PDF Metadata, and How to Strip It
After reading this you will know what a PDF records about itself, where those fields hide inside the file, how to read them, and how to remove them before you send a document to anyone.
What PDF metadata is
Every PDF carries a small block of text that describes the file rather than showing on any page. This is the metadata. It typically names the person who authored the document, the program that created it, the program that produced the final PDF, and the exact date and time of creation and last edit.
Here is the hook. You write a resume in Word, export it to PDF, and send it to a company. The visible pages say your name and nothing else. The metadata block says Author: jsmith, Creator: Microsoft Word for Microsoft 365, Producer: Acrobat Distiller 21.0, and CreationDate: D:20240115094212-05'00'. You have just told the reader your network username, your office software version, and that you wrote the file at 9:42 in the morning in a timezone five hours behind UTC. None of that shows on the page. All of it travels with the file.
Where the fields actually live
A PDF stores metadata in two places, and a thorough strip has to clear both.
- The Info dictionary
- An old-style object near the end of the file holding key-value pairs:
/Title,/Author,/Subject,/Keywords,/Creator,/Producer,/CreationDate,/ModDate. Values are plain strings. - The XMP packet
- A newer block of XML, embedded as a stream, that can duplicate the Info fields and add more: tool history, document IDs, sometimes GPS data or editing timestamps. Many programs write both, so the same author name can appear twice.
If you clear the Info dictionary but leave the XMP packet, a reader who opens the raw bytes still finds your name. A tool that claims to strip metadata but skips XMP does half the job. The stripper on this page clears both.
The date format D:20240115094212-05'00' reads left to right: year 2024, month 01, day 15, then 09 hours, 42 minutes, 12 seconds, then the offset -05'00' from UTC. So 09:42:12 in a zone five hours behind UTC, which is 14:42:12 UTC.
When to strip, and when to leave it alone
Strip metadata before you publish or share a document with anyone outside a group that already knows who you are. Job applications, legal filings, public reports, forum uploads, and anything posted to a website are all cases where the author name and software fingerprint help you not at all and can leak something.
Leave metadata in place when it is doing useful work for you. Inside a company, the author and modification date help colleagues track a document. An archive of scanned records may rely on creation dates for sorting. A signed PDF stores signature data that overlaps with metadata, and stripping can break the signature, so do not run a stripper on a document you need to keep verifiably signed.
What stripping does and does not remove
Removing metadata is a targeted operation. It rewrites the file so the Info dictionary is empty or absent and the XMP packet is gone. It does not touch a single glyph on any page.
That distinction matters. If a page contains a phone number in a footer, or a name in the body text, or an image with a face in it, stripping metadata does nothing to any of that. The content is still there, still selectable, still searchable. Some PDFs also hide text behind images or in layers that do not display; a metadata strip leaves that hidden text intact too. Metadata stripping is about the file's description of itself, not about redacting content.
Do not treat this as redaction. If you need to hide text or images that appear on the pages, deleting metadata will not help. You must remove or cover the content itself in an editor that flattens the change into the page.
A worked example with the demo file
Reading a typical set of fields
Load the demo file with the sample button and you will see a set of fields like the ones below. Walk through what each one tells a reader.
| Field | Value | What it reveals |
|---|---|---|
| Title | Quarterly Report | Document topic |
| Author | jsmith | A login or short name |
| Creator | Microsoft Word | Authoring program |
| Producer | Acrobat Distiller 21.0 | PDF engine and version |
| Created | 2024-01-15 09:42 | When writing finished |
| Modified | 2024-01-16 17:03 | Last save |
- The
Authorvaluejsmithis almost certainly a system username, the same one that appears in file paths and email addresses. - The gap between
CreatedandModifiedis one day and about 7 hours, so the document was revisited the next afternoon. - The
Producerversion21.0narrows the software to a release window, which is the kind of detail an attacker uses to pick an exploit. - After you download the stripped copy and reload it, every field above reads empty. The pages look identical.
Interpreting what you find
The fields tell a small story, and the story is often more specific than people expect. Three patterns show up again and again.
First, usernames. When Author is a short lowercase string with no space, it is usually a login rather than a full name. That login often matches an email prefix, so jsmith hints at [email protected].
Second, software fingerprints. Creator and Producer together identify the toolchain. A mismatch, such as a Word creator with a non-Adobe producer, tells you the file passed through a converter, which can hint at a workflow or a leak path.
Third, timestamps and timezones. The UTC offset in a date field reveals a rough location. An offset of -05'00' points at the eastern United States or western South America, and +05'30' points at India. That is coarse, but it is more than most authors intend to share.
Common mistakes
The recurring errors are easy to avoid once you name them.
- Clearing only the Info dictionary and leaving the XMP packet, so the author name survives in the XML block.
- Assuming stripping is redaction, then publishing a file whose pages still contain the sensitive text.
- Re-saving in the original editor after stripping, which writes fresh metadata right back in. Strip as the last step before sharing.
- Running a stripper on a digitally signed PDF and breaking the signature.
- Uploading the file to an online cleaner you do not control. This tool runs entirely in your browser, so the file never leaves your device.
Related tools
Metadata often needs cleaning as part of a larger workflow. If you assembled a document from parts with the PDF Merger, the merged file may inherit metadata from the first source, so strip afterward. When you pull pages with the PDF Page Extractor or cut a file with the PDF Splitter, the pieces keep the original metadata. Files built from photos with the Images to PDF tool can carry camera metadata worth reviewing. To shrink a file, use the PDF Compressor; to turn pages the right way up, the PDF Rotator; and to render pages as pictures, PDF to Images, which drops the document-level metadata by design.
Frequently asked questions
Does stripping metadata change how the PDF looks?
No. The pages, fonts, images, and layout are untouched. Only the file's self-description is removed. Open the stripped copy and it renders identically.
Can someone recover metadata after I strip it?
Not from the stripped file itself, because the values are no longer stored anywhere in it. They can still exist in your original file, in backups, or in earlier copies you already sent. Strip before the first time you share.
Why does the author name still appear after I stripped it elsewhere?
Almost always because the other tool cleared the Info dictionary but not the XMP packet. Open the raw file and search for your name in the XML block to confirm. This tool clears both.
Will stripping remove text hidden on the pages?
No. Metadata is separate from page content. Hidden or invisible text inside the pages stays in place. For that you need to edit or redact the page content itself.
Is it safe to strip a signed PDF?
Usually not. A digital signature covers the file, and rewriting it to remove metadata invalidates the signature. Keep signed documents as they are, or re-sign after stripping if the signature is yours to reissue.
Related tools
Images → PDF
Bundle JPG/PNG images into one PDF, one page each.
PDF Merger
Combine several PDFs into one, in the order you choose.
PDF Splitter
Split a PDF into single pages or custom ranges.
PDF → Images
Render PDF pages to PNG or JPEG, single or ZIP.
PDF Booklet Imposition
Reorder a PDF into fold-and-staple booklet spreads, 2-up for duplex printing.
PDF Compressor
Shrink a PDF by re-encoding its images in the browser.