PacketReady

Tagged PDF, explained

Tags are an invisible outline of a PDF. Screen readers depend on them to know what is a heading, a list, a table or an image. Here is what they are, how to tell whether a file has them, and how to add them.

Reviewed September 23, 2026.

What tags are

A PDF page is essentially a set of drawing instructions: put these letters at these coordinates in this font, draw this line, place this image. Nothing in those instructions says "this is a heading" or "these cells form a table". A person looking at the page works that out from the layout. Software cannot.

Tags add that missing meaning. They form a separate tree that labels the content in reading order:

  • H1 to H6 for headings, and P for paragraphs
  • L, LI, Lbl and LBody for lists and their items
  • Table, TR, TH and TD for tables, rows, header cells and data cells
  • Figure for images, carrying their alternate text
  • Link for links, and Form for form fields

Things that are not part of the content, such as running headers, page numbers and decorative lines, are marked as artifacts so that assistive technology skips them.

Inside the file, a tagged PDF has a structure tree (the StructTreeRoot entry in the document catalog) and a MarkInfo dictionary whose Marked flag is true. Each piece of page content is linked to its tag through a marked-content identifier. PacketReady's Tagged PDF check looks for all three, and also checks that the text on each page is actually inside tags or marked as an artifact.

Why tags matter

  • Reading order. Screen readers follow the tag order, which is how a two-column page or a sidebar is read in the right sequence.
  • Navigation. Screen reader users jump by heading, list or table, the way sighted readers scan a page.
  • Tables. Header cells let a listener hear "Amount, $1,000" rather than a string of numbers.
  • Images. Alternate text lives on Figure tags; an untagged PDF has nowhere to put it.
  • Reflow. Features that re-wrap text for small screens and magnification, such as Acrobat Reader's Reflow view, rely on tags.

In WCAG terms, tags are how a PDF meets success criterion 1.3.1, Info and Relationships, and 1.3.2, Meaningful Sequence. W3C techniques PDF9 (headings), PDF6 (tables) and PDF21 (lists) show the details.

How to tell whether a PDF is tagged

  • Run it through PacketReady: the Tagged PDF check reports untagged files, empty tag trees, and pages whose text is outside the tags.
  • In Adobe Acrobat or Acrobat Reader, open Document Properties; the Description tab shows "Tagged PDF: Yes" or "No".
  • In Acrobat Pro, open the Tags panel to see the tree itself.

"Tagged: Yes" only means that tags exist. A packet made by combining a tagged agenda with scanned or untagged attachments will say it is tagged while whole sections are invisible to a screen reader. That is why PacketReady checks the text on every page.

How tags get into a PDF

Best: from the source document

Tags are most accurate when the program that created the document writes them. Use real heading styles, list buttons and tables in Word or Google Docs, map paragraph styles to tags in InDesign (Paragraph Style Options > Export Tagging), and export with tagging turned on. In Word for Windows that is the "Document structure tags for accessibility" option (Microsoft support).

Next best: automatic tagging

Acrobat Pro can add tags to an untagged file (All tools > Prepare for accessibility > Automatically tag PDF, according to Adobe's help). It guesses the structure from the layout, so review the result: headings detected as paragraphs, tables split or merged, and decorative items tagged as content are all common.

Last resort: tagging by hand

Acrobat Pro's Tags panel and reading order tool let you build or correct the tree element by element. It is precise but slow, which is why fixing the source template is usually the better investment.

Common tagging mistakes

  • Everything is a paragraph. The file is tagged, but visual headings are P tags, so there is nothing to navigate by.
  • Fake tables and lists. Columns lined up with tabs, or bullets typed as characters, come out as plain paragraphs.
  • Wrong reading order in multi-column layouts, text boxes and sidebars.
  • Untagged pages after combining files, and lost tags after printing to PDF.
  • Artifacts that carry information, such as a footer containing the only statement of the meeting date.

Tags and PDF/UA

PDF/UA (ISO 14289-1) is the international standard for accessible PDF. It requires tags plus a set of other properties: a displayed document title, a document language, alternate text for figures, header cells in tables, and more. The Matterhorn Protocol lists every way a file can fail it. Some files carry a PDF/UA identifier in their metadata; that is a claim by the software that produced them, not proof, so check the file anyway.

Check your own files

PacketReady checks a whole packet folder in your browser, with nothing uploaded, and explains every fix in plain English.

Check PDFs now Get packets remediated

Related guides: How to check a PDF ยท Accessible board packets

Sources

  1. PDF Techniques for WCAG 2.1 (PDF3, PDF6, PDF9, PDF21). W3C
  2. The Matterhorn Protocol 1.1. PDF Association
  3. Create and verify PDF accessibility (Acrobat Pro). Adobe
  4. Create accessible PDFs. Microsoft Support