BASALT · JOURNAL
Making a scanned PDF searchable on a Mac
A scanned page is a photograph. Searching it finds nothing, because there is no text in the file, only an image of text.
Check whether you already have a text layer
Open the document and try to select a word. If your cursor draws a box rather than highlighting characters, there is no text.
Faster, from the command line:
pdftotext document.pdf - | headEmpty output means no text layer. A document that returns nothing for a common word like "the" is a scan.
Adding one
OCR reads each page image and writes an invisible text layer aligned over the picture. The page looks identical and becomes searchable, selectable and extractable.
On macOS this runs entirely on device through the Vision framework, so it needs no internet connection and nothing is uploaded. Any native application can use it.
What to expect from accuracy
Good on clean typed documents. Worse on poor scans, unusual fonts, handwriting, stamps and anything skewed.
The practical consequence is that a search over an OCRed document can miss pages where the term is plainly visible, because recognition got a character wrong. Treat search as an aid rather than proof of absence, particularly when you are searching for material to remove.
The trap that matters
Once OCR has run, the document contains real machine readable text sitting invisibly under the image.
If you later redact that page by destroying pixels but leave the text layer alone, the result looks perfectly redacted and remains fully extractable. Anyone can recover the words with a command line tool.
This is a documented failure, not a theoretical one. The order that avoids it is: OCR first, then review and mark, then redact both the image and the text, then verify by extracting text from the finished file and searching it.
Before you OCR a large production
It is the longest single step on a big scanned set, because recognition analyses every page. Run it as a background job rather than waiting on it, and do it before rather than after any redaction work.
Frequently asked questions
How do I make a scanned PDF searchable on Mac?
Run OCR, which reads the page images and adds an invisible text layer aligned over them. On macOS this runs on device through the Vision framework, so no connection is needed and nothing is uploaded.
How do I tell whether a PDF has a text layer?
Try to select a word: if the cursor draws a box instead of highlighting characters, there is no text. From the command line, pdftotext document.pdf - returns nothing for a scan.
Does OCR need an internet connection?
No. Recognition can run entirely on device. Cloud OCR exists and is sometimes more accurate on difficult material, but a connection is not required to make a scan searchable.
Can OCR text survive redaction?
Yes, and it is a real failure. Destroying the pixels of a scanned word while leaving the invisible OCR text in place produces a document that looks redacted and stays machine readable. Both layers must be handled, and the output verified.
Doing it in Basalt
Basalt is a native macOS PDF toolkit: eighteen tools in one window covering merge, split, page organisation, compression, OCR, passwords, Bates numbering, forms, signing, watermarks and comparison. Every file is processed on your Mac, and the engine that opens documents holds no network entitlement at all, which macOS enforces at the code-signature level. Redaction destroys content rather than covering it, and an independent verifier proves the material is gone before a file is written. A one time $29 licence covers up to three Macs, free for the first 24 hours. Get it at basaltformac.com, or brew install --cask chipmunk1101/tap/basalt.
Redaction that proves itself
Basalt destroys the content you mark, then re-opens the file it wrote and proves the content is gone before it saves anything. Your documents never leave your Mac.