A black box is not redaction
This is the single most consequential misunderstanding about PDFs, and it has leaked witness names, salaries, medical details and unredacted contracts into the public record more times than anyone has counted. Drawing a black rectangle over a word does not delete the word. It draws a rectangle. The word is still there, underneath, in full.
What a PDF page actually is
A PDF page is not a picture. It is a list of drawing instructions: put this glyph at this position in this font, draw this line, fill this rectangle. When you add a black rectangle in a word processor, a PDF viewer's markup tool or most "redaction" plug-ins, you append one more instruction to the end of that list. The text instruction that came before it is untouched. Visually the rectangle covers the glyphs because it is painted later. Structurally the text is exactly where it was, and everything that reads the file rather than looking at it — copy and paste, a text extractor, a search index, an accessibility tool — reads straight past the rectangle.
How the text comes back out
Three ways, none of them advanced. Open the PDF, select the blacked-out area with the cursor, copy, paste into any text editor: the hidden words appear. Run any text extraction tool over the file and it returns the page's text in full. Or open the file's raw content, where the strings often sit in plain view. This is not hacking; it is the file working as designed. The uncomfortable part is that the person who made the document has no way of noticing, because on screen it looks redacted — which is precisely why it keeps happening to people who were being careful.
What real redaction has to do
The text object must not exist in the output. There are two honest ways to get there: remove the specific text operators and rebuild the content stream, or convert the affected page to an image, paint the boxes onto the image, and rebuild the page from that image. This app takes the second route deliberately, because it is the one that cannot be half-done. A page with a redaction box is rasterised, the boxes are painted onto the raster, and the page is rebuilt from the result — there is no text object left underneath to recover. Pages with no boxes are copied from the original untouched, so they keep their vector text, their sharpness and their selectability.
The trade, stated plainly
Rasterising a page costs something and we would rather say so than hide it: on a redacted page the text stops being selectable and searchable, and the file gets larger. That is the price of the guarantee. It is why only pages you actually redacted are converted — a fifty-page contract with two redacted pages keeps forty-eight pages of real text. And it is why the output is always assembled into a new document rather than saved over the original: a library that writes back its whole object table can leave the original text objects in the file as orphans even after they have been removed from the page. Building fresh makes that structurally impossible.
Frequently asked questions
Is drawing a black rectangle over text enough to redact it?
No. The rectangle is drawn on top; the text stays in the file and comes back with copy and paste or any text extractor. This is the most common cause of accidental disclosure in PDFs.
Can't I just highlight the text in black?
No — a highlight is an annotation drawn over the page, and the text underneath is untouched. Some viewers will even let you remove the highlight.
Does flattening the PDF fix it?
Not reliably. Flattening usually merges annotations into the page content; it does not necessarily delete the text operators. See the article comparing methods.
Why does the app turn my page into an image?
Because that removes the text object rather than covering it. Only pages that contain a redaction box are converted; the rest keep their real, selectable text.
Does that make the file bigger?
Yes, on the pages that were redacted. That is the honest cost of the text genuinely being gone.
Can the text be recovered from the image?
The glyphs are gone, so there is nothing to extract. Someone could of course read whatever is still visible on the page — redaction removes what you covered, not what you left.
Remove the text, then prove it is gone. Redact takes the words out of the page rather than covering them, then reopens the file it just made and searches it twice — by text extraction and by raw stream scan — before letting you download. The file never leaves your browser.
Redact a PDF in the browser