PDF metadata, and what it says about you
Every PDF carries a small dossier about how it was made. Most of it is harmless, some of it is a name and a timestamp you did not mean to send, and clearing the properties panel does not remove it.
What does a PDF store about its author?
A short list of fields, filled in by whatever produced the file. Title, author, subject and keywords come from the document; creator names the program the content was written in, producer names the one that wrote the PDF, and two timestamps record when it was created and last changed.
None of that is on the page, and all of it opens in any reader’s document properties. The author field is usually the account name of whoever installed the software, which is how a submission meant to be anonymous arrives with a full name attached.
Why does clearing the properties not always work?
Because a PDF stores those facts twice. Alongside the old information dictionary, files since PDF 1.4 carry an XMP packet: the same fields as XML, plus whatever custom schemas the producing application added. Most current readers show the XMP copy.
Editing one copy therefore leaves the other on display, which is why a file whose author you just cleared still shows the old name. PDF 2.0 settled the question by deprecating most of the old dictionary in favour of XMP and keeping only the creation and modification dates, but files in circulation still carry both, and anything that claims to remove metadata has to remove both.
What else travels with the file?
More than the properties panel shows. Saving over a PDF often appends the change rather than rewriting the document, so the previous version can still be sitting inside the same file. Attachments live inside a PDF too, comments keep their authors and times, and bookmarks keep names from whatever the file used to be.
Images bring their own history. A photograph placed on a page keeps the EXIF block the camera wrote, including the model and, on a phone, the coordinates where it was taken. Nothing on the page hints at any of it.
How do I check and strip it before sharing?
Read it first. Any reader’s document properties shows the visible fields, which is enough to catch a name or an internal project code. Then strip the file rather than editing the fields one at a time, so both copies go together.
Where a file has a history you would rather not send, rewrite it completely instead of saving over it: a full rewrite drops earlier revisions rather than carrying them along. Strip last, because most tools stamp their own producer line on the way out.
Where to go from here
Remove PDF metadata clears both copies at once, Edit PDF metadata sets the fields you do want a reader to see, and Redact PDF is for content on the page rather than facts about the file.
Redaction that actually removes text covers the page itself, and what leaves your device answers the question a stripped file raises next.