How to redact a scanned PDF on a Mac
A scanned agreement shows the payment account clearly, but Find returns no match. That result can occur before you edit anything. It cannot tell you whether a later redaction removed the account.
For a scan, inspect the image and any searchable text separately. You need to know what the recipient can see as well as what the file lets them extract.
Check what the scan contains before marking it
Try selecting a line and pasting it into a plain text editor. If nothing comes out, the page may be image-only. If words appear, compare them with the image, especially the numbers you plan to remove. An OCR error can make an account number searchable under the wrong characters.
OCR means optical character recognition. It can add a searchable text layer to scanned page images; Adobe’s documentation describes that layer and asks users to review recognition errors. The scan image can still contain the same information as the recognized text.
A rectangle painted over the page does not establish that its text layer was removed. Deleting the text alone does not establish that the visible image was changed. Use a tool with image-area redaction on a copy. The general Mac PDF guide covers Preview’s redaction tool and the difference from ordinary shape annotations.
Match the marked area to the visible words
On a crooked or faint scan, the recognized characters and their boxes may not match the image closely. Zoom in around both ends of the account number and along the top and bottom of the selection. Include any exposed fragments in the area to remove. Review signatures and codes separately according to what the recipient needs.
Inkfish’s scan FAQ documents on-device recognition for pages without a text layer in English, Chinese, and German. Review suggestions against the image and mark missed areas manually. Its usage guide covers the controls; the extra work with scans is checking that each selection covers the visible content you intended to remove.
Inspect the saved output after redaction
Close the editing window, then reopen the copy you intend to attach. Keep the original separate. Check each page for repeated details that may not have been selected, and inspect the edges of every changed area at a useful zoom.
Then search for the removed number and a distinctive part of it. Copy the page text into a plain text editor and look for remnants or repeated values. These checks address different failures:
| Result | What still needs review |
|---|---|
| No searchable text | The visible page image may still show the account |
| Text extraction looks clean | Unmarked image content may remain |
| Boxes look complete | A searchable duplicate may still exist elsewhere |
Inkfish checks extraction beneath marked areas after saving. That does not inspect an image-only page visually. Review document properties, comments, and attachments for information outside the page image too.
Watermarks and paper copies are separate delivery choices. Once the scan is checked, the contract-sharing guide helps choose the required format.
Common questions
Do I have to run OCR before redacting a scanned PDF?
No. A tool with image-area redaction can remove selected image content directly. OCR helps locate words. If the file contains an OCR text layer, check that layer too.
Why inspect the image if search and copy return nothing?
An unedited image-only scan can return no extracted text. Search and copy therefore cannot establish that its visible details have disappeared. Review every saved page.
