Free tools that run in your browser. Use any tool without an account.

Favourites Cart
Sign in
Join free
PDF and Documents

How to Redact a PDF Permanently and Safely

By FreeToolr editors8 November 20268 min read
A PDF page with sensitive text covered by permanent redaction marks beside a verification checklist

Redact a PDF with a dedicated redaction function, apply the marks, save a new copy, and verify the result. A black rectangle, highlight, blur effect, or annotation can leave the original text searchable and recoverable.

What PDF redaction really does

Redaction removes information from a copy of a document before you share it. It can remove text, image pixels, form values, comments, attachments, or other PDF objects.

A shape placed over text only changes what you see. The text underneath may still exist in the document structure. Someone could select it, copy it, search for it, or extract it with another application.

Redaction is different from encryption. Encryption limits access to a file, while redaction removes selected information from a file that others can view. Password protection does not replace redaction.

Sanitisation is a separate step. It removes hidden data such as metadata, comments, embedded files, scripts, layers, form data, and document history. A safe workflow uses both redaction and sanitisation.

How to redact a PDF safely

Use a desktop application or a trusted PDF redaction tool that has a real redaction feature. Do not cover information with shapes or use a highlighter.

1. Keep the original file secure

Make a working copy before you redact anything. Keep the original in a restricted folder, with access limited to people who need it.

Save the output under a new name, such as case-file-redacted.pdf. Do not overwrite the source. You may need the original for legal, audit, or record-keeping reasons.

2. Identify every sensitive item

Review the whole document before marking it. Search for names, addresses, phone numbers, email addresses, account numbers, dates of birth, signatures, identity numbers, and internal reference codes.

Check headers, footers, tables, bookmarks, comments, attachments, form fields, and images. Sensitive information can appear more than once or in a different format.

Search for both full and partial values. For example, search for a surname, email domain, account-number ending, and reference code. A search for only one spelling may miss a second occurrence.

3. Mark the sensitive content

Use the application's redaction mark tool. Select the full text or image area, including spaces and nearby characters that could reveal the hidden value.

For a printed signature, select the visible pixels. For a scanned identity document, select the image area, not only any searchable OCR text.

Review the page at 200% to 400% zoom. This helps you catch small text, narrow table cells, and marks that stop short of the sensitive content.

4. Apply the redactions

Marking content is not the same as removing it. Choose the command named Apply Redactions, Permanently Redact, or a similar option.

The application should remove the underlying content and replace it with a redaction appearance. Some applications let you add a label such as REDACTED or a reason code.

Apply the redactions only after checking every mark. In many applications, this action cannot be undone in the working copy.

5. Remove hidden information

Run the application's sanitisation or remove-hidden-information feature. Remove metadata, comments, attachments, scripts, hidden layers, embedded indexes, form data, links, and document history where appropriate.

You can also use FreeToolr's Remove PDF Metadata tool to clean common document properties. Check the result after processing, because metadata cleaning does not replace visible redaction.

Sanitisation may remove useful features. It can affect bookmarks, links, digital signatures, comments, forms, and accessibility tags. Confirm that the recipient still needs those features before deleting them.

6. Save and rename the output

Save the redacted document as a new PDF. Close the application, reopen the new file, and check that it opens normally.

Do not send the original file with the redacted copy. Also check email attachments, shared folders, cloud links, and temporary exports. A secure redacted PDF can still be exposed if the original travels with it.

Advertisement

Using an online PDF redaction tool

Online processing can be convenient, but confidentiality comes first. Uploading a PDF may send its contents to a third-party server. The file may also pass through different countries or remain in backups and logs.

Before uploading a confidential document, check the provider's privacy policy, retention terms, processing location, and deletion controls. Your organisation may also require approval for external processing.

For lower-risk files, you can try FreeToolr's PDF Redaction Tool. Use the tool to mark and remove sensitive content, then download the new file and complete your own checks. Do not assume that using any online tool alone proves legal compliance.

For identity documents, legal records, medical information, financial records, or customer data, an approved local application may be a better choice. Your organisation's data protection rules should guide that decision.

Redacting scanned PDFs and images

A scanned PDF usually contains page images. It may also contain an OCR text layer that lets you search the scan. These two layers can differ.

If you redact only the OCR text, the sensitive words may remain visible in the page image. If you cover only an assumed position, the OCR layer may still contain searchable information elsewhere.

Use this process for a scanned document:

  1. Open the page at high zoom and locate the printed information.
  2. Mark the actual image area containing the information.
  3. Search the OCR text for the same name, number, or phrase.
  4. Apply the redaction to both visible content and searchable content.
  5. Review the page image and extracted text separately.

OCR can misread Arabic, French, Hausa, Yoruba, Igbo, and other languages. Right-to-left text can also affect selection and mark placement. Review every page visually when the document uses non-Latin or mixed scripts.

Automatic detection can help find likely personal data, but it is not a final review. It can miss handwriting, names without labels, indirect identifiers, and sensitive information inside images.

A worked example

Suppose you must share a 12-page tenancy file. Page 4 contains a tenant's bank account number, pages 4 and 9 contain the tenant's phone number, and page 12 contains a scanned signature.

First, save the original as tenancy-file-original.pdf in a restricted folder. Create a working copy named tenancy-file-redaction-work.pdf.

Search for the phone number and account number. Mark both phone-number occurrences, the entire account number, and the signature pixels on page 12. At 300% zoom, extend the signature mark slightly beyond the visible ink.

