BASALT · JOURNAL
How to redact a PDF without Adobe Acrobat
You need text permanently removed from a document and you do not have, or do not want, a subscription. The useful version of the question how to redact a PDF without Adobe Acrobat is not which app has a redaction button, it is which approach destroys the content and lets you prove it afterward. Most of the free options fail the second half.
What redaction has to do
Redaction is not a drawing operation. Removing a word means finding the glyphs that compose it inside the content stream, excising them from the text-showing operators, and re-emitting the surviving text so the rest of the line still renders in place. If the area is part of an image, the pixels have to be destroyed and re-encoded, not covered.
Then there is everything not on the page: annotation /Contents strings, AcroForm field dictionaries, /ActualText entries, bookmark titles, optional content groups holding switched-off layers, the OCR text layer scanned pages carry in rendering mode 3, the document information dictionary, the XMP packet, and embedded file streams. A word can survive in any of those after the page looks clean.
The file also has to be written as a single generation, because PDF supports incremental updates where a revision is appended and the earlier bytes remain. Judge every tool against that list; most do the first part and none of the rest.
Preview: a rectangle is not a redaction
Preview ships with macOS and can draw a filled black rectangle over anything. That rectangle is a vector shape added to the page or an annotation attached to it, and the text underneath is untouched. Select across the black box and paste into TextEdit and the words come back. Open the file in another viewer, delete the annotation, and they are visible again.
Treating markup tools as redaction is the most common way documents get produced with everything still in them, covered in why redacting in Preview is not safe. Preview also saves in ways that can append rather than rewrite, so even a genuine deletion can leave an earlier revision behind.
One variation gets recommended often enough to name: print to PDF after covering, or export the pages as images and rebuild. That destroys the text layer, because the page becomes pixels, and every other useful property with it. The result is not searchable, not accessible, several times larger, and still carries the metadata.
Browser based tools: the file leaves your machine
A web page cannot process a document without receiving it. When you drop a PDF onto a site that redacts, converts, or compresses, the file is uploaded to a server you do not control, processed there, and offered back. Some services do their work in the browser and some delete after an hour, and you cannot verify either claim from outside.
For a utility bill that may be an acceptable trade. For privileged material, patient records, or anything under a protective order, uploading is itself the disclosure, and it happens before any redaction is applied. More in the risk in cloud PDF tools.
The practical test is simple. If the tool is a web page, assume the document was sent somewhere. If it is a native application you can check, because on macOS an app's entitlements are part of its code signature and codesign -d --entitlements :- /Applications/Whatever.app reports whether it may open network connections.
Command line tools: capable, and not a redaction workflow
Poppler, qpdf, mutool, and exiftool are excellent and you should have them installed: pdftotext extracts, qpdf --qdf --object-streams=disable decompresses the structure, exiftool reads metadata, and pdfdetach -list finds embedded files.
What they do not give you is a review step. Redaction is a decision about content, made per instance, in page context, by a person who understands the matter. A pipeline stripping a regular expression cannot know that one match is the client's own name in a caption that must stay, and has no interface for marking a region on a scan. Use them as the check, not the redaction.
Native Mac PDF applications
There are several native PDF editors for macOS and they vary considerably. Some offer a redaction feature that removes text from the content stream, and some offer a feature named redaction that draws a filled shape. Reading the description will not tell you which you have, so test it: take a document you do not care about, redact a distinctive word, save, and run the verification below.
The questions worth asking are whether it destroys image pixels or only covers them, whether it strips metadata and attachments in the same operation, whether it writes a single generation, and whether it can tell you anything about the output that comes from something other than itself. More in choosing an Acrobat alternative on a Mac.
The procedure for how to redact a PDF without Adobe Acrobat
Whatever tool you settle on, the sequence is the same. Inspect the file first: read the metadata with exiftool -a -u -g1, list attachments with pdfdetach -list, and decompress with qpdf --qdf --object-streams=disable to see annotations, field values, and switched-off optional content.
