How to add alt text to images in a PDF

Alt text in a PDF is added figure by figure. In PDF by Zaweba you open the accessibility tools, go through the document's figures one at a time, and type a short description for each, or mark a figure as Decorative if it carries no meaning. A screen reader then reads your description out where the picture sits.

Alternate text is the words that stand in for a picture when someone can't see it. A blind reader using a screen reader hears it in place of the image. Without it, a chart that carries the main finding of a report is just silence, or the word "figure".

Before you start

Alt text attaches to figures, and figures come from tags. If the document has never been tagged, run Tag the document automatically first. It wraps untagged content in tags and marks pictures as figures, along with paragraphs, headings, links, form fields and comments. Once that is done, every picture has a place to hold its description.

A scanned document is a single picture per page, so it needs OCR and tagging before individual figures can be described.

Describing each figure

Open the PDF and go to the alternate text tool in the accessibility section. It works figure by figure, so you deal with each picture in turn and type its description.

For each figure, decide first whether it means anything. A photo of the product being discussed does. A decorative border, a divider line or a background flourish doesn't, so mark it Decorative and move on. That keeps the screen reader from announcing clutter.

For the ones that matter, write the description. Keep it to a sentence or two in most cases. Then move to the next figure until you reach the end.

Finally, run the check under Tools > Accessibility. Alternate text is one of the things its 32 rules look at, so it will catch any figure you passed over. Check again after fixing and the count should fall to nothing.

Writing good alt text

Describe what the picture is there to say, not every detail of it. For a chart, give the finding: "Sales rose each quarter, with the biggest jump in the third" tells a listener far more than a list of colours and bars. For a photo of a person, say who they are if that is the point.

Don't start with "image of" or "picture of". The screen reader already announces that it's a figure, so those words are wasted.

Leave out text that is already in the body. If the caption below the picture says what it shows, the alt text can be brief and add only what the caption doesn't.

Be consistent. When the same logo appears on every page, mark it Decorative everywhere rather than describing it thirty times.

Doing it as part of a bigger job

If you are making the whole document accessible, Make accessible runs the built-in action, then takes you to alternate text and then to the check, so describing figures sits in the right place in the sequence. The setup assistant guides the same steps for first-timers. While you are there, set the document language and title, since those are checked too.

PDF by Zaweba does all of this offline, so documents with private or unpublished pictures never have to leave your PC. Describing pictures well takes a little thought, but it is the step that makes the most difference to a reader who can't see them.