Your PDF still has your name on it: where it hides and how to remove it

A PDF is meant to look the same everywhere, which is why people treat it as a finished, sealed thing. It is not sealed. Behind the page is a set of records the programs along the way have written about the file and about you, and they travel with it every time it is attached to an email or uploaded to a portal.

For a business letter that is harmless. For a job application, a complaint, a whistleblowing report, a legal draft or a document you wanted to share without attaching your identity, it can matter a great deal. This post looks at what people actually run into, lists the places a PDF can name you, shows what we measured on a test file, and then walks through the right way to deal with it, including the cases where cleaning the file is not enough.

What people actually run into

Look across the five and the same pattern shows up again. Each person found one place their name appeared, went looking for a way to change that one place, and either failed or could not tell whether it was enough. Several were told the answer depends on which program you own. Nobody was given a complete list of where to look, and nobody was told how to check afterwards. That missing list is what the rest of this post supplies.

A PDF can name you in more than one place

The Author field in the Properties window is only the most visible record. A typical PDF can carry all of the following, and a cleaning method that handles only the first leaves the others readable.

• The author field. Usually filled from the name registered to your office software or your computer account, not from anything you typed.
• The title. Some programs fill this with the original file name, such as a Word file called something like CV_Final_v3_YourName.docx. That file name can carry a name, an employer or a version history all by itself.
• The software fields. Which program made the file and which converted it. These are not personal, but they tell a reader a lot about your setup.
• The dates. When the file was created and when it was last changed, down to the minute, taken from your computer's clock and often stored with your time zone.
• An editing history block. Many programs write a second, separate record in a format called XMP that repeats the author, lists the tools used and sometimes keeps a trail of earlier versions.
• Comment and markup authors. Every highlight, sticky note and review comment stores the name of whoever made it, and often when.
• A document identifier. A fingerprint that links a file to its earlier versions, which can show that two files came from the same source.
• Private data left by the application. Extra records that the program that created the file keeps for itself, sometimes attached to individual pages or images.

A note on the "file location" worry in the 2012 thread. The properties window in a PDF reader also shows where the file sits on your own computer, which is a fact about your machine and not something stored inside the document. Some software does write original file names or paths into the document's own fields, and cleaning removes those. But the location line you see in the reader is not what travels when you send the file.

Why the obvious fixes miss things

The threads show three reasons the common fixes disappoint, and all three are worth knowing before you pick a method.

1. 1The name is written when the file is made, not when you open itChanging your name in a reader's preferences or your computer's login changes what future work is labelled with. It does not rewrite a file that already exists. Both the 2010 and 2018 threads hit exactly this.
2. 2The free reader reads, it does not editAs the 2018 thread's accepted answer says, the free Reader cannot change the document's Author property. It can change the name on comments, which is why people sometimes think it worked.
3. 3The source file keeps the name tooIf you clean the PDF but keep sending the Word or Pages file it came from, the name is still there. Clean the thing you actually send, and fix the source if people will receive that as well.

What we measured on a test file

To see what a thorough clean really changes, we built a small test PDF that behaves like a document exported from Word: a title taken from a file name, an author, software fields, two dates, an XMP editing record and one review comment with an author name. The name is made up. Then we ran it through the Remove PDF metadata tool and looked at the file before and after.

The last two rows matter most. Looking in the properties window is a good first check, but it only shows some of the records. Searching the raw bytes of the file for the name is the stricter test, and it is the one an investigator or a curious recipient would run. In our test the name was in the file before cleaning and nowhere in it after.

The right way, depending on your situation

You are about to export a document

Deal with it at the source. In Word, the Inspect Document feature under the File and Info menus lists personal information and can remove it, and there is an option to remove personal information from file properties on save. Set the Title to something you are happy to be seen, clear the Author field, and then export. It takes a minute and means the exported PDF starts out clean. Remember to do the same for the Word file if that is also going to be sent.

You already have the PDF

Write a clean copy of the file. That is the case the forum threads kept failing at, because it needs a tool that removes every record and not only the Author field. Use one that shows you what it found before and after, so you are not taking its word for it.

The document contains photos

Photos can carry their own camera model and, if location was on, GPS coordinates. If you are building a document from phone photos, clean the photos before you put them in. That is far easier than trying to deal with them afterwards.

You need it to be completely anonymous

Then cleaning the file is only the first step. Read the pages themselves for anything that identifies you, and think about the file name you are going to send, because the name of the file on disk is visible to whoever receives it. The screenshot method suggested in the 2012 thread does remove every record, but it also removes all selectable text, so it is a last resort and not a routine.

A word on online metadata editors

The Microsoft thread's answer was a free website that edits the properties. There is an irony worth noticing. To get your name out of a document you are being asked to send that document, name and all, to a stranger's server. For a casual file that may be fine. For the kind of document where you wanted the name gone in the first place, it is the opposite of the goal. A tool that runs in your browser does the same job without the file going anywhere, which is how ours works, and you can confirm that by watching your network activity while you use it.

Mistakes that undo the work

• Cleaning, then editing again. Opening a cleaned file in an editor and saving it can write fresh author and software records. Clean last, after every edit.
• Forgetting the file name. A document called something like Application_YourFullName.pdf has told the recipient what it needs to. Rename it.
• Cleaning one copy and sending another. Keep a folder for the final version and send from it.
• Trusting only the properties window. It shows a subset. If the file matters, check again with a tool that lists the XMP and comment records too.
• Cleaning a signed document. Any change breaks the signature. If a signature matters, clean before signing.
• Assuming it is only a PDF problem. The same names sit inside Word files, spreadsheets and photos, and a PDF made from them inherits whatever they carried.

A routine you can keep

Clear the personal details in the source program before you export. Finish all edits. Give the file a plain name. Run it through a tool that lists what it removes, and read that list once. Open the result and make sure the pages look right. Send that file and keep the original for yourself. Five steps, about two minutes, and you will know exactly what the recipient can and cannot see.