Redacting a document
Choose what to find
Tick the kinds of data to remove: National Insurance numbers, UTRs, sort codes, 8-digit account numbers, card numbers, IBANs, email addresses and UK phone numbers start ticked; UK postcodes and dates are optional.
Add your own words
List names, addresses or reference numbers to remove, separated by commas or one per line.
Open the file
Choose a PDF, CSV or text file. A text file is redacted on screen with blocks or [REDACTED] labels; a PDF is processed page by page.
Download and check
For a PDF, the tool reports how many areas it blacked out on how many pages. Open the result and look at every page before sending it.
Example: the text the tool would remove
These fictional details were run through the tool's matcher with names added to the custom list, NI numbers, sort codes, accounts, emails, phones and postcodes ticked, and block replacement chosen. Each match is replaced by blocks of the same length, which keeps the layout of text files intact.
In a PDF the same matches become solid black boxes over the page image.
Contact: Sam Taylor NI number: AB 12 34 56 C Sort code 20-00-00, account 41234567 Email [email protected], mobile 07700 900123 Postcode BS1 4DJ
Contact: ███ ██████ NI number: ██ ██ ██ ██ █ Sort code ████████, account ████████ Email ████████████████████████, mobile █████ ██████ Postcode ███ ███
Before you send a redacted file
- Scanned pages have no text layer, so nothing on them can be found automatically. Make the file searchable with OCR first, or redact those pages by other means.
- Signatures, handwriting, photos and logos are not detected. Check each page by eye.
- Names written differently (initials, a nickname, a maiden name) need adding to the custom list in every form.
- Remember the file's metadata: the title or author can contain a name. Check it with the metadata viewer.
- Keep the unredacted original somewhere safe; the redacted copy cannot be reversed.
Questions people ask
How do I permanently redact a PDF for free?
Tick the kinds of data to remove, add any names or words, then open the PDF here. Each page is rebuilt as an image with black boxes over the matches, so the original text is removed from the file rather than covered. Download the result and check every page before sharing it.
Why can't I copy text from my redacted PDF?
Because each page is now an image. That is what makes the redaction permanent: there is no text layer left in which hidden words could survive. If the recipient needs to search the document, run OCR on the redacted copy, which reads only what is still visible.
What personal data can it find automatically?
National Insurance numbers, 10-digit UTRs, sort codes, 8-digit bank account numbers, card numbers, IBANs, email addresses, UK phone numbers, UK postcodes and dates. Anything else, such as names, street addresses or customer references, can be added to the custom words list.
Does redaction work on scanned PDFs?
Not automatically. Scanned pages are images with no text to search, so no matches are found on them and nothing is blacked out. Run the file through an OCR tool first to add a text layer, then redact the OCR copy and check the result carefully.
Where your data goes
Redaction runs entirely in your browser. The original PDF or text file, the words you list and the redacted copy stay on your computer; nothing is uploaded, which matters when the document holds exactly the data you are trying to protect.
Checked against the tool on