How to Remove Metadata from a PDF Before Sharing It

The Hidden Layer Inside Every PDF

When you save a PDF, you are saving far more than the visible content. Quietly embedded in the file is a metadata layer that can include your full name, your organisation, the software you used to create the document, the date you first drafted it, and sometimes a revision history stretching back months.

For most internal workflows this is harmless. But when you share a PDF with a client, publish it on your website, or submit it as part of a tender or legal process, that hidden data becomes a liability. This guide explains exactly what PDF metadata contains, why removing it matters, and how to clean it before the file leaves your hands.

What Metadata Is Stored in a PDF?

A PDF stores metadata in two places: the Document Information Dictionary (a legacy key-value store) and the XMP stream (an XML-based format introduced by Adobe that most modern applications use). Together they can contain:

Why This Matters: Real Privacy Risks

Leaking Internal Names and Roles

The Author field is populated automatically by most applications using your operating system username or the name registered in your Office suite. A file drafted by "j.smith@acmecorp.com" reveals both the employee and the organisation. In competitive or sensitive negotiations, this kind of accidental disclosure can be exploited.

Revealing Your Toolchain

The Creator and Producer fields disclose exactly which software generated the file — down to the version number. Knowing that a company is running an unpatched, three-year-old version of a PDF editor can be useful intelligence for someone probing for vulnerabilities.

Exposing Timeline and Revision History

Timestamps tell a story. If a document was "created" six months ago but the current version was saved yesterday, that gap raises questions. In legal contexts, metadata timestamps have been used as evidence in disputes over when a document was actually drafted or altered.

Unintended Keywords and Tags

Document management systems often write internal project codes, workflow tags, or classification labels into the Keywords or Subject fields. These are invisible to the reader but trivially extractable — a reminder that "hidden" in PDF parlance means hidden from the default viewer, not from anyone who looks.

Use PDFCheck to Remove or Edit Metadata

Step 1 — Check What Is Already There

Before removing anything, you need to know what you are dealing with. Use PDFCheck's metadata checker to inspect the visible document properties and software fingerprints in your PDF. Look for fields that contain a real person's name, an internal username, a company-specific code, a software version you would rather not disclose, or timestamps that do not match the story the document is supposed to tell.

If you already know the PDF must be cleaned before sending, you can skip straight to the metadata remover and download a clean copy after processing.

Step 2 — Remove Metadata with PDFCheck

Open the PDF Metadata Remover, upload your file, and click Remove All Metadata. PDFCheck creates a cleaned copy of the document and gives you a download link for the metadata-free version.

1. 1

   Upload the PDF you want to clean

   Drag the file onto the remover page or select it from your device. The tool accepts PDFs up to the current upload limit shown on the page.

2. 2

   Click Remove All Metadata

   PDFCheck strips common document properties such as title, author, subject, keywords, creator, producer, creation date, and modification date.

3. 3

   Download the cleaned PDF

   Use the cleaned copy for external sharing, client delivery, public downloads, or any workflow where author and software details should not travel with the file.

If the cleaned file also needs to meet an email or portal size limit, compress it after metadata cleanup using the PDF compression workflow. That keeps privacy cleanup and size reduction as two deliberate checks instead of one risky export step.

Step 3 — Edit Metadata When You Need Safer Values

Removing everything is the simplest privacy move, but it is not always the right one. If a PDF should keep a public-facing title, neutral author name, project subject, or keyword set, use the PDF Metadata Editor instead.

Upload the PDF, fill only the fields you want to keep or replace, then save the edited copy. This is useful for sales sheets, public reports, templates, proposals, and documents where metadata should describe the published file without exposing the person or tool that created the draft.

Step 4 — Verify the Result

After removing or editing metadata, run the cleaned file through the metadata checker again. Confirm that sensitive fields are empty, neutral, or intentionally rewritten. Pay particular attention to:

• The XMP xmp:CreateDate and xmp:ModifyDate fields — these survive many "simple" metadata clear operations
• The xmpMM:InstanceID and xmpMM:DocumentID fields — unique identifiers that can be used to link revisions of the same document
• The Producer field — often reset to the name of the tool used for sanitisation itself, which may reveal your internal toolchain
• Any custom application namespaces — some enterprise document management systems write proprietary XMP schemas that generic cleaners miss

Alternative Metadata Removal Methods

PDFCheck's metadata remover is the primary path when you want a quick clean copy, and the metadata editor is the best fit when you need controlled public values. The alternatives below are still useful when your organisation requires a desktop application, a fully offline workflow, or automated batch processing.

Adobe Acrobat Pro: Sanitize Document

Acrobat Pro includes a Sanitize Document function under the Redact tools. It can remove metadata, hidden layers, embedded content, and scripts in one pass. Editing only Document Properties → Description is not enough for privacy because it may leave XMP data or other hidden structures intact.

Print to PDF for Simple Read-Only Files

Opening the PDF and printing it to a new PDF can create a fresh rendered copy without the original metadata. This is practical for simple read-only documents, but it can remove bookmarks, form fields, accessibility tags, links, comments, and other interactive features.

LibreOffice or ExifTool for Offline Workflows

LibreOffice can re-export some PDFs with personal data cleared, and ExifTool can strip writable metadata fields from the command line:

exiftool -all= -overwrite_original document.pdf

This is useful for technical users, but always verify the output afterward. Some PDF structures and proprietary XMP namespaces can survive simple metadata commands.

Python Automation for Internal Pipelines

For generated reports, invoices, or document pipelines, Python libraries can clear common metadata before files are published. A basic pypdf pass clears the document information dictionary:

from pypdf import PdfReader, PdfWriter

reader = PdfReader("input.pdf")
writer = PdfWriter()

for page in reader.pages:
    writer.add_page(page)

writer.add_metadata({
    "/Author": "",
    "/Creator": "",
    "/Producer": "",
    "/Title": "",
    "/Subject": "",
    "/Keywords": "",
})

with open("output.pdf", "wb") as f:
    writer.write(f)

For XMP metadata, use a library that can access the XMP packet directly, then run the output back through a metadata checker before publishing.

What You Cannot Remove Without Destroying the Document

Build Metadata Hygiene Into Your Workflow

The most reliable approach is to make metadata review part of your export or publish step rather than an ad-hoc action. Practical ways to do this:

Key Takeaways

• Every PDF embeds metadata that can reveal author identity, software versions, and revision timestamps
• Check first with PDFCheck's metadata checker — you can not clean what you have not seen
• Use PDFCheck's metadata remover when you want to strip hidden fields before sharing
• Strip metadata before signing — removing it after will invalidate digital signatures
• Use PDFCheck's metadata editor when you need safe public title, author, subject, keyword, or creator values instead of blank fields