Yes. Nearly every PDF carries metadata, and it's stored in two separate places, so a file can look clean in one viewer and still contain your name. Comments and attachments can identify you too. We built a test file, cleaned it six different ways and searched the bytes afterward. Two of those methods were our own tools, and neither got it fully clean.
More than the Properties box shows. A PDF has two separate metadata stores, and it can also hold things that aren't labeled metadata but work the same way:
| Where it lives | What it holds | Who can see it |
|---|---|---|
| Info dictionary | Title, Author, Subject, Keywords, Creator, Producer, created and modified dates | Any PDF reader's Properties box |
| XMP packet | A second copy of much the same, sometimes with edit history | Metadata tools, or a text search of the raw file |
| Comments | The author name on every sticky note and highlight | Anyone who opens the comments panel |
| Attachments | Whole files embedded in the PDF | The attachments panel |
| Old saves | Earlier versions kept when a program appends changes instead of rewriting | Anyone who can parse the file |
Most PDFs carry something, even if it's only a Producer name and two timestamps. What matters is whether anything in there identifies a person, a client or an internal project.
Usually you didn't type it. The program that writes the PDF fills it in from the source document and your account:
| Made with | Typical fields | Identifies you? |
|---|---|---|
| Chromium 145, Save as PDF (tested) | Title from the page, Creator "Chromium", Producer "Skia/PDF m145", created and modified times | None |
| Word, Save as PDF | Usually title and author from document properties, Word version | Often your account name |
| Scanner or copier | Device or firmware name, scan date | Sometimes a model or serial detail |
| Design apps (InDesign, Illustrator) | Full XMP, app version, sometimes file history | Can include source file names |
We only ran the first row ourselves. The others are typical behavior, and they vary with version and settings. Printing to PDF from a browser leaves out your name, but it does keep the page's title, which on a web app may be the name of a customer or a case. Word is covered in more detail in our guide to removing metadata from Word documents.
In Acrobat Reader, press Ctrl+D (Cmd+D on a Mac) and read the Description tab. In Preview, use Tools > Show Inspector. In a browser, our PDF Metadata Editor lists the eight standard fields without uploading anything. On Linux or macOS, pdfinfo file.pdf prints the same fields, plus a "Metadata Stream" line telling you whether XMP is present. That line is the quick way to catch a file whose Info fields are empty but which still has XMP.
We built a small test PDF with all eight Info fields filled in, an XMP packet, a sticky note and an attached file. The same fake name went into each. Then we cleaned copies of it and searched the raw bytes for that name. A hit of 0 is what clean looks like, though a name inside compressed data won't show up in a plain search, so read that column as a floor, not a guarantee.
| Method | Info fields left | XMP readable | Comment author | Attachment | Plain-text hits for the name |
|---|---|---|---|---|---|
| Original test file | 8 of 8 | Yes | Yes | Yes | 3 |
| Blank Info only | 0 | Yes | Yes | Yes | 3 |
| Blank Info and XMP, incremental save | 0 | No | Yes | Yes | 3 |
| Blank both, full rewrite with cleanup | 0 | No | Yes | Yes | 1 |
| Also delete comments and attachments, full rewrite | 0 | No | No | No | 0 |
| Our PDF Metadata Editor, Strip All | 2 (both dates) | Yes | Yes | Yes | 1 |
| Our Document Metadata Remover | 0 | No in readers, bytes remain | Yes | Yes | 1 |
Three things stood out:
We included our own tools because you should know what they do. Strip All in the editor empties the eight Info fields, but in this test it left the XMP packet readable and kept the creation date, while setting a new modified date. The remover cleared both stores for PDF readers, but the old XMP bytes stayed in the file, unlinked. Both are fine for tidying a title or removing your name from Properties. If a file is sensitive, use the checklist below.
garbage=4 drops unreferenced data. Our test file reached 0 hits and shrank from 2,647 to 940 bytes this way.pdfinfo, then search the file for your name with grep -a.Printing to a new PDF from a browser or the system dialog is the blunt alternative. It usually drops comments, attachments and old saves, but it can change the layout, break links and remove form fields. It won't remove anything visible, so a black box drawn over text needs a real redaction step: see how to redact a PDF.
Photos inside a PDF can keep their own EXIF data as well. If you're sharing the originals, check them in our EXIF viewer first. Trying to work out a file's history rather than hide it? See how to tell if a PDF has been edited.
Yes, almost always. Most PDFs have at least a Producer name and created and modified dates, and many also list a title and author. A PDF can hold this in two places, the Info dictionary and an XMP packet, so clearing one may leave the other.
Often, but not reliably. The Author field usually comes from the account or document properties of whoever made the source file, so it may name a colleague or a previous owner. Anyone can edit or delete it, so treat it as a clue and not as proof.
No. Exporting usually copies the title and author from the source document into the PDF. In our test, Chromium's Save as PDF wrote no author but did store the page title, the app name and the time it was created.
Not by itself. Flattening merges comments and form fields into the page, but the Info fields and XMP packet generally stay. Clear the metadata as its own step and then rewrite the file.
Open Properties in your reader for the Info fields. To catch XMP, run pdfinfo and look for "Metadata Stream: yes," or search the raw file for "xmpmeta" with grep -a. Check the comments and attachments panels too, since neither appears in Properties.