Draw a black box over a Social Security number in a PDF, or delete a paragraph in Word, and it looks like the sensitive data is gone. It usually isn’t. The text is still sitting underneath, fully selectable and searchable, and anyone who copies the “blacked out” area or runs a search can pull it right back out.
What “Real” Redaction Actually Means
Redaction means permanently deleting the underlying data. A true redaction tool strips the text and image data out of the file itself, so there’s nothing left to copy, search, or extract. Word’s Document Inspector and dedicated PDF redaction tools (Adobe Acrobat, LibreOffice Draw) do this correctly. Drawing a shape in Preview, Word, or PowerPoint does not.
Prerequisites
Gather these before you touch the document:
- Microsoft Word for Mac (Microsoft 365 or a recent standalone version)
- macOS 12 or later
- A dedicated PDF redaction tool, either Adobe Acrobat Pro (paid) or LibreOffice (free)
- The original Word document containing the sensitive information
- A few minutes before you need to send the file; don’t redact under deadline pressure
| Requirement | Details |
|---|---|
| Device | Mac (desktop or laptop) |
| OS version | macOS 12 or later |
| Word app | Microsoft 365 or Office for Mac |
| Redaction tool | Adobe Acrobat Pro (paid) or LibreOffice Draw (free) |
Why a Black Box Isn’t Safe Redaction
A rectangle drawn in Preview, Word, or Acrobat’s markup tools is just an image or shape floating on top of the page. The text layer underneath is untouched: select it with Cmd + A, copy it, paste it into TextEdit, and the “redacted” text shows up in plain view. The same problem hits a deleted sentence in Word if you haven’t cleaned up afterward. If Track Changes was ever turned on, or the file has version history, that deleted text can still be recovered by anyone who opens the revision history or rejects your edit. Real redaction works on the file’s underlying data, not the pixels displayed on screen.
Step-by-Step Guide
Step 1: Clean Hidden Data With Word’s Document Inspector
Word documents carry more than the text on the page. Author names, comments, tracked changes, and hidden text can all travel with the file, even after you think you’ve deleted them. Run this before you convert anything to PDF.
- Open the document in Word for Mac.
- Accept or reject all tracked changes first: go to the Review tab, click the arrow under Accept, and choose Accept All Changes. If you don’t want the edits kept, reject them instead.
- Go to File > Reduce File Size or, on newer builds, Tools > Protect Document, and select Check for Issues > Inspect Document. (On some Word for Mac versions this lives under File > Properties.)
- In the Document Inspector dialog, leave all categories checked: Comments, Revisions, and Versions; Document Properties and Personal Information; Headers, Footers, and Watermarks; Hidden Text.
- Click Inspect.
- Review the results screen. Any category with sensitive data found shows a Remove All button next to it. Click it for every category that applies.
- Click Close, then save the cleaned file with
Cmd + S. This is now your working copy; don’t share the original.
Tip: Run the Document Inspector again after making any further edits. Adding new comments or accepting changes after inspecting can reintroduce metadata.
Step 2: Convert the Cleaned Document to PDF
This step only converts the file to PDF. The actual redaction happens next.
- With the cleaned document open, go to File > Save As.
- In the File Format dropdown, choose PDF.
- Give the file a new name (something like
report-redacted-draft.pdf) so you don’t overwrite your working Word copy. - Click Export.
- Close the Word document. From this point on, you’ll work entirely in your PDF redaction tool.
Warning: Don’t skip Step 1 and jump straight to converting a raw, unedited Word file to PDF. Tracked changes and comments can still appear in the PDF’s layers or be recoverable from the file structure.
Step 3: Redact the PDF With Adobe Acrobat
Acrobat Pro’s Redact tool is the most reliable option if you already have a subscription or your organization provides one.
- Open the PDF in Acrobat Pro.
- Go to Tools and select Redact.
- Click Mark for Redaction, then click and drag over each piece of sensitive text: a name, an account number, an address. Acrobat outlines each selection in red.
- Repeat for every instance of sensitive content across all pages. Use
Cmd + Fto search the document for the term first, so you don’t miss a repeated Social Security number or account number on a later page. - Once every mark is placed, click Apply in the Redact toolbar.
- Confirm the warning dialog. This step permanently removes the underlying data as well as the visible text.
Warning: Marking text for redaction does nothing on its own. Nothing is actually deleted until you click Apply. This is the single most common way “redacted” PDFs still leak data: someone marks the text, saves the file, and never applies the redaction.
- Save the file with a new name once redactions are applied.
