How PDF Info dictionaries and XMP metadata streams work
PDF files commonly store document properties in two separate places: the classic Document Information Dictionary and embedded XMP XML packets. When you save changes here, the worker updates the 4 standard Info fields and strips XMP Metadata and PieceInfo streams across the entire file, as described in the Library of Congress PDF 1.7 Format Description for ISO 32000-1.
Why old author names survive in hidden XMP packets
Basic metadata tools often edit only the Info dictionary while leaving the XML-based XMP stream intact. Because modern PDF readers and operating system search indexers read XMP first, an author name or draft title you thought you deleted can still show up in file properties.
Deleting /Metadata from the document catalog and every indirect dictionary ensures viewers fall back to the clean values you just saved. Even embedded image or page-level XMP packets are cleared during the same pass so no hidden author strings linger inside sub-objects.
Clearing Creator, Producer, and PieceInfo objects on save
If your file was exported from a word processor or design suite, the tool displays a Made with notice showing the existing Creator and Producer strings. Saving clears both fields to empty strings and removes application-specific PieceInfo data.
Afterward, unreachable objects are pruned before saving with compressed object streams. Removing unused indirect objects also trims leftover revision baggage from desktop design tools. If you want to audit attachments, annotations, or hidden layers too, you can inspect and sanitize hidden PDF streams before publishing.
Comparing the 4 editable fields and hidden document tags
Metadata lives inside the file header rather than on the printed page. Knowing what each field controls helps you decide what to populate for searchability versus what to clear for privacy.
| Property field | UI control | Saved behavior | Typical use |
|---|---|---|---|
| Title | Title input | Trimmed string written to Info | Browser tabs and search titles |
| Author | Author input | Trimmed string written to Info | Attribution or cleared for anonymity |
| Subject | Subject input | Trimmed string written to Info | Short abstract or document category |
| Keywords | Keywords input | Split by comma and trimmed | Internal archive and desktop search |
| Creator / Producer | Read-only notice | Cleared to empty on save | Strips exporting software version |
| XMP / PieceInfo | Automatic on save | Deleted and pruned | Eliminates stale hidden XML tags |
Metadata never prints on paper. To add visible titles, dates, or page numbers onto the pages themselves, use running headers and document pagination instead. Combining clean internal properties with visible page headers makes long technical or legal binders easy to index both digitally and in print.
Where PDF titles and keywords appear in browsers and search
Setting an accurate Title improves how readers and search engines identify your PDF when it opens in a browser tab or appears in search results. If a PDF browser tab shows the wrong title, it almost always comes from an outdated template string left inside the metadata rather than the filename on disk.
Browser tab titles, OS search, and search engine indexing
Desktop file managers such as macOS Finder and Windows File Explorer index Author, Subject, and Keywords for instant filtering. Under ISO 32000-2, covered in the Library of Congress PDF 2.0 Format Description, consistent metadata helps archival systems classify documents without parsing full page text.
Updating a 4-field header takes seconds compared to re-exporting a 50-page report from desktop publishing software. Clear titles also prevent shared links in Slack, Teams, or email previews from showing an internal draft name.
Keeping metadata consistent after merging files
When you combine multiple documents into a single binder, the resulting PDF often inherits properties from the first input file or resets them. Loading the final merged packet here lets you set one unified Title, Author, and Keywords list across the whole deliverable.
Standardizing keywords across quarterly packets makes multi-year document repositories much easier to search without renaming hundreds of individual files on disk.
Wiping metadata privately before sharing sensitive files
Legal filings, blind peer-review manuscripts, and whistleblowing disclosures often leak employee names or internal software versions through document properties. Scrubbing metadata locally prevents third-party servers from logging either your file or its author history.
Using Clear all details without leaving server traces
Your browser reads the PDF bytes into local memory through the MDN Web File API documentation interface, transmitting 0 bytes over the network. Clicking Clear all details empties all 4 inputs in one click so Save details outputs a clean, anonymous file.
Unlike cloud utilities that log filenames or temporarily cache uploaded documents, every dictionary edit happens in your browser tab and disappears the moment you close the page. You can run this cleanup offline once the site is loaded.
Locking finalized documents after metadata cleanup
Once your document properties are accurate or wiped clean, you can lock finalized PDFs with password encryption so recipients cannot alter metadata or extract pages without permission. Restricting modification permissions keeps public reports and client deliverables consistent after distribution.
Before sending a file externally, verify that the Title, Author, Subject, and Keywords fields match your organization’s naming policy. Keeping a clean, minimal header avoids leaking internal project codes, contractor names, or software build numbers in public document repositories.
Tip: Always reopen your saved PDF in a browser tab or file inspector after clearing properties to confirm the window title displays the filename or your new Title, and check that no old author name remains.