OCR: how to make a scanned PDF searchable and copyable

You search for a word in a 30-page scanned contract, but Ctrl+F finds nothing. You want to copy a paragraph from an old scanned letter, but selection doesn't work. You try to convert a PDF to Word and get an empty document. All these problems have the same cause: your PDF contains no text, only images of pages. The fix is called OCR.

This guide explains how OCR works and what affects its quality. This version of PDFtoWorld does not offer OCR and sends no documents to an external AI service. Its text extraction and conversion tools use text already selectable in the PDF.

What is OCR?

OCR (Optical Character Recognition) analyses the image of a page to recognise the letters, words and lines it contains, and turn them into computer text.

A PDF made by a scanner or from a photo is a series of images: you can read the text, but to the computer it is just pixels. After OCR, the document contains real text you can search, copy, read with a screen reader or reuse.

How do I know if my PDF needs OCR?

Open the PDF and try to select a word with your mouse:

• if the text highlights, the PDF already contains text: no OCR needed;
• if the whole page gets selected like an image, or nothing happens, it is a scan: OCR will help.

On a scan, the PDFtoWorld editor lets you add text on top with the Text tool. The image's contents do not become editable.

What you can do in this version

• Ask the author for the original digital file to obtain selectable text.
• Add text or annotations to a scan with Edit PDF.
• Convert a PDF that already has a text layer with PDF to Word.

Text conversions reject entirely scanned pages without usable text to avoid an empty result. A subscription does not unlock OCR.

Getting the best recognition

OCR quality depends mostly on image quality. For your scans:

• Resolution: 300 dpi is ideal for a text document; below 200 dpi, small characters become hard to recognise.
• Contrast: crisp black text on a white background gives the best results. Avoid greyish photocopies.
• Alignment: a straight page is recognised better than a skewed one.
• Phone photos: document flat, good light without glare, tight framing.

Printed text is recognised very well. Handwriting, stamps and highly stylised fonts are more hit-and-miss: always proofread important information (amounts, names, numbers).

What to do with a PDF already processed by OCR

If your document already has a recognised text layer before you import it, you can:

• Search the document with Ctrl+F in any PDF reader.
• Copy passages into an email or report.
• Convert to Word to rewrite the text, with PDF to Word.
• Edit the text directly in the PDF with Edit PDF.
• Manually redact personal information with Redact PDF, whether the document contains text or only images.

Common uses

Situation	What OCR brings
Digitised paper archives	Find a document by keyword
Scanned contract or lease	Search for a clause, copy an article
Invoices and receipts	Pick up amounts and references
Accessibility	Make the document readable by a screen reader
Theses, documentation	Quote passages without retyping them

Frequently asked questions

Does the searchable PDF look different?

No. The page stays the original image; the recognised text is added as an invisible layer aligned with the words.

Does OCR handle several languages?

Depending on the OCR software, several languages may be recognised, with or without selecting them first. Always proofread important passages. PDFtoWorld does not perform this recognition in its current version.

Can I run OCR on a simple photo?

Some OCR software can analyse a JPG or PNG photo. In this version of PDFtoWorld, converting a photo to PDF preserves its image without recognising its text.

Does OCR read handwriting?

Results vary greatly depending on the OCR software and the legibility of the handwriting. Proofreading is essential; this recognition is not offered here.

Why can't my scan be converted to Word?

A scan without a text layer contains only images. This version rejects entirely scanned pages without usable text. Use the original digital file or a PDF that already contains selectable text.