You need to send a document with some parts hidden: an account number, a name, a salary figure, a paragraph that isn't anyone else's business. The obvious move is to draw a black box over it. On screen, that looks done.
It usually isn't. In most PDF editors, a black box is a new shape sitting on top of the text. The text is still in the file, right underneath. Real redaction removes it.
The black box mistake
A PDF page is built in layers. The text is stored as characters with positions. When you draw a rectangle in an editor, or highlight text with the colour set to black, the editor adds a shape or annotation on top. It doesn't touch the characters below.
So anyone who gets the file can:
- Select the area under the box and copy it. The hidden text pastes perfectly into an email.
- Search for a word and watch the reader jump to it, under the box.
- Open the file in another app, or convert it with a PDF-to-text tool, and get all the text without the box at all.
- Delete the box, if it's an annotation, in any editor.
Court filings, government reports and company documents have all been published with "redacted" text that readers recovered in seconds by copying and pasting. It's one of the most common document security mistakes there is, and it keeps happening because the result looks right.
The same applies to "Whiteout" and "Edit text" in Edit PDF. Both cover the old text with a white box and put new text on top. Use them for typos, never for hiding anything: the original text stays in the file.
How Redact PDF removes the text
Redact PDF takes a blunter approach on purpose. For every page where you've drawn a box, it renders the whole page as an image (at about 180 dpi), paints your box into that image, and replaces the page with the image.
After that, there's no text left on that page. Not under the box, not anywhere else on the page. There's only a picture that happens to show a black rectangle. Images under the box are gone too, because the page is now one flat image. Copying, searching and text extraction have nothing to find.
Pages without any boxes aren't touched. They keep their real text, links and quality.
The trade-offs
Turning a page into an image has costs, and you should know them before you redact a 40-page report.
- Text on a redacted page is no longer selectable or searchable, including all the parts you didn't redact.
- The page can look slightly softer than the original, since it's an image at about 180 dpi rather than sharp vector text. It's fine for reading and printing, but you'll notice it if you zoom in a long way.
- The file can get bigger. A page of text might be a few kilobytes; the same page as an image is usually a lot more.
If you need the rest of a redacted page to be searchable again, run the result through OCR PDF afterwards. OCR only sees what's visible, so it can't bring back anything under a box. If the file has grown too big to email, Compress PDF will usually bring it back down.
I think the trade-off is worth it. A redaction method that keeps text selectable has to be very careful about what it removes, and that's where mistakes slip in. An image can't leak text it doesn't contain.
Redacting a PDF on EddiesPDF
- Open Redact PDF and add your file.
- Drag across each piece of text, number or image you want to hide. A box appears; drag its handles to adjust it so it fully covers the area, with a little margin.
- Choose "Fill color": "Black" is the usual choice for legal and official documents. "White" hides the area without a visible box, which looks cleaner but doesn't tell the reader something was removed.
- For something in the same place on every page (a header with a client name, an account number in a footer), draw it once and turn on "Repeat on every page".
- Click Redact and save and download the result.
You can draw boxes on the first 30 pages. For content on later pages, "Repeat on every page" applies your boxes to all pages at the same position, including ones not shown. If the content you need to hide moves around in a long document, split out the relevant pages with Split PDF, redact them, and merge them back.
There's no search-and-redact here; you mark areas by hand. For a document with a name on fifty different pages in fifty different places, go through it carefully, or use a desktop tool built for bulk redaction.
Redaction runs in your browser, so the unredacted original never leaves your device.
Don't forget the parts you can't see on the page
The visible page isn't the only place information lives.
- The file name. "Smith divorce settlement final.pdf" gives away the name you just redacted. Rename the file before sending.
- The document title and other properties (author, subject, keywords). Redact saves a brand-new PDF, so these aren't copied across, and neither are bookmarks or attached files. It's still worth opening the downloaded file's properties in Acrobat Reader (File, then Properties) or Preview on Mac (Tools, then Show Inspector) to check.
- Comments. A comment with the original wording survives on pages you didn't redact.
- Other pages. A name redacted on page 2 might appear in plain text on page 9.
How to check the redaction worked
Do this every time, on the downloaded file, not the original:
- Open it in a normal reader like Acrobat Reader or your browser.
- Try to select text under one of the boxes. You shouldn't be able to select anything on that page.
- Search for a word you redacted. It shouldn't be found (unless it also appears on a page you didn't redact, which is worth knowing).
- Copy a paragraph from just above a box and paste it into a plain text editor. If you get nothing, the page is an image, as intended.
Sometimes the better move is not to send the page at all. Removing it with Delete pages is cleaner than a page that's 80% black boxes.