Apply the redactions. Then remove metadata, comments, attachments, and form data. Save the final file as tenancy-file-redacted.pdf.

Search the final file for the phone number and account number. Try selecting and copying the marked areas. Open the file in a second PDF viewer and check pages 4, 9, and 12.

If the final PDF has 12 pages and no sensitive text appears in search, that is a useful check. It is not a reason to skip visual review or hidden-data cleaning.

How to verify permanent redaction

Verification should test what a recipient can see, search, copy, and extract. Use the following checks on the final file:

  1. Search for every redacted name, number, phrase, and identifier.
  2. Try selecting and copying text from each redacted area.
  3. Paste the copied content into a plain-text editor.
  4. Open the PDF in a second viewer or browser.
  5. Inspect scanned pages at 200% to 400% zoom.
  6. Check comments, attachments, form fields, bookmarks, and document properties.
  7. If the file is high risk, use a trusted text-extraction or PDF inspection tool.
  8. Ask a second person to review the redacted pages against the original.

Do not rely on the redaction colour. A black mark can look correct while the original content remains in the file.

A 2022 study tested 11 popular PDF redaction tools. It found that two tools did not remove the text under the marked area. Other tools removed text but retained glyph-position information that could provide clues about redacted content.

The researchers reported an example where a Word-generated PDF leaked about 13 bits of information about a redacted surname. That amount could narrow a surname search to about 8,000 candidates in their example.

Rasterising a PDF can reduce some risks, but it is not a universal guarantee. The same study found that converting pages to images did not remove every possible information leak. Treat image conversion as a fallback, not proof of safety.

Common redaction mistakes

Drawing a black box over text

This hides the text visually but may leave the original object below the shape. Use a real redaction command instead.

Using a white box

A white box can fail in the same way. It may also become obvious when someone changes the page background or edits the PDF.

Blurring or pixelating information

Blurred data can sometimes be reconstructed, especially when the source has high resolution or repeated patterns. Remove the content instead.

Redacting only one occurrence

Names and identifiers often appear in headers, footers, tables, and attachments. Search for repeated occurrences and inspect every page.

Forgetting metadata and attachments

A file can show no visible secret while its properties, comments, or embedded files reveal one. Run a sanitisation check before sharing.

Treating OCR as accurate

OCR can miss text, misplace characters, or select the wrong area. Review the printed page and the text layer separately.

Overwriting the original

Overwriting removes your clean reference copy and can complicate later checks. Keep the original secure and create a separate redacted file.

Redaction and privacy duties

A technically redacted PDF does not automatically meet every legal duty. The correct process depends on the document, recipient, organisation, and country.

Under Article 32 of the EU GDPR, organisations must use suitable technical and organisational security measures. Redaction can support data minimisation and safer disclosure, but the GDPR does not require one named PDF application.

The UK Information Commissioner's Office defines redaction as permanently removing information from a copy because it cannot be disclosed. Its guidance covers public disclosure, customer documents, subject access requests, and Freedom of Information responses.

Nigeria's Nigeria Data Protection Act 2023 established the Nigeria Data Protection Commission. Nigeria's guidance also addresses processors, cross-border transfers, and compliance duties.

Saudi Arabia's Personal Data Protection Law covers personal data processing, disclosure, destruction, security, retention, and transfers. It describes destruction as removing data so it cannot be accessed or recovered.

These sources provide legal context, not a substitute for advice on your case. Ask your data protection officer or legal adviser when the document contains regulated information or a cross-border transfer.

Sources

  1. Adobe, How to redact a PDF
  2. Kumar and colleagues, Story Beyond the Eye: Glyph Positions Leak Information in Redacted Documents
  3. UK Information Commissioner's Office, Disclosing documents to the public securely
  4. EU GDPR, Article 32
  5. Nigeria Data Protection Commission
  6. Nigeria Data Protection Act 2023
  7. Saudi Arabia Personal Data Protection Law
  8. Saudi data disclosure guidance

Questions people ask

Is putting a black box over text a real PDF redaction?

No. A black box may cover the text while leaving the original content selectable, searchable, or extractable. Use a dedicated redaction feature, apply the marks, and verify the saved copy.

How do I permanently remove text from a PDF?

Mark the text with a real redaction tool, apply the redactions, and save a new PDF. Then search, copy, and inspect the redacted areas in another PDF viewer.

Can I redact a scanned PDF?

Yes, but review the page image and any OCR text layer separately. Mark the printed pixels, search the OCR layer, and check for missed or misaligned text.

Does removing PDF metadata redact sensitive information?

No. Metadata cleaning removes hidden properties such as author names or software details, but it does not remove visible text or images. Use metadata cleaning alongside proper redaction.

Is an online PDF redaction tool safe for confidential files?

It depends on the provider's security, retention, and processing terms. Check those terms before uploading, and use an approved local application for highly confidential or regulated documents.

Should I keep the original PDF after redaction?

Keep it only in a restricted location if your records policy requires it. Save the redacted version under a new filename and never send the original with the shared copy.

FreeToolr articles are written and checked by our editors. Found a mistake? Email [email protected]. Read our editorial policy.

More in PDF and Documents

Help keep every tool free

FreeToolr stays free because people help it grow. Share a tool you like, tell us what to build next, or support us on Ko-fi.

Advertisement
Advertisement