How to tag a PDF on a Mac

To tag a PDF on a Mac, first try exporting a tagged file from the source document. If you only have the PDF, open it in a tool that can edit the structure tree, generate or correct the tags, check the reading order against the page, and then validate a separate output. The steps below follow that order, because each one saves work in the next.

What tags do, and why Preview is not enough

Tags are a hidden outline stored inside the PDF. They say which text is a heading, which cells are table headers, what order the content should be read in, and which images have a description. A screen reader uses that outline rather than the look of the page, so a PDF can appear perfect and still be unusable.

The Preview user guide covers viewing, annotating, signing, and combining files. It describes no way to create or edit tags. Saving or exporting a PDF from Preview is therefore not a remediation step, and you should check any file it has saved, because you cannot correct its structure there.

Step 1: Check whether you can fix the source

If you own the Word, Pages, InDesign, or similar source, start there. Set real heading styles, real lists, table header rows, the document language, and alternative text, then export a tagged PDF. On a Mac, check the exact export path for your version, because some export routes produce a better structure than a generic print to PDF. See the Word guide, and test what comes out.

A corrected source also means that the next edition starts accessible. Work done only on the PDF is lost as soon as someone exports a new version.

Step 2: Choose a tool that edits structure

Step 3: Generate tags, then check the order

Automatic tagging gives you a draft. Read the proposed order next to the page and look at the places where it usually goes wrong: columns, sidebars, captions, footnotes, and anything placed in a text box. Our guide to making a PDF accessible explains what to look for. Repeated page headers, footers, and page numbers should usually be marked as artifacts so that they are not read out on every page, as the artifact guide describes.

1. Confirm there is one logical heading structure, with no skipped levels that hide meaning.
2. Check that lists are tagged as lists, not as paragraphs that begin with a dash.
3. Set table header cells and scope, and split or join tables only where the content supports it.
4. Write alternative text for each meaningful figure, and mark decoration as an artifact.
5. Set the document title and language, and tag any passage in another language.

Step 4: Handle forms, links, and scans

Every form field needs a label that a screen reader announces and a sensible tab order. Links should say where they go. A scanned page has no text at all until it has been through OCR, so it cannot be tagged usefully first. Run recognition, correct the recognized lines, and only then tag the result. On a Mac, Apple's Vision framework can do that recognition on the device.

Step 5: Validate, then test with a person

Run a PDF/UA check on the output with a tool such as veraPDF, which the Mac validation guide covers in detail. A clean machine result does not prove that a heading is the right level or that a description is accurate. Finish by reading the document with VoiceOver on the Mac, and with JAWS or NVDA on Windows if your readers use them, because viewers and screen readers do not all behave alike.

Frequently asked questions

Sources

The statements above rest on the primary sources below. Where a source is a law or standard, the version and date named in the text are the ones checked on the review date. This guide is general information, not legal advice.

1. Apple: Preview User Guide for Mac . What Preview documents as its PDF features.
2. Apple: VoiceOver User Guide for Mac . The screen reader built into macOS.
3. Adobe: Create and verify PDF accessibility . Acrobat's documented tagging and checking workflow.
4. veraPDF . An open-source PDF/A and PDF/UA validator.