How to redact a PDF for free

Updated October 14, 2026. Written for PdfCraft 0.7.1 (what changed since).

Why a black box isn't redaction

A PDF page is a list of drawing instructions: put these letters here in this font, paint this image there. A black rectangle, a highlight set to black or a shape comment adds one more instruction on top. The letters underneath are still in the file, so anyone can select them, copy them, search for them or delete the box. Turning every page into an image does get rid of the letters, but then nobody can search or copy anything in the document, including the parts you meant to keep.

Real redaction removes the instructions themselves: the characters, the image pixels and the vector shapes inside the marked area, plus any comments and form fields that overlap it. Then the file has to be rewritten, because many PDF tools, PdfCraft included, normally save by appending changes and leaving the original bytes intact. After you apply redactions, PdfCraft's save rewrites the whole file, so the old content isn't left behind in an earlier version inside it.

Redact a PDF step by step

1. Open a copy of the PDF at pdfcraft.run/editor. Keep the original somewhere safe; redaction can't be undone once saved.
2. Set how the marks will look in Redaction properties (press Ctrl/⌘+K and type "redact" to find every redaction tool): the fill colour, and optional overlay text such as "Redacted" or a legal code. U.S. FOIA and Privacy Act code sets are built in. The settings apply to the marks you make next.
3. Mark the content with Redact text and images. Select text to mark words, or drag a box to mark an area such as a photo or a signature. Hold Alt (Option on a Mac) while dragging to mark a column instead of whole lines.
4. Mark whole pages with Redact pages when an attachment or an appendix has to go entirely.
5. Apply redactions. PdfCraft lists what will happen: everything under the marks, and comments and form fields that overlap them, is removed permanently.
6. Save with Menu › File › Save (Ctrl/⌘+S) and give the file a new name. The redacted PDF downloads to your computer.

Until you apply them, marks are only marks: you can move them, delete them or undo them. Applying removes the content from the open document, and saving writes that to the file.

Find every copy of a name or number

One missed instance of a name is a failed redaction, and on a 60-page bundle you will miss one by eye. Find text and redact searches the whole document and marks every match in one step. It takes a single word or phrase, a list of words and phrases (one per line, typed or imported from a .txt file, up to 1,000 entries), or built-in patterns for phone numbers, email addresses, credit card numbers, Social Security numbers and dates.

Read through the marks before applying them. A pattern search finds what looks like a phone number, which sometimes includes an invoice number you meant to keep, and it misses a client's name spelled differently in the signature block.

Scanned pages

A scanned page is a photo of text, so there are no letters for a search to find, and the browser version of PdfCraft can't run text recognition on it. Area marks still work: drag a box over the part of the scan to remove and PdfCraft deletes those pixels from the image. Go through a scanned document page by page.

Clean out what you can't see

Visible text is half the job. A PDF can also carry the author's name and editing software in its metadata, file attachments, old comments, hidden text, hidden layers, links and scripts, and private data left by the program that made it. Remove hidden information lists these for you to tick and removes them. Sanitize document removes all of it in one go and rewrites the file on save. For a document leaving your organisation, run Sanitize after applying redactions.

Check the saved file

• Open the downloaded PDF in a different viewer, such as your browser's built-in one.
• Search (Ctrl/⌘+F) for each name and number you removed. It should find nothing.
• Drag-select across a redacted area and paste into a text editor. Nothing from under the mark should come through.
• Look at the document properties for the author and title fields if you ran Remove hidden information.

What PdfCraft can't redact yet

Part of a match. Find text and redact marks whole matches, with no setting yet to keep part of one, such as the last four digits of an account number. Local formats. The patterns have no country settings yet, so check non-U.S. phone numbers and ID numbers by eye, or put them in a word list. Vertical text, as in some Japanese documents, can't be removed under a mark; PdfCraft says so and changes nothing. The project is in alpha, which is one more reason to run the checks above every time.

Preview and Acrobat Pro, for comparison

On a Mac, Preview's Redaction Selection removes the text you select, which is fine for one or two lines. The redaction becomes permanent once you close the document, and table-of-contents entries pointing at the redacted text are not removed, so a heading with a person's name can survive in the sidebar. For anything longer, the search and hidden-content tools above are what save you.

Acrobat keeps redaction for Pro: Acrobat Standard at US$14.99 a month doesn't include it. If you redact a handful of documents a year, the free route covers it. If you redact scanned files every week, paying for a tool that can search scans saves the page-by-page pass.

Before sending anything redacted, do the two-minute search test on the saved copy. It is the only check that tells you what the recipient will be able to find.