Redacting a PDF properly in PDF by Zaweba
Mark text, areas, pages or personal information like emails, card numbers and NHS numbers, then Apply so the text and pixels underneath are removed.
Drawing a black box over a name in a PDF does not hide the name. The text is still there underneath, and anyone can select and copy it. Proper redaction removes the information from the file. In PDF by Zaweba, when you apply redactions, the text and picture pixels under the marks are removed from the file, not just covered.
The job happens in two stages: mark what should go, then Apply. Keeping those separate lets you check everything before anything is removed for good.
Marking by hand
You can mark selected text, an area of the page or whole pages. Selected text suits a name or a figure in a paragraph. An area suits a photo, a signature or a block of a table. Whole pages suit appendices or pages that shouldn't be shared at all.
Marks are only proposals until you apply them. You can look through the document, add more, and remove any you placed by mistake.
Finding and marking automatically
The faster route for long documents is Find and mark. It finds words you type, and it finds kinds of personal information, chosen by country.
The general kinds include email addresses, payment card numbers (checked with the Luhn test that card numbers follow, so random digit strings are less likely to be caught by mistake), IBANs, dates, IP addresses and web addresses. For the UK it adds phone numbers, postcodes, National Insurance numbers and NHS numbers. For the US it adds phone numbers, Social Security numbers and ZIP+4 codes.
That makes it practical to redact a bundle of letters before disclosure, or a spreadsheet printout full of customer details, without reading every line by eye. Do still read the result. Automatic finding is a great help, but you know the document and the context better than any pattern does.
The look of a redaction
You choose the colour of the redaction marks and any words that appear over them once applied. Black is traditional. Some organisations prefer a wording such as a reason code over each redaction so readers know why something was removed. Pick a style and keep to it across a set of documents.
Applying
Apply is the step that matters. Once you apply, the text and picture pixels under the marks are removed from the file. Save the redacted version under a new name and keep the original safe, since there's no getting the removed content back from the redacted copy.
Making sure nothing slips through
Redaction deals with what you can see on the page. A PDF can also carry things you can't see, such as document details, old comments, attached files, invisible text, content off the page and earlier versions. Run Protect, then Remove hidden information, after redacting, or use Sanitise document to remove all of it at once. Otherwise a name you redacted from the page could still be sitting in the file's details.
If you receive a document someone else says is redacted, the check for unsafe redactions looks for boxes drawn over text that can still be copied and redaction marks that were never applied, and offers to mark them so Apply redactions removes the text.
A routine worth following
Open a copy of the document. Use Find and mark for the personal information you need to remove, then read through and mark anything else by hand. Choose the colour and words. Apply. Remove hidden information or sanitise. Save under a new name. Then open the saved file, try selecting text where a redaction sits, and search for a name you removed. If nothing comes up, the job is done.
Find and redact is also a step that Actions can run, so a team that redacts the same kinds of information every week can save the routine and run it on a folder of files. All of it works offline in PDF by Zaweba, which matters when the whole point is that the information shouldn't go anywhere.
