If you've been researching PDF accessibility, you've likely seen both "WCAG 2.1" and "PDF/UA" cited as the requirement — sometimes on the same page, sometimes contradicting each other. They're not interchangeable, and understanding the difference will save you a lot of confused searching.
WCAG 2.1: the general web accessibility standard
The Web Content Accessibility Guidelines (WCAG) are the internationally recognised standard for accessible digital content of any kind — websites, web apps, and documents. WCAG 2.1 at Level AA is the version referenced by most accessibility law in Europe, including the standard behind the European Accessibility Act. It's written as general principles (perceivable, operable, understandable, robust) that apply whether you're building a webpage or a PDF.
PDF/UA: the PDF-specific technical standard
PDF/UA (ISO 14289-1, "Universal Accessibility") is a separate ISO standard that specifies exactly how a PDF file needs to be technically structured to be accessible: a proper tag tree, correct reading order, alternative text on figures, table headers, a declared document language, and dozens of other concrete, machine-checkable rules. Where WCAG says "content must be perceivable by assistive technology," PDF/UA says precisely what that means in PDF's internal object structure.
How they fit together
The EU's harmonised standard, EN 301 549, is what the European Accessibility Act actually points to legally. For documents, EN 301 549 aligns with WCAG 2.1 AA success criteria, and for PDF specifically, conforming to PDF/UA is the recognised, practical way to meet the PDF-related requirements. In short: PDF/UA conformance is how a PDF satisfies most of the WCAG-derived legal requirement. A PDF/UA validator (like the veraPDF engine behind our checker) is checking machine-verifiable technical structure — it's the right tool for PDFs specifically, not a general WCAG website audit tool.
What automated tools can and can't verify
Being straight about this matters. A PDF/UA validator can definitively check things like: does a tag tree exist, is there a declared title and language, do images have alt text present, are table headers marked. What it cannot fully verify — and no automated tool can — includes whether reading order matches the visual layout in every edge case, or whether alt text is actually meaningful rather than just present. Those need a human judgement call. A tool that claims to fully automate WCAG compliance for PDFs is overstating what's technically possible today.
General information, not legal advice.
← Back to all guides