PDF/UA pre-check — accessible document
Since June 2025 the European Accessibility Act requires digital services — including published PDF documents — to be accessible to people with disabilities. In practice the reference standard is PDF/UA (PDF Universal Accessibility): a tagged document that screen readers can render. This pre-check reviews the points that actually block assistive reading: document tagging (without structure, a screen reader only sees a stream of glyphs), the declared language (otherwise speech synthesis picks the wrong pronunciation), the title and how it is displayed, the heading hierarchy, figure alternative texts, table headers and tab order. Each checkpoint returns a verdict — pass, deviation, warning — with the fix to apply. Metadata deviations (title, language) can be corrected right here; an untagged document must be regenerated from its source. The analysis runs in your browser: a medical file or a confidential report never leaves your device. The verdict is an indicative pre-check: full conformance is validated with a reference checker like veraPDF. The check runs on your file’s actual structure — not a generic checklist — so the report reflects what a screen reader would really find in this document. When the title or the language attribute is missing, the tool can write it for you directly, which fixes two of the most common PDF/UA failures in one click.
FAQ
Does my document need to be PDF/UA?
If you publish documents for the public or for customers in the EU — administrations, banks, online services — the European Accessibility Act (in force since June 2025) requires it. PDF/UA is the technical standard that implements it.
How is this different from veraPDF?
veraPDF is the reference validator: exhaustive and formal, but technical and barely actionable. This pre-check aims at fast triage: it flags what blocks assistive reading and states the fix, in plain language, without installation.
Can an untagged PDF be fixed?
Not automatically — and any tool claiming so produces dishonest tagging. Tagging is created at generation time (“tagged PDF” export in Word, InDesign, LibreOffice). Here, only title, language and title display are corrected.
Is my document uploaded anywhere?
No. The analysis runs in your browser: no file, no extract, no metadata ever leaves your device.
Is a scanned PDF accessible?
A picture of text can be read by no one: run the scan through OCR first (selectable text), then tag it — OCR alone does not create the structure PDF/UA requires.
What is a tagged PDF?
A PDF whose content carries a logical structure tree — headings, paragraphs, lists, figures — so assistive technology can navigate it instead of reading raw painted glyphs.
Can the tool fix everything it reports?
Missing title and language can be written directly here; tagging and logical reading order must come from the authoring tool — the report tells you where the source needs rework.
Does accessibility work help search engines?
Yes — tagged structure, declared language and a proper title also help search engines and AI tools parse the document; accessibility work doubles as machine-readability work.
Who should run this pre-check?
Anyone publishing PDFs for the public in the EU — administrations, banks, online services — before release: it triages what blocks assistive reading and what can be fixed on the spot.
What does the report look like?
A checkpoint-by-checkpoint verdict — pass, deviation, warning — each with the concrete fix to apply, so you act on what actually blocks assistive reading.
How to proceed
- Drop the PDF to publish or audit
- Read the checkpoint-by-checkpoint report — each deviation states its fix
- Fix title and language on the spot if needed, or regenerate the tagged document from its source
- Validate the final document with veraPDF before official publication