PDF metadata editor
See and change what a PDF says about itself: title, author, subject, keywords, the software that made it and when. Both copies of the metadata are rewritten together, and the file is never uploaded.
Choose a PDF to read its title, author and other properties.
Document
What a PDF says about itself
Besides its pages, a PDF carries a handful of properties: a title, an author, a subject, keywords, the application that created the original document, the software that wrote the PDF, and when it was created and last changed. Readers show them under File › Properties, operating systems index them for search, and document management systems file things by them.
They are also where documents leak. A report exported from Word carries the name of whoever's account it was written on; a contract drafted from a template can still have the title of the template, or of the client it was first written for; a scan may name the office copier that made it. Choose a file and every field is shown, including any custom entries the producing software added, before you decide what to keep.
Two copies of the same information
Since 2001, PDF has stored metadata twice: in the old info dictionary, a simple list of named text strings, and in an XMP packet, a block of XML embedded in the file. Different programs read different copies. Acrobat and most desktop search tools prefer XMP when it is present; many browsers, command-line tools and older libraries read only the info dictionary.
Editing one and not the other is the most common way metadata edits fail: you change the title, check it in one program, and a colleague's copy still shows the old one. So this tool writes both. When the file already has an XMP packet, it is replaced with a new one that says exactly what the info dictionary says; the panel tells you if the two copies disagreed in the file you opened. If the original had no XMP, none is added.
Files that declare themselves PDF/A — the archival variant many courts and public bodies require — keep that declaration in the new packet. The rest of the file is not touched, so a file that was valid PDF/A stays so.
Dates
PDF dates are stored as text like D:20240131140502+08'00', meaning 14:05:02 on 31 January 2024 at UTC+8. They are shown here converted to your own time zone, and written back with your zone's offset. A date with no zone at all is read as local time, which is what it almost always means in practice, though strictly the standard calls it unknown.
Set Modified to the moment I save is on by default, because that is what the field means. Untick it to keep whatever is in the Modified box, including a date you have typed in yourself. Nothing checks that the dates are true; they are statements the file makes, not a tamper-evident record, and they carry no more authority than any other editable field.
Removing everything
Remove all metadata writes a copy with no info dictionary and no XMP packet. It is the sensible thing to do before sending a document outside your organisation, or before publishing a PDF whose history is nobody's business. The pages are not affected.
It is worth being precise about what this does not reach. Author names and comments can also live inside the content: in annotation author fields, in form field values, in the text of the pages themselves, or in the properties of an embedded file. Those are part of the document rather than its metadata, and they are copied across unchanged. If you need a document scrubbed of every trace of its origin, look at those too.
Nothing else moves
The saved copy is the whole document — pages, fonts, images, bookmarks, links, form fields, attachments — copied across as it was, with new metadata in place of the old. Page content is not re-compressed, and the file comes out within a percent or two of its original size. A file that was password-protected is read with your password and saved without encryption; digital signatures cannot survive any edit and will show as invalid on the copy.
Common questions
Why does my PDF still show the old title in some programs?
Probably because it had two copies of the metadata and only one was changed. PDF keeps properties both in an info dictionary and in an XMP packet, and programs differ in which one they read. This tool rewrites both so they agree. If you edited a file elsewhere and see an old title, open it here: the panel flags a mismatch between the two copies.
How do I remove the author name from a PDF?
Clear the Author field and save, or use Remove all metadata to drop every property at once. Both copies of the metadata are updated. The name can also appear in comment author fields or on the pages themselves, which are content rather than metadata and are not changed.
Can I change the creation date?
Yes. Both the Created and Modified dates are editable, in your local time zone, and are saved with your zone offset. Untick Set Modified to the moment I save if you want to keep a Modified date you entered. The dates are simply what the file claims; they are not evidence of anything.
What is the difference between Creator and Producer?
Creator is the application the document was originally written in, such as Microsoft Word or InDesign. Producer is the software that turned it into a PDF, such as a PDF library, a print driver or macOS Quartz. Both are free text and both can be changed or cleared.
Does editing metadata change the pages?
No. The whole document is copied into the new file exactly as it was, with only the metadata replaced. Text, images, bookmarks, links and form fields are untouched, and nothing is re-compressed.
Will a PDF/A file stay PDF/A?
The PDF/A declaration in the XMP packet is carried over, and the new packet is written to agree with the info dictionary, which PDF/A requires. The rest of the file is copied as is. If the original was valid PDF/A, the edited copy should validate too, though for anything going to a court or an archive it is worth running it through a validator.
Is my PDF uploaded?
No. The file is read and rewritten by this page, in your browser. Nothing is sent to a server, which matters for the kind of document whose metadata you are worried about in the first place.
What are the other entries it offers to keep?
Some programs add their own properties to the info dictionary: a company name, a source file path, a document ID, a template name. They are listed in the panel and kept by default. Untick the box to drop them while keeping your edits to the standard fields.