How to Make a Scanned PDF Searchable
Turn a flat scan into a PDF you can Ctrl+F through and copy text from, without changing how it looks.
Published by Filevolo · Report a correction
A scanned PDF is really just a picture of a page. It looks like a document, but there's no actual text behind it - Ctrl+F finds nothing, and you can't select or copy a single word. Fixing that takes two tools, and you only need the first one if the scan came in crooked.
Straighten it first, if it needs it
A page scanned slightly off the bed, or photographed at an angle, throws off text recognition more than you'd expect. If your pages have a visible tilt, run the file through Deskew PDF first - it detects each page's skew angle and straightens it automatically. Skip this step entirely if the scan is already straight; a page with no real tilt is left untouched rather than nudged for no reason.
Add the searchable text layer
Next, run the file through OCR PDF. It recognizes the text in each scanned page and adds it back in as an invisible layer, positioned exactly where the real words sit on the image. The page still looks identical to the original scan - nothing visible changes. What changes is that the text underneath is now real, machine-readable text.
What you actually get
A PDF that looks exactly like your scan but behaves like a normal document: Ctrl+F finds words on the page, you can select and copy text, and it shows up correctly in another program's search results. How precisely the invisible text lines up depends on scan quality - a clear, straight, well-lit page lines up closely, while a blurry or low-resolution one can shift slightly. Since the added layer is invisible, a small misalignment never affects how the page looks, only how exactly a search or selection lands.
Just want the text, not a searchable PDF?
If you don't need to keep the document as a PDF and just want the words themselves, use PDF to Text instead - the same on-device OCR engine, but it pulls the recognized text out into a plain text file rather than embedding it back into the original.
Common questions
Does this change how the page looks? No. The recognized text is invisible - it sits behind the image, not on top of it.
Does OCR work on any scanned PDF? It's built for full-page image scans, the typical result of scanning paper. A page that already has real text, or several separate images rather than one full-page scan, is left untouched instead of guessed at.
Is my file uploaded anywhere? No. Both tools run entirely in your browser - deskew detection and OCR recognition happen on your device, nothing is sent to a server.