How to Tag a PDF, in the Order That Works

Tags are how a PDF says "this line is a heading, this block is a list, this box is a table header". A sighted reader infers all of that from size and position. A screen reader cannot infer anything. It reads the tag tree, and if there is not one, it guesses from layout and usually guesses wrong.

There is good news. Build the source document properly and the export does most of the tagging for you. The work below is mostly about not throwing that away.

Step 1: Fix the Source, Not the PDF

Every minute spent repairing tags inside Acrobat is a minute you will spend again the next time somebody re-exports the file. Go back to the Word document or the InDesign file.

In the source, use real structure instead of visual imitation:

• Apply the built-in Heading 1, Heading 2, Heading 3 styles instead of making text big and bold. The style is what becomes a tag.
• Build lists with the list buttons, not by typing a hyphen and a space at the start of each line.
• Build tables with the table tool, and mark the top row as a header row. In Word that is Table Design, then Header Row, then in Table Properties, Row, tick "Repeat as header row at the top of each page".
• Put alt text on images in the source. In Word, right-click the image, then View Alt Text.
• Use the columns feature for columns. Never fake a column with the tab key or a text box, because the reading order will follow the boxes and not the eye.

Step 2: Export With Tagging Switched On

This is the step people miss. It undoes step one entirely.

Microsoft Word

File, then Save As, then choose PDF. Click Options and tick "Document structure tags for accessibility". On Mac, choose "Best for electronic distribution and accessibility". Do not use Print to PDF: it flattens the file to ink and throws every tag away.

Adobe InDesign

Set up the Articles panel to control reading order before you export. Then File, then Export, then Adobe PDF (Print) or (Interactive), and tick Create Tagged PDF. Map your paragraph styles to export tags first, under Paragraph Style Options, then Export Tagging, so an H2 style exports as an H2 tag.

Google Docs

File, then Download, then PDF Document. Google Docs exports basic tags from real heading styles, but it will not carry table headers or alt text reliably. For anything beyond a simple text document, finish the file in Acrobat Pro.

Step 3: Set the Title, Language, and Initial View

Three settings, all in Acrobat Pro under File, then Properties. They take about a minute and they are three of the eight failures we find most.

1. 1Description tab, Title field. Type the document's real name. "Spring 2026 Price List", not "final_v3_USE_THIS.pdf".
2. 2Advanced tab, Reading Options, Language. Set the document language. This is what tells a screen reader which voice to use.
3. 3Initial View tab, Show. Change it from File Name to Document Title. Without this, the title you just set never gets announced.

Step 4: Check the Reading Order

Open the Accessibility tools in Acrobat Pro and choose Reading Order. Acrobat draws numbered boxes over the page showing the order a screen reader will read in. Read the numbers in order. If they run 1, 2, 5, 3, 4, the document reads out of sequence.

Fix order in the Order panel by dragging entries up and down, or in the Tags panel for finer control. Pull sidebars, pull quotes, and captions to where they belong in the flow, which is usually right after the paragraph they relate to.

Step 5: Alt Text and Artifacts

Every image in the tag tree gets one of two treatments. There is no third.

Images that carry meaning get alt text: a chart, a photo of the product, a diagram, a signature. Write what the image tells the reader, not what it looks like. Our alt text decision helper walks any image to the right answer in four questions.

Images that carry nothing get marked as artifacts: rules, borders, background textures, decorative flourishes, the logo repeated in the page header. An artifact is skipped entirely, which is exactly right. In Acrobat, use the Reading Order tool, select the item, and click Background/Artifact.

Step 6: Tables and Form Fields

Tables need header cells marked as headers, not just styled to look like them. In Acrobat's Reading Order tool, select the table, click Table, then use the Table Editor to set each header cell's type to Header and give it a scope of Row or Column. A data cell that does not know its header is a number with no meaning attached.

Form fields need names. In Acrobat's Prepare Form tool, open each field's properties and fill in the Tooltip field with the label the user sees on the page. That tooltip is what a screen reader announces. An unnamed field is read as "edit text, blank", which tells the user nothing about what to type.

Step 7: Verify, Then Have a Person Read It

Run Acrobat's Accessibility Check, which reports against its own rule set, and then run our checker or PAC for a second opinion against PDF/UA. Two checkers disagree often enough to be worth the minute.

Then read the document with a screen reader, or better, watch somebody who uses one every day read it. That is the step that finds the heading tagged as body text and the alt text that describes the wrong thing.