How to remove metadata from a PDF

Every PDF you create carries a silent passenger: metadata. Your name, the software you used, the exact time you saved it, sometimes the full file path on your machine including your username. Open a PDF you sent last month and check its properties — you’ll probably be surprised what’s in there. This guide covers what metadata is actually stored, when stripping it matters, what to strip, and how to verify you got it all.

What’s actually hidden in a PDF

The standard metadata fields in every PDF are Title, Author, Subject, Keywords, Creator (the application that made the source document — e.g. “Microsoft Word”), Producer (the library that produced the PDF — e.g. “Acrobat Distiller 15.0”), Creation Date, and Modification Date.

On top of that, PDFs can contain XMP metadata (a more extensible block), embedded file attachments, form field defaults with your name still in them, and occasionally — this is the one that catches people — the full file path of the source document. Export a Word doc from C:\Users\jsmith\Desktop\confidential.docxand that path can end up in the PDF’s XMP block.

When it matters: journalism and source protection

If you’re a journalist and a source sends you a leaked document as a PDF, the metadata may identify them before you’ve even read the contents. The Reality Winner case is the textbook example — printer tracking dots and metadata helped narrow the leak to one person. Always strip and re-export before publication, and ideally before opening the file on a networked machine.

When it matters: job searches and anonymization

You export your resume from Word. The Author field reads “Jane Doe” and the Creator reads “Microsoft Word 2019 for Mac.” Fine. But if you’re submitting anonymous writing samples, pseudonymous portfolios, or applying to a company where you already work and want a quiet parallel search, that metadata breaks the anonymization you thought your filename provided.

When it matters: litigation and discovery

In legal discovery, metadata is evidence. Modification dates can contradict claims about when a document was authored. Author fields can suggest who actually wrote something despite the byline. If you’re producing documents under a protective order, check whether metadata is in scope — sometimes you must preserve it, sometimes you must strip it. Get this wrong and you’re in trouble either way.

The field list to strip

At minimum, clear these: Title, Author, Subject,Keywords, Creator, Producer, CreationDate,ModDate. Also strip the XMP packet entirely if your tool offers it, and check for embedded files and form field defaults. A quick pass in our PDF metadata remover handles the standard fields; verify with a separate viewer afterward.

What stripping metadata does NOT do

Stripping metadata does not redact visible content. If your name is typed in the document body, on a signature line, in a header, or embedded in an image, it’s still there. Metadata removal is about hidden fields only. For visible content, you need actual redaction — blacking out text at the pixel level, not just drawing a black box on top of it (which anyone can remove).

Similarly, metadata removal doesn’t touch tracked changes, comments, embedded fonts (which can carry identifying info), or form field values. Many “stripped” PDFs still leak identity through one of these channels.

How to verify you scrubbed it

Don’t trust a single tool’s “cleaned” confirmation. Re-open the output in our PDF metadata viewer or any other inspector and look at all fields. The XMP block is the one most tools forget. On Linux or Mac, exiftool dumps everything including custom fields. On Windows, right-click properties shows the standard fields but not XMP.

A belt-and-braces approach for sensitive releases: strip metadata, then print the PDF to a new PDF (which re-generates the file and drops most remaining fields), then verify again. It’s the closest thing to a clean slate without re-creating the source document from scratch.