can activate and use interactive elements 

Scenario: User has cognitive or reading impairments 

• Given: I have difficulty reading 
• When: I access the website  
• Then: All text should be written in plain English 
• And: copy should have an average reading age of 13 or less 

Scenario: User using speech-to-text software

• Given: I am using STT software
• And: I ask it to click link <llnk text>
• Then: the link opens

Scenario: User using speech-to-text software (2)

• Given: I am using STT software
• And: I ask it to click <button text>
• Then: the action specified by the button is carried out