Question 1

What is PDF metadata?

Accepted Answer

The descriptive information a PDF carries about itself: the title, the author, the subject, the keywords, the application that made it and the dates it was created and changed. It is not part of the visible page — you cannot see it when you open the file — but every reader shows it in the document properties panel, and a search engine may read it. It is also how a document names the person who made it, which is the thing people usually want gone.

Question 2

Does this remove everything, or only some of it?

Accepted Answer

Everything, and from both places it can hide. A PDF keeps its metadata in two independent stores: the Info dictionary in the trailer, and an XML packet called XMP hanging off the document catalog. A file written by a modern tool usually has both, and they routinely disagree — one measured file had a title in the XMP and no title at all in the dictionary, and its author's name in both. The tool reads both, names both in the report, and empties both.

Question 3

Are my files uploaded to a server?

Accepted Answer

No. Every byte stays on your machine — the file is read and rewritten in the browser tab, and the copy you download is built there too. That matters more here than it does for a converter: a metadata remover is the one PDF tool where the file you are handing over may be a contract, a payslip or a medical record, and sending it to a server to be cleaned means it now exists on somebody else's disk.

Question 4

What is the XMP packet, and why does it matter?

Accepted Answer

It is a block of XML that sits inside the PDF and carries the same facts in a modern form: dc:creator for the author, xmp:CreatorTool for the application, xmp:CreateDate and xmp:ModifyDate for the timestamps, plus a document UUID. On the one-page file this was measured on, the Info dictionary came to 283 bytes and the XMP packet to 1,748 bytes — six times the size of the dictionary it duplicates — and it held a second copy of the author's name. A cleaner that only empties the dictionary leaves that name in the file.

Question 5

Can I keep the document title?

Accepted Answer

Yes. Choose "Keep the title, remove everything else" and the title stays in the document properties, which is what a screen reader announces when it opens the file and what a search engine reads. The XMP packet is still removed in full, because it carries a second copy of the title as well as the author, and hand-editing XML metadata is how files get broken. The report says so rather than quietly leaving your name behind.

Question 6

Will removing the metadata damage the PDF?

Accepted Answer

No. The document is edited in place rather than rebuilt, so the pages, the bookmarks, the page labels, the links and any embedded files all survive exactly as they were. The pages still render and the text is still selectable and searchable, because nothing about the content is touched. The clean copy is also written with a plain cross-reference table, so it opens in old readers as happily as in new ones.

Question 7

Can it read a password-protected PDF?

Accepted Answer

No, and it says so rather than guessing. The metadata of a protected file cannot be read without the password, which the browser does not have. Remove the password first, or print the file to a new PDF, and then this will work on it.

Question 8

What if the file has no metadata at all?

Accepted Answer

The report says exactly that — no Info dictionary, no XMP packet, nothing to remove — and the file is handed back byte for byte rather than being rewritten for no reason. It happens more often than you would expect: a PDF printed straight from a browser often carries nothing at all.

Where it is	What it holds	What the tool does
The Info dictionary	Eight standard fields in the trailer: Title, Author, Subject, Keywords, Creator, Producer, CreationDate and ModDate. 283 bytes on the file this was measured on.	Every field is read, named in the report with its value, and deleted. A title can be kept on request.
The XMP packet	The same facts again, in XML: dc:creator, xmp:CreatorTool, xmp:CreateDate, xmpMM:DocumentID. 1,748 bytes on the same file.	Removed in full. It is six times the size of the dictionary it duplicates, and it holds a second copy of the author.
Both at once	A PDF written by a modern tool usually has both stores, and they disagree. One measured file had a title in the XMP and none in the dictionary.	The report shows the two side by side, so the disagreement is visible instead of guessed at.
Inside an object stream	A compressed container that can hold several objects at once, which is where some producers put the Info dictionary.	Checked rather than assumed: the container is expanded on load, so the dictionary is deleted from the bytes and not merely unhooked from the trailer.
The page content	The text, the images and the pages themselves — the part you can see.	Never touched. Pages, bookmarks, page labels, links and embedded files all survive untouched.
What is left	A document with no Info dictionary and no XMP packet, and nothing else changed.	Checked afterwards with independent readers: the pages still render and the text is still selectable.

Remove PDF Metadata Online

About the Remove PDF Metadata Online

How to use the Remove PDF Metadata Online

Where a PDF actually hides its metadata

Frequently asked questions