Redact PDF — Permanently Black Out SSNs and Personal Data

Permanently black out SSNs, bank details, and names in a PDF — burned into the pixels, never an overlay.

PII Redactor's core decision is that a black box drawn over text on a page isn't a redaction — it's a redaction-shaped picture with the real text still sitting underneath it, extractable by anyone who copies the page. So instead of drawing over the original page, this tool rasterizes any page that has a confirmed redaction into a flat image first, burns the black rectangles into that image's actual pixels, and rebuilds the page from the image alone, with zero text and nothing left to select or extract.

How to redact a PDF

Drop the PDF onto the Load PDF zone so its pages show in the preview.
Click Scan for PII; the text layer and an OCR pass both look for Social Security numbers, account and card numbers, and capitalized names that start with a common first name.
Work through the candidate list, jumping to each page with its page button, and press Accept on real hits and Reject on false alarms.
Drag across the page preview to draw a rectangle over anything the scan missed, such as a signature, an address or an unlisted name.
Click Redact & download; each page carrying a confirmed box is flattened into an image with the black area burned in, and untouched pages stay as they were.

When to redact before sharing

Court filings and discovery productions, which regularly contain Social Security or account numbers that have no business entering a public docket.
Medical records and insurance paperwork going to a new provider, employer or lawyer who only needs part of the history.
Bank statements, pay stubs and tax forms handed to a landlord, lender or accountant, where the full account number serves no purpose.
HR files such as offer letters, performance reviews and payroll forms before they leave the personnel folder.
Redaction helps you meet privacy obligations, but it doesn't make a document compliant on its own; check what your court, regulator or contract actually requires.

Two detectors, run together, every time

Every page is scanned twice: once by reading the PDF's own text layer directly, and once by running OCR against a rendered image of the same page, regardless of whether the text layer already found something. Both passes always run — a hit from one detector never skips the other — because a page can carry sensitive text as real selectable characters, as pixels inside a scanned image, or in some documents both at once, and relying on only one detector would silently miss whichever form the other one alone would have caught.

The pattern matching itself looks for Social Security numbers in both the dashed and undashed forms, labeled bank or account numbers, credit-card-length digit runs validated with a checksum, and capitalized word pairs that match a name list — each flagged candidate records which detector found it, and every one waits for you to accept or reject it before it's ever actually redacted.

Why the redacted page has to become an image

A confirmed redaction doesn't edit the page's existing content — it rasterizes that entire page into a bitmap at 200 DPI, paints the black rectangle directly into the bitmap's pixels, and replaces the page with nothing but that flattened image. No font is embedded into a redacted page and no text-drawing call happens on it, which is the actual guarantee here: there is no text object left anywhere on that page for a copy-paste or a script to pull out from underneath your black box.

Pages with no confirmed redaction at all are left completely alone — copied straight through from the original document, still fully text-searchable — so only the specific pages you've actually redacted lose their text layer, not the whole document.

Drawing your own redaction rectangles

Beyond the automatic candidates, you can draw a rectangle directly on the rendered page yourself for anything the pattern matching missed — a name that isn't on the built-in list, a number format the detectors don't recognize, or a signature. Manual rectangles go through the exact same rasterize-and-burn process as confirmed automatic candidates; nothing about how they're applied is any weaker.

A warning about other copies of the same file

If you're working inside a PDF Studio project and another file with the same name and size sits elsewhere in that project, this tool flags it as a possible un-redacted copy before you finish. That check matches on filename and size, not a content hash, so it's a helpful nudge rather than a guarantee — it can't tell you about a copy that's been renamed, or one that lives outside the project entirely.

Redaction is free for documents up to 300 pages, with Pro covering longer files.

What this tool doesn't do for you

It never redacts anything automatically — every pattern-matched candidate needs your explicit accept before it's burned in, precisely because a false positive redacted by default would be worse than a false negative you catch yourself.
The name-matching heuristic checks capitalized word pairs against a fixed list of common first names; a name that isn't on that list, or one written in lowercase, won't be flagged automatically and needs a manual rectangle instead.
This tool doesn't search other files, other tabs, or your device for copies of the sensitive document beyond the one same-project check described above.

Questions people ask about PII Redactor

If I black out text, is it actually gone from the file?: Only if it went through this tool's redaction process. A hand-drawn black box in a generic PDF editor usually just paints over the text, leaving the original characters extractable underneath; this tool instead rebuilds the entire page as a flattened image with no text object left on it at all.
Can a redaction be undone?: Not from the saved file. A redacted page is rebuilt from pixels that already contain the black box, so there is no hidden text or original image beneath it to recover. Your source file is left unchanged, so keep it somewhere safe if you may need the full version, and take care not to send it by mistake.
Does redacted text survive in the PDF's metadata?: Document properties don't come along: the redacted file is assembled as a new PDF that receives only the pages, so the original's Title, Author, Subject, Keywords and XMP metadata are left behind. Pages you didn't redact are copied as they were, including any comments or annotations, so review those, and open the result in Metadata Editor to confirm what remains.
When should you redact a PDF?: Whenever a document is going to someone who shouldn't see every detail in it — a court, a public-records request, a landlord or a new doctor. If only part of a page is sensitive, redacting lets you share the rest rather than withholding the whole file or retyping it.
Why does the redactor run OCR even on a normal, non-scanned PDF?: Because sensitive text can exist as an image inside an otherwise normal PDF — a scanned signature block pasted into a digital document, for instance — and the text-layer scan alone would miss it. Both detectors always run on every page so neither kind of hidden text slips through.
Will redaction make the rest of my document blurry or lower quality?: No — only pages that actually have a confirmed redaction get rasterized into an image; every other page passes through completely unchanged, keeping its original vector content, text, and quality exactly as it was.
Is redacting a PDF free?: Yes, for documents up to 300 pages, with no daily limit on how many you redact, no watermark and no account. Longer documents need Pro, and detection and flattening happen on your own device on either plan.
Does the tool know if I have another unredacted copy of this file?: It checks for one, but only within an active PDF Studio project, and only by matching filename and size rather than actual content — a renamed copy or one stored outside that project won't be caught by this check.

Related tools on BrakeTools

Metadata Editor & Eraser — Edit PDF title, author and keywords, or remove document metadata and JPEG EXIF, GPS, IPTC and comments locally.
PDF Stamp — Add page numbers, Bates numbering, watermarks, headers/footers, and edit or strip metadata — entirely on-device.
PDF Page Remover — Delete unwanted pages from a PDF and export the rest — drag to reorder first if needed, all locally with no upload.
PDF OCR Reader — Extract selectable, searchable text from scanned documents, fully client-side.

Content reviewed 2026-09-27.