How to OCR a PDF: Make a Scanned PDF Searchable

To make a scanned PDF searchable, run it through OCR: choose the language of the text, click Make PDF searchable and download the result. The page looks the same, but you can now search, select and copy the text.

You scanned a contract, a bank statement or an old book, and now you cannot search it. You press Ctrl+F and nothing is found. You try to copy a paragraph and the whole page gets selected as one picture.

That is because a scanned PDF contains no text at all, only a photo of each page. OCR fixes this. This guide explains what OCR does, how to run it for free, and how to get accurate results.

What does OCR do to a PDF?

OCR (optical character recognition) looks at the picture of each page, recognises the letters and words, and adds them back into the PDF as an invisible text layer behind the image.

The page still looks exactly like your original scan. But now you can:

• Search for a name, number or phrase with Ctrl+F (or Cmd+F on Mac).
• Select and copy text to paste into an email or a document.
• Let screen readers read the document aloud, which helps people with visual impairments.

If you want to understand why a scan cannot be edited in the first place, read Why you can't edit text in a scanned PDF.

How do I make a scanned PDF searchable?

1. Open OCR PDF.
2. Click Select PDF file, or drop your scanned PDF onto the page.
3. Choose the language of the text in your scan. If the document uses more than one, choose each of them.
4. Click Make PDF searchable and wait a few seconds per page.
5. Click Download to save your new searchable PDF.

OCR runs inside your browser. Your document is not sent to our servers, and the reading software and language files come from our own website. If you want to know why that matters, see Are online PDF tools safe?

How do I get the best OCR results?

The quality of the text depends almost entirely on the quality of the scan. Before you run OCR:

• Scan straight. A page that is tilted by even a few degrees can confuse the reading software.
• Use at least 300 DPI for normal text. Lower resolutions make small letters blur together.
• Choose black text on a white background. Coloured paper, stamps and watermarks over the text reduce accuracy.
• Fix sideways pages. If any page is rotated, turn it first with Rotate PDF (our guide: How to rotate a PDF permanently).
• Pick the right language. The wrong language makes the software guess names and accents badly.

What kinds of documents work best?

Whatever the document, always check the important parts, such as amounts, dates and names, against the page image. OCR is very good, but never perfect.

How many pages can I do at once?

PDFixed reads up to 100 pages in one go. For a longer scan, split it into parts with Split PDF, run OCR on each part, then join them again with Merge PDF. Our guides on splitting and merging show how.

If your scan is made of separate photos or several files, you can first combine them into one PDF, then make the result searchable.

Can I turn the scan into an editable Word file?

OCR makes the text searchable and copyable, but the document is still a PDF. To get fully editable text, run OCR first, then convert the result. See How to convert a PDF to Word.

What if the text is still not accurate?

Poor scans, old documents, handwriting and unusual layouts are hard for software. When accuracy really matters, such as a legal contract or a document full of numbers, our experts can retype the pages by hand and match the original layout. Ask for a quote and we will tell you the price before any work starts.