Strip Hidden Metadata With Sanitize Document
Visible redaction doesn’t touch the PDF’s own metadata. Author name, title, embedded scripts, or attachments can still carry sensitive information.
- In Acrobat, go to Tools > Redact, then look for Sanitize Document (sometimes listed as Remove Hidden Information).
- Click it and confirm. Acrobat strips metadata, embedded fonts data, JavaScript, and attachments in one pass.
- Save the file again.
Step 4: Redact the PDF With LibreOffice Draw (Free Alternative)
No Acrobat license? LibreOffice Draw handles real redaction for free, on Mac, Windows, and Linux.
- Download and install LibreOffice from libreoffice.org if you don’t already have it.
- Open LibreOffice Draw, then open your PDF (File > Open, select the PDF you exported in Step 2).
- Go to Tools > Redact.
- Use the redaction tool to draw boxes directly over each piece of sensitive text or image content you want removed.
- Once every area is marked, use the Export Redacted option in the Redact toolbar. LibreOffice flattens the page and strips the underlying data in the marked areas, producing a new PDF.
- Open the exported file separately to confirm the original, unredacted version isn’t the one you’re about to send.
Tip: LibreOffice Draw’s redaction works on both text and scanned/image-based PDFs, since it treats the marked area as pixels to remove rather than relying only on selectable text. That matters if your document is a scanned form.
Verify the Redaction Actually Worked
Don’t take any tool’s word for it. Run this two-part test on the final, exported PDF before you send it anywhere.
- Open the redacted PDF in Preview or Acrobat.
- Press
Cmd + Fand search for the exact term you redacted: a name, an account number, part of an SSN. - Confirm the search returns zero results.
- Press
Cmd + Ato select all text on the page where the redaction appears, then copy and paste it into a blank TextEdit document. - Confirm the redacted term does not appear anywhere in the pasted text.
If either test surfaces the sensitive text, the redaction wasn’t applied correctly. Go back to Step 3 or 4 and redo it on the original cleaned PDF, not the file that already failed the test.
Common Mistakes to Avoid
- Marking without applying. In Acrobat, clicking Mark for Redaction is not the same as removing the data. You must click Apply.
- Ignoring metadata. Visible text can be gone while the PDF’s Author, Title, or Company fields still name real people. Run Sanitize Document (Acrobat) or check File > Properties in LibreOffice/Preview to confirm those fields are blank.
- Using Preview’s shape tool. macOS Preview lets you draw a black rectangle over PDF text using the Markup toolbar, but it’s an annotation. The text layer underneath is fully intact and copyable. Don’t use Preview alone for anything sensitive.
- Redacting a scanned image with a text-search tool. If your document is a scanned form or screenshot, text-based redaction won’t find anything to remove because there’s no selectable text layer. You need a tool (like LibreOffice Draw or Acrobat’s image redaction) that flattens the pixels themselves.
- Skipping the verification step. Every redaction workflow above is reliable when done correctly, but “correctly” is easy to get wrong on the first attempt. Always run the search-and-select test before sending the file.
Configuration Reference
| Setting | Where to Find It | What It Does |
|---|---|---|
| Document Inspector categories | Word: File > Check for Issues > Inspect Document | Scans for and removes comments, tracked changes, hidden text, and personal metadata |
| Mark vs. Apply | Acrobat: Tools > Redact | Marking only previews the change; nothing is deleted until you click Apply |
| Sanitize Document | Acrobat: Tools > Redact > Sanitize Document | Strips PDF-level metadata, attachments, and scripts after visible redaction |
| Export Redacted | LibreOffice Draw: Tools > Redact | Flattens marked areas and exports a new PDF with the underlying data removed |
Wrapping Up
Run the Document Inspector on the Word file first, then finish in Acrobat if you have it. Redact-and-Apply plus Sanitize Document handles metadata and visible text in one place, and it’s the most reliable path of the two. No Acrobat license? LibreOffice Draw’s Tools > Redact does the same core job for free, and it works on scanned images too, which Acrobat’s text-search redaction can miss on non-searchable pages.
Whichever tool you use, finish with the search-and-select test. It takes thirty seconds, and it’s the only way to know the sensitive data is actually gone instead of just hidden from view.
Resources
- Microsoft: Remove hidden data and personal information by inspecting documents
- Adobe: Redact information from PDFs in Acrobat
- LibreOffice Discover page
- For more information on redacting in Word, visit the official Microsoft support page or third-party resources.
- Learn how to delete a Microsoft Word document securely.
