By PDFKits Team — Published February 19, 2026
TL;DR: To edit PDF metadata — title, author, dates, keywords, and the hidden XMP stream — open the free Clean Metadata tool on PDFKits, load your file, change or delete any field, and download the cleaned copy. Processing runs 100% in your browser, so the document never leaves your device. Adobe Acrobat Pro charges $14.99/month for the same edit; PDFKits does it for free.
Every PDF carries hidden information beyond what you see on the page: who wrote it, when it was created and revised, which software produced it, and sometimes the original file path on the author's computer. According to the ISO 32000-2 specification, this data lives in two places — the document information dictionary and the XMP metadata stream — and it survives most edits, merges, and exports unless you remove it deliberately.
That persistence is the problem. A proposal exported from Word can name the analyst who drafted it. A "final" contract can reveal it was modified two hours before signing. Cloud editors like Smallpdf or iLovePDF can change these fields, but only after you upload the document to their servers — an awkward trade for a privacy task. PDFKits gives you a free pdf metadata editor that reads, edits, and strips every field locally, with no account and no upload. For the broader risks, see our guide to PDF privacy and data protection.
Every PDF exposes a basic set of pdf properties through any reader's properties dialog: title (often the original working title, not the filename), author (auto-filled from the OS account or Office settings), subject, keywords, and creation and modification dates. A Word-exported PDF typically carries 8–12 populated fields out of the box. Dates alone can be strategically sensitive — in a bid, they reveal when you started and how long you took; in a dispute, they document when terms changed.
Beyond the standard dictionary, applications embed their own data. Microsoft Word records the template, total editing time, and revision count; Adobe tools may include layer names, color profiles, and editing history. XMP (Extensible Metadata Platform), stored as XML inside the file, can hold copyright terms, version history, document IDs, and custom properties — sometimes hundreds of lines the author never intentionally added.
Strictly outside metadata but equally risky: hidden layers, comments, embedded attachments, form values, and cosmetic "redactions" drawn as rectangles. Metadata cleaning does not fix those — pair it with Redact PDF for content and Flatten PDF for forms and annotations. Under GDPR Article 4, an author name or a GPS coordinate inside metadata counts as personal data, so European organizations must include metadata in their data-protection assessments.
Best for: anyone who sends PDFs outside their organization — legal teams, journalists, HR, sales, and grant writers top the list.
Most paid suites treat metadata editing as a footnote inside a much bigger subscription. Here is how the options compare on the axes that matter for this specific job.
| Axis | PDFKits | Adobe Acrobat Pro | Smallpdf | iLovePDF |
|---|---|---|---|---|
| Price | Free | $14.99/month | $9/month | $48/year Premium |
| Where files are processed | In your browser | Desktop app | Cloud servers | Cloud servers |
| View full XMP stream | Yes | Yes | No | No |
| Edit individual fields | Yes | Yes | Limited | Limited |
| One-click remove all metadata | Yes | Yes (Sanitize) | No dedicated tool | No dedicated tool |
| Account required | No | Adobe ID | Yes (paid tiers) | Yes (paid tiers) |
The structural difference: uploading a document to clean its privacy data is self-defeating if the document is the sensitive thing. Browser-side processing removes the vendor from the equation entirely.
Cleaning the copy but sending the original. The cleaned file is a new download; if you attach the old file from your desktop, nothing was removed. Verify the file you actually send by re-opening its properties.
Assuming metadata removal redacts content. It does not. Names printed on the page stay on the page. Use true redaction for visible content, metadata cleaning for hidden fields — most sensitive documents need both.
Forgetting embedded image EXIF. Photos inside a PDF can carry GPS coordinates and camera serial numbers. Check images separately with Extract Images if the PDF contains phone-taken photos.
Stripping fields your DMS depends on. Document-management systems route and index files by Title, Subject, and Keywords. For internal archives, edit fields to consistent values instead of deleting them; clean fully only for external copies.
Trusting "Print to PDF" as a cleaner. Re-printing drops some fields, but the new file gains fresh Producer and Creator entries, and Adobe Acrobat exports re-add XMP. Explicit cleaning is the only reliable method.
A typical PDF contains a title, author name, subject, keywords, creation and modification dates, the creator application, and the producer library, plus an XMP stream with software versions and document IDs. Word and InDesign exports usually populate 8–12 fields automatically.
No. Metadata lives in separate objects from the page content streams. Editing or removing it leaves text, images, layout, links, and bookmarks untouched — the pages render identically before and after.
Not from the cleaned file: the fields are rewritten out of the document, not hidden. Your original copy still has its metadata, so keep it as a backup if you may need the history later.
Yes, when it identifies a person. An author name, a username in an edit trail, or GPS coordinates in embedded photos all qualify as personal data under GDPR Article 4(1), and processing them carries the same obligations as visible content.
The document information dictionary is a small set of key-value pairs (Title, Author, dates) dating back to PDF 1.0. XMP is a richer XML packet introduced by Adobe that can mirror those fields and add version history, rights data, and custom schemas. A thorough cleaning must strip both.
Partially and unreliably. Re-printing discards the original fields but writes new Producer, Creator, and date values, and some print pipelines copy the title across. It also destroys bookmarks, links, and tags. A dedicated metadata editor is cleaner and non-destructive.
Yes. Google indexes PDF files and can use the Title field for the result headline when no better signal exists, as documented in Google Search Central. For PDFs you publish deliberately, a descriptive title and keywords help discoverability; clean everything else.
No — a merged document gets one new set of properties, usually from the merging tool, while the originals' fields are dropped. However, embedded images and any per-page hidden content carry over, so clean the merged output before distribution.
Yes. PDFKits runs in mobile browsers, so you can review and strip fields from an iPhone or Android device before forwarding a file. Processing stays on the device, which matters when the document arrived through a confidential channel.
It is the first step of three. Clean the metadata, verify any redactions are content-stream (the select-and-copy test), and flatten forms or annotations. Published PDFs are downloaded and inspected far more often than people expect.
Clean Metadata — view, edit, and strip document properties and XMP. Redact PDF — permanently remove sensitive visible content. Flatten PDF — bake form fields and annotations into static pages. Extract Images — pull embedded images for EXIF checks. Protect PDF — add password protection after cleaning. Edit PDF — adjust visible content before sanitizing. Compress PDF — shrink the cleaned file for email.