How to check a PDF for hidden data before you send it

A PDF looks like paper, which is why people trust it. Underneath, it is a container that can hold a great deal the page does not show.

1. The document properties

Every PDF can state its title, author, the program that created it and the dates it was created and changed. The author is often the name on your computer account. The “creator” can give away which template or internal system the file came from. A second, longer block of the same kind of information, called XMP, usually sits beside it.

2. Earlier versions

When a PDF is edited and saved, many programs do not rewrite the file. They add the changes to the end and leave the old content in place. The reader shows only the newest version, but the old one can be recovered. A figure you corrected or a paragraph you deleted may still be there.

This is also why “removing” properties with a quick edit is not enough: the old values stay in the earlier layer.

3. Comments and markup

Sticky notes, highlights and text comments carry the name of the person who made them and the time. A comment collapsed to a small icon is easy to miss on screen.

4. Attachments

A PDF can have other files attached to it: the spreadsheet behind a chart, or the original Word document. Readers tuck these away in a side panel.

5. Invisible text

Text can be white on white, set in a size too small to see, placed outside the edge of the page or on a layer that is switched off. Scanned documents often have an invisible layer of recognised text behind the picture of the page, which is useful for searching but means the words can be copied even if the picture is hard to read.

6. Form fields and scripts

A filled-in form keeps its values. A PDF can also contain small programs that run when it is opened.

7. Black boxes that are not redaction

A black rectangle drawn over a name hides it from the eye and nothing else. The text is still under it and can be selected, copied or searched. This has exposed names in court filings and government reports more than once.

How to check and fix

1. Run Find hidden content on the file. It reports white, tiny and off-page text, hidden layers, attachments, scripts, form values and comments, without changing anything.
2. Run Remove metadata to drop the properties and rewrite the whole file, so earlier versions and leftover objects are gone.
3. If something on a page must not be readable, use Redact PDF. It replaces the page with a picture that has the box burned in, then reads the result back to confirm the words are gone.
4. Open the final file and look at it once more before sending.