Tags say what a heading is, what a list is and what a table is. Without tags, a screen reader does not know this. You cannot see the difference on the page.
The easiest way: tags from the source file
The program you make the document in can add the tags. You do have to save the PDF the right way.
In Word for Windows
Choose Save As PDF, not Print to PDF. Under Options, turn on the document structure tags for accessibility.
In InDesign
When you export, turn on the option for a tagged PDF.
Even then, the PDF is usually not entirely right. So always check the result.
If you no longer have the source file
Then you add the tags to the PDF afterwards. In Acrobat that is manual work. Our tool does it like this:
- 1The tool reads the text on every page and turns it into headings, paragraphs, lists and tables.
- 2Tags that do not identify anything, such as the lines of a chart, are removed.
- 3A row only becomes a table header if that is certain. A header on the wrong row is worse than no header.
- 4The title, the language and the settings are set correctly.
A scan gets no tags
A scan contains no text, only an image of text. So there is nothing to put a tag around. First ask for the original file. Text recognition (OCR) can help, but it makes mistakes. Only use it if there is no other option.
After tagging
A few things no program can judge. You check those yourself:
- Is the reading order right?
- Is every heading at the right level?
- Does the alt text of each image describe what is on it?
The tool never fills in alt text by itself. If an image contains short text that it can read reliably, it suggests that text, and you decide whether to use it. The person who made the document knows what else is on the image.
Does your PDF have tags?
Check it for free, without an account.