What a PDF's metadata reveals about you
Short answer: A PDF's metadata can name the author, the program and its version, sometimes the operating system and its build, and the exact time it was created and changed. Edited files can keep old values that viewers no longer show. Open the file in a metadata viewer to see what it carries, and remove it by rewriting the file.
Three producers and one edit example
We made the same three-line meeting note with three common PDF producers, left their metadata exactly as each program wrote it, and added a fourth file we constructed ourselves to show an incremental save. All four were opened in our metadata viewer (Chromium, 12 November 2026); you can check them yourself:
| File | What the viewer shows |
|---|---|
| metadata-chromium.pdf — Chromium “Save as PDF” | Title from the web page title; Created with: the full browser identification (Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) … HeadlessChrome/149.0.0.0 …); PDF producer Skia/PDF m149; creation and change time to the second |
| metadata-macos-quartz.pdf — macOS built-in PDF engine, through the print system | PDF producer macOS Version 26.5.2 (Build 25F84) Quartz PDFContext — the exact system version and build; creation and change time |
| metadata-pdf-lib.pdf — pdf-lib, a library many web PDF tools use | Created with and PDF producer pdf-lib (https://github.com/Hopding/pdf-lib); creation and change time |
| metadata-incremental-save.pdf — constructed by us: a draft whose fields were later changed and saved incrementally (the editor name and the person are fictitious) | Title Meeting note; nothing else personal. But the file still contains the old title Salary review 2026 - draft for A. Sample and the author A. Sample |
What this means in practice:
- Software and system versions are written by default, but how much they say depends on the producer. The macOS engine wrote the real system version and build. Chromium wrote its browser identification, whose “Mac OS X 10_15_7” is a fixed compatibility value, not the system the note was made on (the same computer as the Quartz file). Other producers write only their own name. Read these fields as what a program chose to report, not as proof of the system or of how up to date it is.
- Times are exact to the second and include a time-zone offset when the program writes one.
- Titles often come from a file name or web page title you never meant to share.
- Old values survive editing. Many editors change a field by appending a new version to the end of the file (an incremental save). Viewers show only the latest value; the earlier one is still in the file, readable with a text editor.
The first three are what we could produce and check on our own computer. Office documents also store the author and other personal information in their properties; Microsoft documents how to remove it with the Document Inspector before you export to PDF.
See what your file carries
- Open the file in the metadata viewer. It runs in your browser; the file is not uploaded.
- Read the document information fields, the XMP block (a second, XML copy of metadata that can hold more, including edit history), and the facts below them: attachments, form fields, signatures, scripts.
- If something should not travel with the file, select Remove this metadata.
Remove it — and the old versions with it
Remove metadata empties the document information, deletes the XMP streams, and writes the whole file anew, so earlier versions from incremental saves are not carried over. It then reads the new file again and offers it only if none of these fields is left.
We ran it on metadata-incremental-save.pdf: in the result, neither the old title nor the old author name appears anywhere in the file bytes.
Metadata is only one of several places hidden information can live: attachments, comments, form values, and text hidden under black boxes are separate. For files that leave your hands, also run the redaction check, and see removing PDF metadata and hidden information for the full list.
FAQ
Where does the author name come from when I never typed one?
Many programs fill it in from the user or account name of the computer. In Word, the Document Inspector (File › Info › Check for Issues) can remove it from the document before you export the PDF.
Does removing metadata change how the pages look?
No. Page content, fonts and images stay as they are. Removing metadata also does not touch names written on the pages; those need redaction.
Is metadata proof of who wrote or changed a file?
No. Any field can be edited, and many programs fill them automatically. Metadata is a hint, not evidence — in both directions: unchanged fields do not prove a file was not edited.
Sources
- Microsoft Support: Remove hidden data and personal information by inspecting documents · Checked on 2026-11-12