Automated scanning software is a useful tool for identifying clear violations (such as missing alternative text or low color contrast). Scanners typically detect only a portion of actual accessibility barriers, leaving complex interactions (such as modal behaviors and keyboard navigation order) unexamined. Combining automated tools with manual testing using screen readers and keyboard controls yields a complete assessment of website usability.
Automated accessibility scanners are software programs designed to parse the code of a webpage and check it against a set of predefined rules. These rules are typically based on the Web Content Accessibility Guidelines (WCAG) success criteria.
Scanners are highly efficient at identifying objective, syntax-based violations:
- Missing Alt Attributes: The tool scans all
<img>tags to verify if thealtattribute is present in the HTML code. - Contrast Math: The scanner calculates the color contrast ratio between text and its background, flagging areas that fall below the required 4.5:1 ratio for standard text.
- Form Input Associations: The software checks if form fields include an associated label or
aria-labelattribute. - Language Declarations: The scanner verifies if the main
<html>tag includes a language attribute (such aslang="en"), which is required for screen readers to load the correct pronunciation rules.
For IT departments and managed service providers, automated scanning is an essential baseline. These tools can evaluate thousands of pages in seconds, run continuously in deployment pipelines, and identify regressions when new code is published.
While automated tools are necessary for baseline monitoring, they have significant technical limitations. A website that receives a clean report from an automated scanner—or one that relies on automated accessibility widgets—can still be completely inaccessible to individuals with disabilities, leaving the business exposed to Title III litigation.
The primary limitation of scanning software is its inability to evaluate human context and logical interactions:
1. Alternative Text Context
A scanner can verify that an image includes alternative text, but it cannot determine if the text is meaningful. If an image of a company’s executive team is labeled alt="image", the scanner will mark it as compliant. However, this description fails to provide any useful information to a blind user, violating the core intent of WCAG Success Criterion 1.1.1.
2. Slashed Keyboard Focus Indicators and Focus Traps
Automated tools can detect if a focus outline is disabled in a stylesheet using CSS properties like outline: none;. However, they cannot determine if a custom focus ring is visually distinct to a user navigating the page. Furthermore, scanners cannot predict if a user will encounter a keyboard trap, where focus becomes stuck in a dropdown menu or live chat widget and cannot navigate back to the main content.
3. Asynchronously Rendered Page Elements and Modal States
When a user triggers an action (such as opening a mobile menu or a popup registration form), the page content changes without a reload. Managing accessibility during these real-time page updates requires developers to shift keyboard focus programmatically to the new dialog and trap it there until the user closes the menu. Scanners cannot simulate these multi-step user actions, leaving modal interactions unexamined.
4. Custom Component Markups (WAI-ARIA)
If a developer builds a custom interactive component, such as an accordion tab layout, using generic HTML <div> tags, a standard scanner will not flag the container as broken. Because a <div> tag is structurally neutral, the checker does not know that the element is supposed to behave like a button. Without manual verification, developers may omit critical ARIA attributes (such as role="tab" or aria-expanded="false"), leaving the control invisible to screen reader navigation paths.
Because of these limitations, web accessibility experts agree that automated scanners typically catch only thirty to forty percent of total compliance errors. The remaining sixty to seventy percent of errors require human intervention and manual testing to identify.
Manual usability testing involves evaluating a digital product using the same devices and navigation methods that individuals with disabilities rely upon.
A standard manual audit includes several evaluation methods:
- Keyboard-Only Walkthroughs: The evaluator navigates the website using only the Tab key, Shift+Tab keys, Enter key, Space bar, and Arrow keys. This test checks for logical focus order, visible focus outlines, and the complete absence of keyboard traps.
- Screen Reader Testing: The evaluator conducts key user journeys (such as searching for a product, adding it to the cart, and completing checkout) using standard screen readers. These include NVDA and JAWS on Windows, and VoiceOver on macOS and iOS devices.
- Reflow and Zoom Audits: The auditor scales browser text to two hundred percent to ensure that layouts do not overlap, text remains readable, and functionality is preserved (WCAG Criterion 1.4.4).
- Cognitive Usability Audits: The evaluator checks the site’s forms and login procedures to verify that they do not rely on cognitive tests, such as solving equations or typing transcribed characters.
Establishing a sustainable testing lifecycle requires combining automated scans with manual reviews. Automated testing provides immediate feedback for developers during coding, while manual testing provides the deep validation required before major updates or annual audits.
To achieve reliable coverage and prevent regression issues, organizations can address the limits of automated tools by combining scans with manual reviews from an ADA web accessibility provider using screen readers.
Manual evaluation data is also essential for compiling accurate Accessibility Conformance Reports (ACRs) based on the VPAT framework. Because automated scanners cannot verify conformance on complex interactive standards, manual audits provide the verification data required to document compliance. This hybrid approach ensures that digital platforms are both programmatically compliant and usable by all visitors, protecting the business from legal risk and expanding its market reach.