To remove metadata from a PDF: see what yours gives away, then get a copy that no longer does.
How to remove metadata from a PDF
- Add the PDF. One file at a time.
- Read what it holds. Before you press anything, a list shows what is hidden in the file, with the actual values: author, software, dates, earlier revisions.
- Press Remove the metadata. The document is written again from scratch, without that information.
- Download the clean file. The note under the result lists what was removed and what stayed.
What a PDF says about you
Almost always: the name of whoever created it, the program and its version, the creation and modification dates. Often, the previous versions of the document as well.
Word processors, scanning apps and design software fill these fields in on their own. The author is usually the account name on the computer, which may be yours or that of whoever set up the template. Anyone who receives the file can see it under the document properties.
Beyond the document information there is XMP, a second copy in XML that tends to repeat the same facts and add a few; private data some applications leave on the pages (a PDF saved from a design program can carry the whole working file); page thumbnails; an identifier that ties copies of the same document together; and the metadata of embedded photos, with the location and phone model.
Earlier versions, still inside the file
Many programs do not rewrite a PDF when you save a change: they append the change at the end. The previous version, deleted text included, stays in the bytes.
A PDF reader only shows the latest version, so you never notice. Anyone opening the file with the right tool can still read what was there before: a corrected figure, a removed paragraph, a name that was replaced. The panel tells you when this is the case (“Earlier revisions”).
Here the document is written again from scratch and holds a single version, the latest: objects that no page uses any more are not copied.
What this tool does not do
- It does not touch the pages. Text, images and layout stay identical, names written in the document included: to hide those you need Redact PDF.
- It keeps what you use. Bookmarks, links, fillable forms with their values and accessibility tags stay where they are.
- It does not open attachments. Files embedded in the PDF stay as they are, with their own data; the panel lists them.
- It does not recompress images. Location and other metadata are taken out of embedded photos without changing a single pixel.
- It cannot keep a digital signature. Rewriting the file invalidates it, yet the signature stays inside with the signer's name; likewise a PDF/A file stops claiming PDF/A, because the claim lives in the metadata.
- It does not open password-protected PDFs. Unlock the file first with Unlock PDF, if you know the password.
Check it yourself
Open the clean file and look at its document properties: author, title and application are empty, and the producer field says only filesshifters.com.
Most readers show them under File, Properties (in macOS Preview: Tools, Show Inspector). For the
revisions, open the PDF in a text editor and search for %%EOF, which closes every
save: in the clean file it appears once.
Frequently asked questions
Is my PDF uploaded anywhere?
No. Your browser reads the document and writes the new one on your device: the site is made of static files only, and there is no server that receives files.
Which metadata is removed?
Title, author, subject, keywords, creating application and dates; XMP metadata; private application data; thumbnails; the document ID; earlier revisions; the metadata of embedded JPEG photos. If you choose, the author name on comments too.
Will the document look different?
No. Page content is copied as it is: the pages of the clean file match the original ones, pixel for pixel.
Does it remove names written in the text?
No: it removes what describes the document, not what the document says. To black out a name or a number printed on a page, use Redact PDF, which deletes it for real.
Why does the file still say filesshifters.com?
That is the producer field, which names the program that wrote the file. We fill it with the site name, which says nothing about you or the original document. Leaving it blank would not make the file any more anonymous.
How can I check this is true?
Open the developer tools, Network tab, and clean a PDF: you will see this page’s files, the visit-count requests described at the bottom, and no request carrying your document.