How do I check that a PDF is really redacted?

Open the exported file, not the original. Search it for something you removed, such as a Social Security number or a name. Then try to select and copy the text under a black box. If either returns the content you thought you removed, the box is covering the text rather than removing it. Also check the document properties for metadata, and look for comments, attachments and form fields, which sit outside the page content and are missed by a visual check.

Why does a black rectangle not redact a PDF?

A rectangle drawn with an annotation or markup tool is a shape placed on top of the page. The characters underneath are still in the file, so they can be selected, copied, searched and read by any program that opens the document. Real redaction removes the content from the file and then draws the box.

What does Lex Cloak check before it gives me the file?

Four things, and it checks the exported file itself rather than trusting its own working state. Whether text it targeted is still extractable from the output. Whether interactive form fields survived, since their values sit outside the page content. Whether annotation residue is left in the file. And whether a machine-readable code that was redacted still decodes from the rendered page.

What happens when a check fails?

The export stops, no file is written, and Lex Cloak tells you what it found and what you can do about it. For form fields, for example, it explains that the values sit outside the page content and suggests flattening the form first. You can also decide to go ahead anyway. The check informs the decision, it does not take the decision away from you.

Does this mean nothing sensitive can get through?

No, and it is worth being precise about that. These checks confirm that what you marked for redaction was removed from the exported file. They do not confirm that everything sensitive in the document was found in the first place. Detection is a starting point that you review, which is why Lex Cloak shows you the matches and lets you add, remove and adjust them before anything is burned in.

How to check a redacted PDF is actually redacted

A black box can leave the text underneath intact. The document looks finished on screen and still carries the words you meant to remove, which is how the same mistake has been made in public filings for twenty years. The check takes ten seconds by hand, and Lex Cloak runs a version of it against the exported file every time.

Download Lex Cloak Read what went wrong

The check you can run by hand

Do this on the exported copy, never on the original. It works whatever tool produced the file.

Open the exported file and search it for something you removed. A Social Security number, a client name, an account number. Nothing should come back.
Try to select and copy the text under a black box. If you can select it, it is still there.
Open the document properties and read the metadata. Title, author and subject fields often carry a name that no one thought to check.
Look for comments, attachments and interactive form fields. These sit outside the page content, so a visual pass over the pages never sees them.

If the file is a scan, the text may be in an invisible OCR layer behind the image rather than in the visible page, which is why searching matters even when the page looks like a picture.

What Lex Cloak checks before it hands you the file

Four checks, run against the exported bytes rather than against the app's own working state. Checking its own state would only confirm that Lex Cloak believes it did the job. Reading the finished file back is what confirms it.

Text it targeted is still extractable

Reads the exported file and looks for content that was marked for redaction but can still be pulled out of the output.

This is the failure that produces the headlines. The box is drawn, the characters remain.

Interactive form fields survived

Detects form structure left in the exported file.

A form field's value lives outside the page content, so redacting the page does not touch it and a visual check cannot see it.

Annotation residue is left behind

Looks for annotation objects still carrying readable data in the output.

Comments, highlights and sticky notes are separate objects that survive a page-level edit.

A redacted code still decodes

Renders the page and tries to decode machine-readable codes that were supposed to be removed.

A barcode or QR code can carry a name or an account number in a form no human reader would notice.

What this does not tell you

These checks confirm that what you marked was removed. They do not confirm that everything sensitive in the document was found. Those are different questions, and only the first one can be answered by reading the file.

Detection is a starting point. Lex Cloak proposes matches and you review them, keep them, remove them, or add your own before anything is burned in. The judgment stays with the person who knows the document. That is also why a failed check stops the export and explains itself rather than silently fixing something, and why you can still choose to go ahead when you know something it does not.

More on that division of labour in a red box is a proposal, not a decision.

Common questions

How do I check that a PDF is really redacted?: Open the exported file, not the original. Search it for something you removed, such as a Social Security number or a name. Then try to select and copy the text under a black box. If either returns the content you thought you removed, the box is covering the text rather than removing it. Also check the document properties for metadata, and look for comments, attachments and form fields, which sit outside the page content and are missed by a visual check.
Why does a black rectangle not redact a PDF?: A rectangle drawn with an annotation or markup tool is a shape placed on top of the page. The characters underneath are still in the file, so they can be selected, copied, searched and read by any program that opens the document. Real redaction removes the content from the file and then draws the box.
What does Lex Cloak check before it gives me the file?: Four things, and it checks the exported file itself rather than trusting its own working state. Whether text it targeted is still extractable from the output. Whether interactive form fields survived, since their values sit outside the page content. Whether annotation residue is left in the file. And whether a machine-readable code that was redacted still decodes from the rendered page.
What happens when a check fails?: The export stops, no file is written, and Lex Cloak tells you what it found and what you can do about it. For form fields, for example, it explains that the values sit outside the page content and suggests flattening the form first. You can also decide to go ahead anyway. The check informs the decision, it does not take the decision away from you.
Does this mean nothing sensitive can get through?: No, and it is worth being precise about that. These checks confirm that what you marked for redaction was removed from the exported file. They do not confirm that everything sensitive in the document was found in the first place. Detection is a starting point that you review, which is why Lex Cloak shows you the matches and lets you add, remove and adjust them before anything is burned in.