How to Make a Scanned PDF Searchable With OCR

Turn a scanned PDF into a searchable, copy-able document by running OCR.

3 min readOCR a PDF

A scanned PDF is a picture. Search returns nothing. Copy-paste returns nothing. The text exists visually but the computer can't read it.

OCR (optical character recognition) reads the visible text and adds a hidden text layer behind it. The scan looks the same; the document becomes fully searchable.

When OCR is needed

If you can't select text in your PDF, it needs OCR. Common cases: scanned contracts, photographed documents, screenshots saved as PDF.

Live-text PDFs (exported from Word, Pages, etc.) don't need OCR — the text is already there.

Run OCR in the editor

Open the PDF in Flint's editor and run OCR. The tool processes every page and adds a recognised text layer.

Specify the language for best accuracy. English defaults to fast and accurate; other languages slightly slower with similar accuracy.

Verify and clean up

Try searching for a word you know is in the document. If results land on the right page, OCR worked.

For key fields (numbers, names, dates), spot-check the recognised text for accuracy. OCR isn't perfect — 90-95% accurate on clean scans, less on messy ones.

FAQ

How accurate is OCR?

90-95% for clean printed text. 70-85% for low-contrast or skewed scans. 40-70% for handwriting.

Will OCR change how the PDF looks?

No. The visible scan is unchanged. The text layer is invisible and sits behind the scan.

Can I OCR a multi-language PDF?

Yes — specify multiple languages. Accuracy drops slightly versus single-language.

Does OCR work on photos of documents?

Yes. Convert the photo to PDF first, then OCR. Cleaner photos give better results.

OCR is the difference between a picture and a document. Run it in Flint's editor and your scan becomes a real searchable file.

Try it now

Drop a PDF in and you'll be done in seconds — no install, files private to your account.

More on this

How to Make a PDF Searchable | Flint — Flint PDF