Mark rather than delete as you go, since review is where over-redaction and missed variants get caught. Search each value in every form it takes, including split and hyphenated forms, as in redacting every instance of a name, and mark repeated page furniture as a region across a page range. Apply once, to a new file, then verify with something that did not write the file:
pdftotext -layout redacted.pdf - | grep -i -c "smith"
qpdf --qdf --object-streams=disable redacted.pdf expanded.pdf
grep -a -i -c "smith" expanded.pdf
exiftool -a -u -g1 redacted.pdf
pdfdetach -list redacted.pdf
grep -c "%%EOF" redacted.pdf
pdfimages -all redacted.pdf ./out/imgZero hits, no residual metadata, no attachments, one %%EOF, and extracted images with the region actually destroyed. The reasoning behind each check is in how to check whether a PDF was really redacted.
Frequently asked questions
How do I redact a PDF without Adobe Acrobat?
Use a tool that removes glyphs from the page content stream rather than drawing over them, apply all marks in one operation to a new file, and verify with an independent parser afterward. On a Mac that means a native application with a real redaction engine rather than Preview's markup tools or a browser upload, checked with pdftotext, qpdf, exiftool, and pdfdetach -list.
Can Preview redact a PDF on a Mac?
No. Preview can draw a filled black rectangle over text, but the rectangle is a shape or an annotation and the text underneath remains in the content stream, recoverable by selecting and copying, by deleting the annotation in another viewer, or by extracting text from the command line. Its markup tools mark up documents, they do not remove content.
Is it safe to redact a PDF with a free online tool?
Not for confidential material. A web based tool has to receive the file to work on it, so the document is transmitted to a server you do not control before any redaction happens, and no claim about in-browser processing or deletion can be verified from outside. For privileged or regulated material the upload is itself the disclosure.
Does printing to PDF remove redacted text?
It removes the text layer when the output is rasterized, because the page becomes pixels, but it is a poor trade. The result is not searchable, not accessible to screen readers, substantially larger, and visibly degraded, and it does nothing about metadata or attachments.
What is the difference between a black box and a real redaction?
A black box is a graphic added on top of the page, while a real redaction removes the underlying glyphs from the content stream and re-emits the surviving text in place. Both look identical on screen, which is why visual inspection tells you nothing. Only extraction, from a parser other than the one that wrote the file, distinguishes them.
How can I check that a redaction actually worked?
Extract the text with pdftotext -layout and grep it for the removed terms, then decompress the file with qpdf --qdf --object-streams=disable and grep the expanded structure, which catches strings in annotations, form fields, bookmark titles, and metadata. Also read metadata with exiftool, list attachments with pdfdetach -list, count %%EOF markers, and extract images with pdfimages -all.
Do I need a subscription to redact PDFs properly?
No. Redaction is a file processing task with no server component, so nothing about it requires recurring payment, and there are one time purchase applications on macOS that do the work locally. What matters is whether the tool destroys content, strips the non-page carriers, writes a single generation, and can demonstrate the outcome.
Can command line tools redact a PDF on their own?
They can remove and rewrite content, but they give you no review step, which is the part of redaction that requires judgment. A script matching a pattern cannot tell which occurrences must stay, cannot mark a region on a scanned image, and cannot produce a reasoned log. They suit inspection and verification, not marking decisions.
Doing it in Basalt
Basalt is a native macOS application built for this problem. Redaction excises text from the content stream glyph by glyph and re-emits the surviving text in place, destroys image pixels in marked regions, and strips metadata, XMP, attachments, and hidden layers in the same pass, written as one generation to a new file while the original is never modified. A verifier re-opens the written bytes with an independent parser and proves the content is gone before anything is saved, and if it cannot prove it, nothing is written at all. The engine has no network entitlement, enforced by macOS at the code signature level, so no document is ever uploaded. It is $29 once, lifetime, up to three Macs, free for 24 hours if you want to run the commands above against its output first, and a wider walkthrough is in how to redact a PDF on a Mac.
Redaction that proves itself
Basalt destroys the content you mark, then re-opens the file it wrote and proves the content is gone before it saves anything. Your documents never leave your Mac.