Pagewright PDF Editor
Buy for $19.99
Guide

How to make a scanned PDF searchable

A scanned PDF is a picture of a page, so you can't search it or copy its text. OCR (optical character recognition) fixes that — and in Pagewright it runs on your own computer.

Updated · Screenshots from Pagewright on Windows

Before you start

You need Pagewright on Windows 10 or 11 and a scanned PDF with printed text. English recognition is built in. For another language, see Other languages below.

1. Open the scan

Open the PDF in Pagewright. If you try to select its text, nothing happens — the page is an image.

A scanned letter: an image of text, with nothing to select or search.

A scanned letter: an image of text, with nothing to select or search.

Turn sideways pages upright first (Organize → Rotate), and use clean, straight scans where you can: OCR is most accurate on clear, upright printed text.

2. Run Recognize text

Choose OCR → Recognize text. In the dialog:

  1. Tick the languages printed on the pages. Fewer languages are faster and more accurate.
  2. Choose the pages (all pages by default).
  3. Keep Add searchable text layer to this document ticked. Skip pages that already contain text avoids adding text twice.
  4. Click Start OCR.
The OCR dialog after recognition, with the number of words found.

The OCR dialog after recognition, with the number of words found.

Pagewright reports how many words it recognized. If some words have low confidence, they are listed, and Outline low-confidence words on the page shows where they are so you can check them.

3. Save a searchable copy

Save the document (Save As keeps the original scan file unchanged). The text layer is invisible: the page looks exactly as before, but now its text can be searched, selected and copied in Pagewright and in other PDF readers.

4. Search the text

Press Ctrl+F and type a word or number from the page. Results are listed with their context and highlighted on the page.

Search finds the order number, which exists only in the scanned image.

Search finds the order number, which exists only in the scanned image.

Other languages

Pagewright includes English. To recognize another language, open Add a language in the OCR dialog (or Preferences → OCR), choose Import file… and pick a Tesseract 4 language file such as deu.traineddata or tha.traineddata.gz. Imported languages appear in the list, ready to tick.

Good to know

  • OCR runs entirely on your computer; the document is never uploaded.
  • Handwriting, tables and badly degraded scans may not be recognized reliably. Check important numbers before you rely on them.
  • OCR text is for searching and copying. To change wording in a digital PDF, use Edit text.
  • Tick Also export recognized text to a .txt file in the OCR dialog if you need the text elsewhere.

Related: OCR and conversion — feature details