PDF structure tags
Silktide checks that your PDF documents contain accessibility tags. Tags describe the document's structure - headings, paragraphs, lists, tables - so software can read it as more than a picture of a page.
A carries a hidden structural layer, similar to the of a web page, telling what each piece of content is and in what order to read it. An untagged PDF is just ink positioned on paper: the words may be extractable, but their structure and are guesswork.
Why this matters
Without tags, a has to guess how to read the document: multi-column layouts get read across columns, tables collapse into meaningless word streams, headings are indistinguishable from body text, and images have no descriptions. In practice many untagged PDFs are simply unusable without sight. Tagging is the foundation every other PDF accessibility feature builds on - reading order, headings, and all live in the tags.
How to fix it
- The best fix is at the source: re-export the PDF from its original
document with tagging enabled.
- In Microsoft Word, use built-in styles for headings and lists, then export with "Document structure tags for accessibility" checked.
- In Adobe InDesign, set up the article structure and export as a tagged PDF.
- If the source is gone, add tags in a PDF editor such as Adobe Acrobat Pro using its accessibility tools, then set the reading order, tag the headings, and add alternative text to images.
- For scanned documents, run text recognition () first - a scan is an image, and tags need real text underneath.
How Silktide tests this
- Open each PDF document found on your website.
- Check whether the document declares a structure tag tree.
- Report the document if it has no tags at all.
Note that this check tests for the presence of tags, not their quality - a tagged document can still have problems, which other PDF checks and manual review cover.