OCR

How to Make a Scanned PDF Searchable

Last updated · By the PDFSolution team

A scanned PDF is a photo album. Every page looks like a document but contains no text at all, which is why searching finds nothing and selecting picks up an image instead of a sentence.

Making it searchable means running optical character recognition over each page and storing the recognised words as an invisible layer behind the picture. The page looks identical; the difference is that search, copy and indexing suddenly work.

Ready to do it now? Open the tool this guide describes.

Make PDF searchable

Step by step

  1. 1

    Open Make PDF Searchable and add the scan

    Recognition runs in your browser using WebAssembly, so the document is not uploaded.

  2. 2

    Choose the document language

    This matters more than anything else. The wrong language turns accented and non-Latin text into noise.

  3. 3

    Run the recognition

    Each page is analysed in turn. Expect this to take real time — OCR is the slowest thing on the site, and long documents take minutes.

  4. 4

    Download and test it

    Open the result and search for a word you know appears on a middle page. That is the honest test of whether it worked.

Useful tips

  • Rotate crooked or sideways pages before running OCR; orientation affects accuracy far more than people expect.
  • 300 DPI black-and-white or greyscale scans recognise better than large colour ones.
  • Compressing a scan heavily before OCR damages exactly the fine detail the recogniser needs. Do OCR first, compress after.

Common problems

The recognised text is gibberish
Usually the wrong language, a very low-resolution scan, or a heavily stylised typeface. Fix the language first, then the scan quality.
Handwriting was not recognised
OCR reads print. Handwritten notes are outside what it can do reliably.
The browser froze on a long document
Split the file into sections, run each through, then merge the searchable parts back together.

Frequently asked questions

Does the page look different afterwards?
No. The original image is kept and the text is added invisibly behind it.
How accurate is it?
On a clean, straight 300 DPI scan of ordinary printed text, high. On a faxed, skewed or faint document, considerably less. Always spot-check before relying on it.
Is my document uploaded?
No. Recognition runs on your own device.
Which languages are supported?
A range of languages is available in the language selector on the tool page; pick the one the document is actually written in.
Can I edit the text afterwards?
Not directly — the visible page is still an image. Convert a searchable PDF to Word if you need to edit the wording.

Related guides