Imagine handing someone a beautifully printed report, except before you hand it over, you feed it through a shredder, shuffle the strips, and tape them back together in random order. The words are all still there. Good luck reading it.

That is what most PDFs are like for someone using a screen reader. The document looks perfect on your screen, but the software that reads it aloud to a blind or low-vision person receives something closer to word salad: fragments in the wrong order, images with no explanation, tables reduced to a stream of disconnected numbers.

The difference between the two experiences comes down to one thing: whether the file is a tagged PDF. Tags are the single most important and most misunderstood concept in document accessibility, so let’s take it from the top, in plain English.

What Is a Tagged PDF, Exactly?

A tagged PDF contains an invisible layer of labels, called tags, that tell software what each piece of the document actually is. A tag says “this line is a heading,” “this block is a paragraph,” “this is a picture, and here is what it shows,” “this is a table, and here are its rows and columns.”

Think of tags as the skeleton under the skin. You never see a skeleton, but it is what gives a body its structure. Without one, you just have a shapeless pile that happens to look right from the outside.

Here is the part that surprises most people: a PDF without tags still looks completely normal. You can open it, read it, print it, and never suspect a thing. That is exactly why the problem is so widespread, the file only breaks for the people who depend on software to read it, and they usually have no way to tell you.

What a Screen Reader Actually Hears

A screen reader is software that reads what is on the screen out loud. It cannot see your document the way you do, it can only work with the structure the file gives it. Give it a tagged PDF and it performs beautifully. Give it an untagged one and it does its best with a pile of loose text.

The difference is not subtle. In a tagged document, a screen reader user can pull up a list of every heading and jump straight to the section they need, the same way you skim with your eyes. In an untagged document, there are no headings to jump to. The only option is to listen from the top and hope, like being forced to read a book with no chapters, no page numbers, and no table of contents.

And that word “hope” is doing heavy lifting, because untagged files often read in the wrong order too. The visual layout, columns, sidebars, pull quotes mean nothing without tags. A two-column newsletter might read straight across both columns, splicing two unrelated sentences together mid-thought.

The Tags That Do the Heavy Lifting

You do not need to memorize the full tag vocabulary, but knowing the main players makes everything else about PDF accessibility click into place:

• Heading tags (H1 through H6) create the outline of the document. H1 is the title, H2s are major sections, H3s live inside them. These are what let readers navigate instead of wading.
• Paragraph tags (P) mark ordinary body text, so it reads as complete, connected thoughts.
• Figure tags mark images and carry the alternative text, the short written description a screen reader speaks in place of the picture. Without it, the reader hears “image” and nothing more.
• Table tags preserve the relationship between headers and cells. With them, a reader hears “Quarter 2 revenue: $40,000.” Without them, it hears “$40,000” with no clue what the number means.
• List tags announce “list of five items” and keep the items grouped, instead of scattering them as stray lines.
• Form field tags give every box a spoken name, “Last name,” “Date of birth”, so people can fill out your form without guessing.

Reading Order: The Other Half of the Story

Tags say what things are. Reading order says what comes next, and it matters just as much. PDF reading order is the sequence assistive software follows through the page, and it does not automatically match what your eyes do.

Your eyes handle a complex layout effortlessly: main story first, then the sidebar, skipping the page number at the bottom. Software has no such instincts. If the reading order is wrong, a screen reader might announce the footer in the middle of a sentence, read a photo caption before the story it belongs to, or dive into a sidebar halfway through the main article.

This is why documents with multiple columns, text boxes, and decorative layouts are the most common accessibility casualties. The fancier the design, the more the reading order needs a human to check it.

How to Tell If Your PDF Is Tagged

Here is a check that takes fifteen seconds. Open the file in Adobe Acrobat (the free Reader works), go to File, then Properties, and look at the Description tab. Near the bottom you will see a line that says Tagged PDF: Yes or Tagged PDF: No.

A “No” is a definite problem. But (and this trips up almost everyone) a “Yes” is not a guarantee. It only means tags exist, not that they are correct. Software can stamp a document with automatic tags that mislabel headings, skip images, and scramble tables. It is the difference between a building having a fire exit sign and the door behind it actually opening. We wrote a whole piece on why automated checkers miss the most important errors for exactly this reason.

One more quick test: try selecting all the text in the PDF and copying it into a blank document. If it comes out in a scrambled order or nothing copies at all, you are likely looking at an untagged or scanned file.

How to Tag a PDF (Start Earlier Than You Think)

The best place to fix a PDF is before it becomes a PDF. Most documents start life in Word, Google Docs, or InDesign, and the choices you make there decide almost everything:

• Use real heading styles (Heading 1, Heading 2) instead of just making text big and bold. Styles become tags when you export; bold text becomes nothing.
• Add alt text to images in the source document, where it takes seconds.
• Build tables with the table tool, never by lining text up with tabs and spaces.
• Export correctly in Word, use “Save As PDF” with accessibility options enabled, not “Print to PDF.” Printing to PDF throws the structure away, like photocopying a document and losing the original.

For PDFs that already exist, the tags get repaired directly in Acrobat Pro, checking each one, fixing the reading order, writing alt text, and rebuilding table structures by hand. For a short, simple document, a patient person can learn to do this. For long documents, forms, or anything with tables and complex layouts, it is genuinely skilled work, and it is what our manual PDF remediation service does all day.

Scanned documents are their own category: a scan is just a photograph of a page, with no text in it at all. Those need text recognition before tagging can even begin, we cover that whole journey in turning scanned legacy PDFs into accessible documents. And if you are wondering what professional remediation costs, we publish real numbers in our PDF remediation cost guide.

Does the Law Actually Require Tags?

In practice, yes. No law contains the sentence “thou shalt tag thy PDFs,” but every major accessibility requirement is impossible to meet without them. Section 508 requires federal documents to be accessible, and tags are how a PDF becomes accessible. The Department of Justice rule for state and local governments explicitly covers documents on government websites; PDFs count under Title II, with deadlines already on the calendar. Courts in ADA lawsuits treat inaccessible documents the same way they treat inaccessible web pages.

So if your organization publishes PDFs, forms, reports, notices, menus, meeting minutes, tagging is not optional extra credit. It is the baseline. And if you are wondering exactly which standard your documents need to satisfy, we break down PDF/UA vs WCAG vs Section 508 in plain English.

Frequently Asked Questions

Are PDFs automatically tagged when I create them?

Sometimes, partially. Exporting properly from Word or InDesign carries much of the structure over, if the source document was built with real styles. “Print to PDF,” scanning, and many older tools produce no tags at all. And even a good export deserves a human check, because automatic tags are frequently wrong in ways software cannot detect.

Does a tagged PDF look different from a regular one?

No, and that is the point. Tags are invisible to sighted readers. The document looks identical before and after tagging. The only thing that changes is that everyone can read it.

Can I tag a scanned PDF?

Not directly. A scan contains a picture of text, not actual text. It first needs OCR (optical character recognition) to convert the image into real, readable text, then that text can be tagged and the reading order fixed. Older scans with faded print or handwriting usually need a professional pass to get right.

Find Out What Your Documents Are Really Saying

Every organization has a folder of PDFs it hands to the public, and almost nobody knows how those files sound to the people who cannot see them. That is fixable. Our PDF accessibility workspace walks through exactly how we take documents from untagged to fully accessible, with humans checking every tag.

If you want to know where your documents stand, send us one — the free audit takes minutes to request, and our pricing is public, so there is nothing to negotiate and nothing to dread. Your PDFs already say something to every reader. Tags just make sure they say the same thing to all of them.