Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more

Effective accessibility testing combines automated checks, expert manual review, and usability testing with disabled people. No single scan can establish that a website or app is accessible. For a formal conformance evaluation, use W3C’s WCAG-EM 2.0 method, which covers websites, apps, and other digital products.

What accessibility testing can—and cannot—tell you

Accessibility testing asks more than whether a tool reports errors. It examines whether a product meets applicable accessibility requirements and whether people with different disabilities can use it to complete real tasks. Those are connected but distinct questions.

WCAG 2 success criteria are testable, but evaluating them requires both automated checks and human judgment. W3C puts the limitation plainly: “Web accessibility evaluation tools can not determine accessibility, they can only assist in doing so.” Tools can flag potential problems, help reviewers inspect a product, and sometimes produce false or misleading results. A clean report is not proof of conformance.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Likewise, meeting applicable WCAG criteria does not guarantee that every person will find a product usable. Include disabled people in usability testing as well as evaluating criteria. W3C recommends that users with disabilities be included in test groups when usability testing is performed.

#1 Best Overall

How to test a website or app: a repeatable workflow

Start before release and repeat the work as the product changes. W3C recommends evaluating early and throughout development, when accessibility problems are generally easier to address. For a formal review, WCAG-EM 2.0 provides a five-step structure.

1. Define the evaluation scope

State what product and experience you are evaluating: for example, a website, mobile app, desktop product, or a defined part of one. Identify intended users, relevant technologies, the WCAG version and conformance level you are assessing, and any areas excluded from the evaluation.

Be clear about boundaries such as authenticated areas, embedded content, documents, or important user journeys. A report can only support conclusions about the scope actually evaluated.

What’s actually slowing this PC down?

Pick the symptom - the matching free tool is one click away.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

2. Explore the product and its functionality

Inventory the views, content types, and features that matter to users. Include key interactions and tasks—not just a list of URLs. Note differences in layout or behavior, such as forms, menus, media, account flows, and error states, so the evaluation can cover more than the home page.

3. Select a representative sample

If you cannot inspect every view, select a sample using the WCAG-EM approach. Choose pages or screens that represent the product’s different templates, content, functionality, and important paths. Record why those items were selected and what remains outside the sample; do not imply that untested areas passed.

4. Evaluate the sample with tools and people

Run appropriate automated checks during development and, where useful, in continuous integration. Treat each finding as a candidate to investigate: confirm whether it is an actual issue, understand its effect, and look for problems the tool cannot detect.

Then manually assess criteria that need context or interpretation. Review keyboard operation, semantics, content, and complete interaction paths. Use relevant assistive technologies and expertise appropriate to the product. W3C’s WCAG-EM guidance says evaluators should understand accessibility standards, accessible design, assistive technologies, and how disabled people use digital products.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

Finally, test usability with disabled participants completing intended tasks. Observe whether they can find information, operate controls, understand feedback, and recover from errors. This evidence answers questions about practical use that a criteria checklist alone cannot settle.

5. Report findings, remediate, and retest

Document the scope, evaluation method, environments, sampled views, findings, and limitations. Give each finding enough context for the team to reproduce and address it, then retest the affected experience after changes. Present results so readers can distinguish what was evaluated from what was not.

Rank #4

The W3C WCAG-EM Report Tool can help structure a report from findings you supply. It does not perform the accessibility checks for you.

Choosing accessibility testing tools

Choose tools for the work your team needs to do, not for a headline score. W3C notes that teams may combine tools; the right mix depends on the organization, content complexity, and available skills. Compare candidates against practical requirements:

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.
  • Method: Is the tool automated, manual, guided, or simulated? What does it actually help a reviewer determine?
  • Product type and content: Does it support your website, mobile app, document, desktop product, or other digital content?
  • Standards and criteria: Which accessibility standards and specific criteria does it address?
  • Coverage: Does it inspect a single page or a broader product, including restricted or authenticated areas?
  • Workflow: Does it fit your development process and integrate with the systems your team uses?
  • Reporting: Can findings be understood, tracked, assigned, and retested?
  • Environment and language: Check operating-system, browser, and language support for your product and users.
  • Accessibility of the tool: Can the people who need to use the tool operate it accessibly?
  • License and cost: Confirm terms and costs for the intended users and use case.

W3C maintains a provider-submitted list of web accessibility evaluation tools, including free and commercial options. W3C explicitly does not endorse specific products, so treat the list as a starting point for evaluation rather than a recommendation.

Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Support on Ko-Fi

Where ACT rules fit

Accessibility Conformance Testing (ACT) rules document how particular accessibility checks can be applied consistently across automated, semi-automated, and manual evaluation. They are principally intended for developers of evaluation tools and methodologies. Individual rules can also help evaluators handle edge cases more consistently; they do not replace product testing or usability research.

W3C’s ACT overview reports that ACT Rules Format 1.1 was published in February 2026.

Why WCAG-EM 2.0 matters for apps

W3C published WCAG-EM 2.0 as a Group Note on 23 July 2026. Unlike the previous version, which focused on websites and pages, the updated method also applies to apps and other digital products. It offers a consistent framework for defining the evaluation, choosing a sample, reviewing it, and communicating results.

Free tools Windows power users keep installed

One-click scans. No signup required.

Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.

WCAG-EM is an evaluation method, not a tool that automatically tests a product or a guarantee that a product is usable by everyone. Pair a structured conformance review with ongoing checks and usability testing.

Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.