WCAG 2.1 AA Compliance Guide

Perceivable

Information and user interface components must be presentable to users in ways they can perceive. This means providing text alternatives for non-text content (images, charts, diagrams), captions and audio descriptions for multimedia, and ensuring content can be presented in different ways without losing information. For PDFs, perceivability requires alt text on every meaningful image, proper tagging so content structure is programmatically determinable, and sufficient color contrast for all text.

Operable

User interface components and navigation must be operable by all users, including those who rely on keyboards, switch devices, or voice control. In documents, this means interactive form fields must be navigable via keyboard with a logical tab order, and documents must have bookmarks or a table of contents to allow non-sequential navigation. Links must have descriptive text so users know where they lead without relying on visual context.

Understandable

Information and the operation of the user interface must be understandable. The document must declare its primary language so screen readers use the correct pronunciation. Content should be organized with clear headings and a logical reading order. Abbreviations and unusual terms should be defined. For forms, labels must clearly describe what input is expected, and error messages must be descriptive and helpful.

Robust

Content must be robust enough to be interpreted reliably by a wide variety of user agents, including assistive technologies. For PDFs, robustness requires proper tagged structure that conforms to PDF/UA (ISO 14289), embedded fonts so text renders correctly on any system, and valid document structure that current and future screen readers, braille displays, and other assistive technologies can parse without errors.

All non-text content — images, charts, graphs, icons, logos — must have a text alternative that serves the equivalent purpose. In PDFs, this means every meaningful image must have alt text in its tag properties. Decorative images must be marked as artifacts so screen readers skip them entirely. This is the single most commonly violated criterion in government PDFs.

Information, structure, and relationships conveyed through presentation must also be programmatically determinable. In PDFs, this means headings must be tagged as headings (H1-H6), lists must be tagged as lists (L, LI), tables must have proper table tags (Table, TR, TH, TD), and the visual hierarchy must match the tag structure. A PDF that looks structured to a sighted reader but has no tags is invisible to a screen reader.

When the sequence of content affects its meaning, a correct reading sequence must be programmatically determinable. In PDFs with multi-column layouts, sidebars, pull quotes, or complex page designs, the reading order in the tag tree must reflect the logical flow. A screen reader follows the tag order, not the visual layout — if tags are in the wrong order, the content is incomprehensible.

Text and images of text must have a contrast ratio of at least 4.5:1 against their background (3:1 for large text, defined as 18pt or 14pt bold). Many government PDFs use low-contrast design elements — light gray text on white backgrounds, colored text on colored backgrounds, or text over watermarks — that fail this criterion.

Documents must have titles that describe their topic or purpose. In PDFs, this means the document Title field in the metadata properties must be set to a meaningful, descriptive title — not just a filename like "report_final_v3.pdf." The document should also be configured to display the title (not the filename) in the viewer title bar.

The default human language of the document must be programmatically determinable. In PDFs, the language must be set in the document catalog. Without a declared language, screen readers cannot determine which pronunciation rules to apply, resulting in garbled or unintelligible speech output for the user.

For all user interface components (form fields, links, buttons), the name, role, and value must be programmatically determinable. In PDF forms, every field must have a descriptive label, the correct form field type (text, checkbox, radio, dropdown), and its current value must be accessible to assistive technologies.

Automated Validation Tools

veraPDF is the industry-standard open-source PDF/UA validator. It checks structural conformance against ISO 14289-1 (PDF/UA-1) and reports every violation with specific rule references. PAC (PDF Accessibility Checker), developed by the Swiss accessibility foundation, provides a user-friendly interface with detailed reports and a screen reader preview. Adobe Acrobat Pro also includes a built-in accessibility checker, though it is less comprehensive than veraPDF or PAC.

Manual Testing with Screen Readers

Automated tools catch structural issues but cannot evaluate the quality of alt text, the logic of reading order, or the overall user experience. Testing with an actual screen reader (NVDA, JAWS, or VoiceOver) reveals how a real user will experience the document. Listen to the entire document being read aloud — does it make sense? Can you navigate by headings? Do table headers announce correctly?

Adobe Acrobat Accessibility Check

Adobe Acrobat Pro includes an accessibility checker under Tools > Accessibility > Full Check. This runs through common issues including document title, language, tags, reading order, alt text, and table structure. While not as thorough as veraPDF, it provides a good first-pass assessment and allows you to fix many issues directly within Acrobat.

Color Contrast Analysis

Use contrast checking tools like the Colour Contrast Analyser (CCA) to verify that text in your PDF meets the 4.5:1 ratio for normal text and 3:1 for large text. Pay special attention to text over colored backgrounds, watermarks, or images. Automated PDF checkers may not always catch contrast failures, so manual spot checking is important.