Redact a PDF — removed for real, not just covered
Drag boxes over what should disappear; the text underneath is truly gone.
Free: 3 jobs per UTC day, 25 MB per file. PDF exports include a visible Made with Bindery watermark. Pro exports clean PDFs.
Prefer the full studio? Open Redact PDF in the app →
Famous redaction failures share one mistake: a black rectangle drawn over text that is still there, one copy-paste away from a scandal. Bindery is built so that mistake is impossible — pages you redact are re-rendered as flat images with the boxes burned in. The text underneath does not become hidden; it ceases to exist in the file.
Draw boxes on a live page preview, page by page, with undo. And because redaction jobs are by nature the ones you least want on someone else’s server: bindery runs entirely in your browser — your file is never uploaded, so there is nothing to delete from a server afterwards.
How it works
- Drop your PDF and open Redact
- Drag black boxes over sensitive areas, on any pages
- Download — redacted pages carry no hidden text
Why black boxes leak
Most redaction failures are the same failure: a black rectangle drawn over text that is still in the file. Select-all in a viewer, copy, paste — and the “redacted” name appears. Court filings, government releases and corporate disclosures have leaked this way for years. The only safe redaction removes the content, and that is what Bindery does: every page that carries a redaction box is re-rendered as a flat image with the boxes burned in. There is no text layer under a box because there is no text layer on that page at all.
Pages without boxes are untouched, so the rest of the document keeps its selectable text. The guide on why black boxes leak goes deeper, and redact a PDF properly is the step-by-step version.
Finding what to redact
- By eye: drag boxes on a live page preview, page by page, with undo.
- Suggested patterns: one click marks email addresses, phone numbers and ID-shaped numbers found in the text layer for you to accept or reject.
- AI suggestions: with the on-device model set up in Ask your PDF — or your own API key — Bindery can suggest names, addresses, dates of birth and identifiers. Suggestions are proposals; you decide what is boxed.
- Scans: a scanned page has no text to search. Run OCR first so suggestions can work, then redact; the redacted pages are rasterised anyway.
Redacting in the browser versus uploading
| Redact PDF | Bindery | Upload-based tools |
|---|---|---|
| Where the file is processed | In your browser tab, on your device | Uploaded to the service’s servers |
| Method | Marked pages re-rendered without the text underneath | Varies — some only draw boxes |
| The unredacted original | Never leaves the device | Uploaded in full, retained for a period |
| Suggestions | Pattern and on-device AI suggestions | Server-side, if offered |
| What exists afterwards | The original and the new file, both on your device | A server copy until the retention window ends |
| Works offline | Yes, once the app has loaded | No |
A redaction checklist
- Check every page — a name usually appears more than once, and the pattern and AI suggestions only find what they are asked for.
- Box the whole line where a partial box would leave context readable.
- Redact, then open the result and try to select text on the redacted pages: there should be none.
- Run Leak Check on the result: it also looks where selecting text cannot reach — earlier saved versions, hidden fields and layers, thumbnails and attachments.
- Strip document properties with Remove metadata; the author field is a classic leak.
- Remove annotations and comments with Flatten; sticky notes are not on the page.
- Keep the unredacted original somewhere safe, clearly labelled.
Questions people ask
Is the text really removed, or just covered?
Really removed. Redacted pages are re-rendered as images with the boxes burned in — there is no text layer left underneath to copy or search.
Do untouched pages change?
No — only pages with redaction boxes are rasterized; every other page keeps its original selectable text.
Is my document uploaded for redaction?
No. The entire process runs in your browser.
Can redacted text be recovered from the file?
No. Redacted pages contain only a flat image with the boxes burned in. The original text objects are not in the output file at all.
Why did the redacted pages stop being searchable?
Because they are now images — that is what makes the redaction safe. Run OCR on the result if the remaining text must be searchable; the boxes stay black.
Does white-out in Markup redact?
No. White-out covers pixels but leaves the text layer. Use Redact whenever information must be gone, not merely hidden.