How to search and copy text in a scanned PDF
You press Ctrl+F in a PDF and it finds nothing, even though the word is right there on the page. The PDF is a scan: a picture of text. OCR fixes that.
Text or picture of text?
A PDF exported from Word contains real text: letters with positions, which you can search and copy. A scanned PDF — or one made from phone photos — contains only images. Try selecting a word: if you can only select the whole page as a block, it is a scan.
What OCR does
Optical character recognition looks at the image, finds the lines and words, and recognizes each letter. Modern engines like Tesseract use neural networks trained on each language, so choosing the right language matters: accents, ñ and ç are recognized much better.
OCR PDF adds the recognized words as an invisible layer exactly over the image: the PDF looks identical, but Ctrl+F, copy and screen readers now work.
Getting good results
Scan at 300 dpi if you can; straight, well-lit pages make a big difference. For phone photos, Scan to PDF removes shadows first. For small print choose the Accurate (300 dpi) setting. The result shows the average confidence: above 90 % the text is usually almost perfect.
From scan to editable document
Once you have the text, PDF to Word rebuilds paragraphs and tables in a .docx you can edit, and PDF to text gives you plain text. PDF to Word can also run OCR by itself when it detects scanned pages.