PDF Metadata

How to use PDF Metadata

Key features of PDF Metadata

About PDF Metadata

PDF Metadata opens the file with pdf-lib, reads the title, author, subject and keywords from the document's Info dictionary, and writes your changes back before saving. The producer and modification date are not editable: pdf-lib sets the producer to its own name and the modification date to the time of saving, and keeps the creator and creation date if the file already has them.

It is useful for giving a scanned or exported file a proper title so most browser PDF viewers show it in the tab, for putting an author name on a document before sending it, or for clearing a template's old title and keywords from a file built from it. One limit: only the Info dictionary is changed. A PDF that also carries XMP metadata keeps its XMP values, and a reader that prefers XMP may still show the old title.

When to use PDF Metadata

A PDF is about to be published on a website. When a visitor opens it directly, most built-in browser viewers show the file's title in the tab if it has one and the filename if it does not, so "Annual report 2026" reads better than "final_v3.pdf". Set the title, author and subject here before uploading it to the site.

A document exported from a word processor carries the name of whoever created the original template as its author, often a former colleague or a person outside the company. Replace the author with the current name or the organisation before the file goes to a client.

A set of deliverables goes into a document management system that indexes the information dictionary. Fill the subject with the project code and the keywords with the terms the system searches on, and every file in the set can be found the same way.

Some changes are not metadata. To change the title printed on the first page, use the PDF Editor or Add Text. To remove a name from the pages themselves, use Redact PDF. And if the file is password-protected, remove the password with Unlock PDF first, because this tool cannot open an encrypted file.

Tips and troubleshooting

• Keywords are split on commas when you save but stored joined by spaces, as pdf-lib writes them. "tax, 2026, returns" is saved as "tax 2026 returns" and comes back that way if you open the file here again, so a two-word keyword cannot be told apart from two keywords.
• If the page shows "This file could not be opened" after you choose a file, the file is password-protected, damaged, or not a PDF. Remove the password with Unlock PDF if there is one, then reload and try again.
• The producer is set to pdf-lib's own name and the modification date to the moment of saving, on every save. The creator and creation date are kept if the file has them and set to the same values if it does not.
• Accented and non-Latin text is fine. The four values are stored as UTF-16 text, so a title in Greek, Hindi or Japanese is written and read back as typed.
• The result line reads "Document metadata updated." even if you changed nothing, because the file is rewritten either way. The output is named metadata_ followed by the original filename.
• The whole document is parsed and re-serialised to change these few fields, so a large file takes a moment to save. Pages, images, fonts and form fields come through unchanged.
• Preview Document opens the result in your browser's own PDF viewer. To check the values themselves, open the downloaded file here again and read the four boxes.
• Metadata is one of several places where a document says more than its pages show. The guide can a redacted PDF be unredacted? lists the others, such as comments, attachments and earlier versions kept inside the file.
• To clear the XMP copy of these details and the other data this tool does not touch, see how to remove metadata from a PDF.

How PDF Metadata works in your browser

When you choose a file, it is read into memory with the File API and parsed by pdf-lib into an object model of the whole document. The tool asks that model for the title, author, subject and keywords, which pdf-lib reads from the trailer's information dictionary, and puts the values into the four boxes; an entry the file does not have shows as an empty box.

When you click Save & Download PDF, the four values are written back through pdf-lib's setTitle, setAuthor, setSubject and setKeywords, each stored as a UTF-16 string in the information dictionary, and pdf-lib serialises the document to bytes, stamping its own producer name and the current time as the modification date on the way. The bytes are wrapped in a Blob for the Download button. No part of the file or the values you type is sent anywhere; the parsed document is held in the page until you leave or click Process Another File, which reloads the page.

Frequently asked questions