Burn it in.
Don’t just cover it.
RedactDrop turns every redacted page into pixels, so the text under a box is destroyed — not hidden behind a rectangle anyone can lift off.
PDF, PNG, JPEG, WebP. In your browser, nothing uploaded.
What makes it hold
The text is destroyed, not hidden
Every page with a box on it is rendered to pixels at the DPI you choose, the boxes are painted over those pixels, and a new PDF is assembled from the images. On those pages there is no text layer left: nothing to select, nothing to copy, no overlay to remove. Pages you did not touch can stay vector, or you can rasterise all of them for a uniform file.
Search finds every occurrence
Type a name, an account number or a regular expression. RedactDrop reads the PDF's own text layer, lists each hit with its page number, previews the boxes on the page, and adds all of them with one click. Scanned PDFs have no text layer; the tool says so instead of pretending, and you draw by hand.
The output is checked before you download it
After applying, the tool re-opens the file it just built, runs text extraction over every rasterised page, counts annotations and reads the metadata. The report shows the numbers — text items found, annotations found, fields still set — and it only passes when all of them are zero.
About RedactDrop
RedactDrop is a browser tool for permanently removing information from PDFs and images. Most "redaction" you see in the wild is a black rectangle drawn on top of a page: the text is still in the file, and a viewer that selects all, an editor that deletes the shape, or a script that dumps the text layer gets it back in seconds. RedactDrop works the other way round. You mark what has to go — by drawing boxes with the mouse or a finger, or by searching the document for a name, a number or a regular expression and redacting every match — and when you apply, each affected page is rasterised: rendered into a bitmap at 150, 200 or 300 DPI, painted over, and placed into a brand-new PDF as an image. The original page objects, fonts and text runs are not carried over, so there is nothing underneath to recover.
Pages without boxes can be copied over as vector so they stay sharp and searchable; comments and markup are dropped from them, form fields and links are kept, and document metadata is cleared. For screenshots and photos the same idea applies: the picture is decoded, the boxes are burned into the pixels and a fresh PNG or JPEG is encoded with no EXIF, no embedded thumbnail and no layers. Every step — rendering with pdf.js, painting on a canvas, rebuilding with pdf-lib, and the verification pass that re-opens the result and extracts text from it — runs inside your browser tab. Nothing is uploaded, there is no account, and the page keeps working with the network switched off. The honest limits: rasterised pages lose selectable text (run OCR on the redacted copy if you need it back), a 300 DPI scan of a long document makes a large file, and pixelation is offered as a style but solid fill is the only one that cannot be reversed.
Guides
How redaction fails, and how to do it so it holds.
Why black boxes fail: redaction mistakes
Courts, ministries and companies keep publishing documents whose redactions come off with a copy-and-paste. The failures follow a few patterns.
6 min read — Updated June 2026
How to redact a PDF properly: a checklist
Rasterise, drop the text layer, clear metadata, strip attachments and comments, kill the revision history, then prove it with text extraction.
7 min read — Updated August 2026
Redacting screenshots: why blur is not enough
Blur and pixelation keep part of the signal, swirls can be unwound, and the EXIF thumbnail may still hold the original. Solid fill, then flatten.
6 min read — Updated July 2026
Questions
Is my PDF or image uploaded anywhere?
No. RedactDrop is a static page: the file is read with the browser's File API, rendered by pdf.js inside your tab, painted on a canvas and rebuilt with pdf-lib — all locally. There is no upload endpoint, no account and no analytics. You can load the page, switch off the network and still redact and download a file.
Is the text under the box really gone?
Yes, on every page that carries a box. That page is rendered to a bitmap, the box is painted into the pixels, and the bitmap is placed in a new PDF as an image. The original page's fonts, text runs and vector objects are not copied into the output at all, so there is nothing to select, extract or reveal. After applying, the tool re-opens the output and runs text extraction to show you that those pages contain zero text items.
Why do redacted pages turn into images?
Because that is the only way to be certain. A PDF page is a program that draws text and shapes; drawing a black rectangle on top just adds one more instruction, and the text instruction is still there. Rasterising replaces the whole program with a picture of its result, minus what you covered. The cost is that those pages no longer have selectable text — run OCR on the redacted copy if you need it back.
Which DPI should I choose?
200 DPI is the default and is fine for ordinary office documents: text stays crisp on screen and prints acceptably. Choose 150 for large files you will only view on screen or email. Choose 300 when the file will be printed or contains small type, stamps or fine drawings. Higher DPI means larger output; a 300 DPI page as JPEG is typically 300–800 KB, and PNG is larger still.
Search found nothing in my scanned PDF. Why?
A scan is a picture of a page; unless someone ran OCR on it, there is no text layer for search to read. RedactDrop checks the first pages when it opens a file and tells you when no text layer exists. In that case draw the boxes by hand — the burn-in works exactly the same way on scanned pages, because they are rasterised too. The tool does not run OCR on your document, on purpose: OCR would only tell you where text probably is, and "probably" is not good enough for redaction.
Will the file get bigger?
Usually, yes — a text page of a few kilobytes becomes an image of a few hundred kilobytes. Pages you did not redact can be kept as vector, so only the pages with boxes grow. JPEG page images are smaller than PNG; PNG is lossless and better for line art. If size matters more than print quality, choose 150 DPI and JPEG.
Can I undo a box before applying?
Yes. Nothing is destroyed until you press Apply. Before that, select a box and press Delete, drag its handles to resize it, nudge it with the arrow keys, or press Ctrl+Z (Cmd+Z on a Mac) to step back through your last changes. The list on the right shows every box with its page and coordinates, and each row has a delete button. After applying, you can still go back to editing — the original stays open in memory until you close it.
Does it work on a phone?
Yes. Boxes are drawn and resized with a finger, the page thumbnails sit in a strip above the canvas, and the search and redaction list open as a sheet from the bottom of the screen. Large PDFs at 300 DPI are heavy for a phone's memory, so prefer 150 or 200 DPI there and keep untouched pages as vector.
Is pixelation as safe as a solid box?
No, and the tool says so when you pick it. Pixelated or blurred text keeps some of the original signal, and research on depixelation has shown that short strings in a known font can sometimes be reconstructed. RedactDrop uses large blocks to make that harder, but solid fill is the only option that leaves nothing behind. Use pixelation for faces or objects where a rough shape is acceptable, and solid fill for anything textual.