How-to
How to redact a PDF so the text is actually gone
Drawing a black rectangle over a name or an account number in a PDF editor often just adds a shape on top of the page: the original text can still be selected, copied, or found by a different reader once that shape is removed or ignored. Real redaction means the sensitive text is no longer part of the document at all. This guide converts the page to a picture and blacks out the sensitive area directly in the pixels, then rebuilds a PDF from the redacted images, chaining three tools that each run in your browser.
Step by step
- Open the PDF to image tool and convert your document. Choose PNG rather than JPEG so the black box comes out with clean, solid edges instead of compression artifacts, and pick 300 DPI so the untouched parts of the page stay legible. Every page downloads as a separate image file.
- For each page that contains something to hide, open the image in the Image Redact tool, draw a rectangle over the sensitive text, and set the mode to Black box. Download the redacted image. Repeat for every page that needs it; a page with nothing to redact can be used as it came out of step 1.
- Drop every page image, redacted and untouched, into the Images to PDF tool in their original order. The tool places one image per page and rebuilds a single PDF.
- Convert and download the rebuilt PDF. Open it and try selecting or searching the text that used to be visible on the redacted pages: there is nothing to select, because the page is now a picture, not text.
Why a black rectangle in a PDF editor is often not enough
Many PDF editors let you draw an opaque shape over a paragraph and call it redacted, but the shape is usually a separate object sitting on top of the page content, not a change to the text itself. The original characters can still be there underneath: selecting the area, running Select All, opening the file in a different viewer, or removing the annotation can bring the covered text back. Because there is no reliable way for a typical user to check whether a given editor actually deletes the underlying text or just paints over it, the safer approach is to remove the text layer entirely before drawing the black box.
Rasterizing removes the text layer before you redact
Converting a PDF page to an image with the PDF to image tool renders the page through PDF.js onto a canvas and exports the pixels as a PNG or JPEG file. Once that happens, the page has no text objects left at all: it is a grid of pixel colors. Drawing a black box on that image with the Image Redact tool in Black box mode overwrites the pixel values inside the rectangle with solid black, discarding whatever was there before. There is no text underneath to recover, because there is no text anywhere in the file anymore.
What you give up by rebuilding the PDF from images
Reassembling the pages with the Images to PDF tool produces a document made entirely of pictures. That is the point for the redacted pages, but it applies to every page you include: none of the text in the final PDF is selectable, searchable, or readable by a screen reader, even on pages that had nothing to hide. Expect the file to be noticeably larger than the original text-based PDF too, since images take more space than the vector text and fonts they replace. Keep the original PDF stored privately and share only the rasterized, redacted copy with whoever needs it.
The tools used in this guide
Frequently asked questions
Does drawing a black box directly in a PDF editor delete the text underneath?
It depends entirely on the editor, and there is usually no easy way for you to verify it. Some tools genuinely remove the covered content when you flatten or export the file; others leave the original text object in the page underneath an opaque shape, recoverable by anyone who knows to look. Because you cannot easily tell which case applies to a given PDF, rasterizing the page first and blacking out the pixels directly removes the ambiguity: once the text layer is gone, there is nothing left to recover.
Will the redacted PDF still have selectable or searchable text anywhere?
No, not on any page, once you rebuild the document from images. Rasterizing is an all-or-nothing operation for the pages you convert: the pages you did not need to redact lose their selectable text too. That trade-off removes all doubt about the sensitive content; if only some pages need this treatment, you can merge them with untouched text pages from the original document afterward using the PDF merge tool, if keeping some pages searchable matters to you.