Accessibility testing in ACCELQ runs as part of your automated web tests. As a functional test moves through your application, ACCELQ checks each page it visits against WCAG 2.0 and 2.1 Level A and AA requirements, plus a set of best practices. You do not write separate accessibility tests or plan out page paths: every page the test touches is checked.
You also decide how accessibility results affect the test outcome. You choose which severity levels fail the test, and violations below that level are reported as warnings. For example, you can fail a run on Critical and Serious violations while still recording Moderate and Minor ones for follow-up.
This article covers enabling accessibility testing, choosing what fails the test, and reading the results in the Test Report.
Enable Accessibility Testing
Accessibility testing is turned on for a run from the Non-Functional Testing settings.
- In the Run dialog, open Non-Functional Testing.
- Select Enable Accessibility Testing.
- In Fail the test on, choose a severity threshold. See Choose What Fails the Test below.
- Click Done.
These settings are saved with the Job, so they apply every time the Job runs: ad hoc, on a schedule, or triggered from CI.
The same dialog also contains Enable Web Performance, which is covered in [Web Performance article].
Choose What Fails the Test
Every violation ACCELQ finds has a severity level: Critical, Serious, Moderate, or Minor. The Fail the test on setting decides which levels mark the test as failed. Violations at levels below your threshold do not fail the test and are reported as warnings. Only severity decides this: the threshold applies the same way to violations in all three classification groups (Needs attention, Probable issues, and Best Practices).
| Fail the test on | Fails the test | Reported as a warning |
|---|---|---|
| Critical only | Critical | Serious, Moderate, Minor |
| Serious and above | Critical, Serious | Moderate, Minor |
| Moderate and above | Critical, Serious, Moderate | Minor |
| Any violation | Critical, Serious, Moderate, Minor | None |
| Never fail | None | Critical, Serious, Moderate, Minor |
The default is Any violation.
Below the dropdown, the four severity levels show how the current selection treats each one. Levels that fail the test are marked differently from levels reported as warnings, and hovering over a level shows which treatment applies.
How violations affect the test
The threshold changes how a violation is recorded and counted, not whether it is reported. Every violation appears in the Test Report, and the test always continues with its remaining steps.
| At or above the threshold | Below the threshold | |
|---|---|---|
| In the Test Report | Recorded as an error | Recorded with a warning icon |
| Test case in the pass/fail count | Counted as failed | Not counted as failed |
| Test execution | Continues | Continues |
Severity levels
| Severity | What it indicates |
|---|---|
| Critical | Prevents some users from accessing content or completing a task |
| Serious | Makes content or a task very difficult for some users |
| Moderate | Makes content or a task harder for some users to use |
| Minor | Causes inconvenience but does not stop users from completing a task |
Which threshold, when
| Situation | Setting |
|---|---|
| Adding accessibility checks to an application with a known backlog of issues | Critical only, then tighten the threshold as issues are fixed |
| Regression runs where serious barriers must block a release | Serious and above |
| Compliance-driven runs where any violation must block a release | Any violation |
| Collecting a baseline or auditing pages without affecting test status | Never fail |
Execution and Reporting
When accessibility testing is enabled, ACCELQ checks every element on the page at the beginning of each step (Action) in the Scenario, right after page sync completes.
The Test Report lists all violations, grouped by severity and category. Select a violation type in the left panel to load the page image with the failing elements highlighted, then move between every instance of that violation on the page. The More info link for each violation explains the rule and possible remediation steps.
Classification of Accessibility Violations
Violations are listed under three groups.
| Group | Contains |
|---|---|
| Needs attention | Issues that require a fix under WCAG 2.0 and 2.1 Level A and AA |
| Probable issues | Issues that cannot be confirmed automatically and need expert review |
| Best Practices | Issues worth fixing that are not regulatory requirements |
Understanding and Remediating Accessibility Errors
Common Remediation Categories
The More info link for each violation in the Test Report explains how to fix it. Most suggested fixes fall into the following categories.
| Category | Typical violations | Typical fix |
|---|---|---|
| Content and semantics | Images without alt text, incorrect heading structure | Add descriptive text, and use the correct HTML elements (for example, <nav> and <main>) so assistive technologies have the context they need |
| Visual readability | Low contrast between text and background | Adjust CSS color values to meet the required contrast ratio |
| Form interactivity | Form inputs without a matching <label> | Associate a label with each input so screen reader users can understand and complete the form |
| Navigation and usability | Links without discernible text, keyboard traps | Give links a clear description (for example, with aria-label), and make every interactive element reachable and usable from the keyboard |
Note: Automated accessibility testing quickly identifies many potential issues on your application pages, but it supplements manual assessment and does not replace it, particularly for regulatory compliance.
Comments
1 comment
The product needs more. I would like to submit bugs for each violation, but it does not seem to allow that. additionally, I would like to be able to refine the type of violations I am interested in so future executions ignore violations I choose. Example, Minor offenses like foreground color. I would like to ignore it, so when Critical issues get fixed, in the next run the test will pass.
Aside being able to report each violation as a bug individually, It would be great if I could export the results with images. I am left manually taking screen shots and interpreting to my UI/UX team the issues I see and the element they are on. The current image download, does not download an image state with the overlay or the pinpointed element.
Lastly, often element locators are truncated, so I am only able to share a portion of the locator for the element and the UI/UX team struggles to find the violation.
Please sign in to leave a comment.