PDF Accessibility — Missing PDF/UA Identifier and XMP Metadata

Who Is Affected

No reader is directly affected. This is a conformance-declaration issue, and it matters to a different audience: automated validators, accessibility auditors reviewing your document library, and procurement or grant reviewers who ask for evidence that your documents meet a standard.

It is worth being straightforward about this. Most items in an accessibility report describe a barrier that stops somebody using your document. This one does not. We report it because PDF/UA requires it and validators fail documents without it — not because a constituent is being excluded.

What This Means

Two separate pieces of metadata are involved, and a document can be missing either or both.

The PDF/UA identifier is a flag inside the document's XMP metadata that says "this file claims conformance with PDF/UA-1". It is an assertion by whoever produced the document. A validator reads it to decide whether to hold the file to the PDF/UA standard at all — without it, a perfectly tagged, fully accessible PDF is simply not a PDF/UA document, because it never claimed to be one.

XMP metadata is the standardised metadata packet that carries the title, language, and conformance claims. Some tools write document properties only into the older document-information dictionary and never into XMP. The properties then appear correct in Acrobat's dialog while remaining invisible to validators and to some assistive technology.

The important thing about both: adding them does not improve accessibility. Declaring conformance on a document that is untagged makes it a document that falsely claims conformance — worse than one that claims nothing, because a reviewer now has grounds to question everything else you have asserted. Fix the structural problems first. Add the declaration when the document actually conforms.

Fix: Document

Do these in order, and only reach step 3 once the document genuinely passes.

1. Resolve the structural findings first. Tags, reading order, alternative text, document language, document title, table headers. If your report lists any of those for this file, they come before this one. Start with untagged document if it appears.
2. Verify with a validator rather than by eye. Run veraPDF or PAC 2024. Both report the PDF/UA identifier separately from the structural rules, so you will see exactly what remains.
3. Add the identifier and XMP metadata.
   • In Adobe Acrobat Pro, the accessibility Full Check followed by saving as PDF/UA writes both. Alternatively use Preflight → "Convert to PDF/UA-1", which validates and stamps in one operation.
   • If the PDF is generated from Word, save via Acrobat's Create PDF add-in with "Enable Accessibility and Reflow" selected, rather than Word's Print-to-PDF, which discards XMP.
   • For a library of documents, do not do this by hand. Acrobat Pro Action Wizard can apply the same step across a folder; for larger volumes a server-side tool that writes XMP directly is the sensible route.
4. Re-validate. The declaration is only meaningful if the validator agrees with it.
5. Record what you did. For Title II and Section 504 purposes, a dated record of remediation and validation is part of demonstrating good-faith effort. A conformance claim you cannot evidence is not worth making.

Standard Reference

PDF/UA-1 (ISO 14289-1) — Clause 5, Conformance

A conforming file shall contain XMP metadata identifying the PDF/UA conformance level. The identifier is a claim of conformance and does not itself create it.

PDF/UA is the document counterpart to WCAG for the web. Both the DOJ Title II rule and the HHS Section 504 rule cite WCAG 2.1 Level AA as the technical standard; PDF/UA is how that standard is applied to PDF documents in practice.

• PDF Association — PDF/UA in a nutshell
• WCAG Technique PDF18 — Specifying the document title using the Title entry
• veraPDF — the reference PDF/UA validator