Does your PDF need OCR?
A scan can show clear words while storing the whole page as a picture. Try selecting a sentence or searching for a word you can see. If neither works, OCR may help add a text layer.
A PDF that already contains correctly selectable text may not need OCR. Recognition reads the visible page again; it does not guarantee a better version of existing text.
Prepare a readable source
Use a straight, sharply focused scan with all page edges included. Shadows, faint print, a tilted photograph, and tiny characters make recognition harder. If you can obtain a clearer scan, use it before running OCR.
In OCR PDF, choose your document and the appropriate recognition language from the available options. Run OCR and allow processing to finish before downloading the searchable PDF. Large documents take longer because the work runs on your device.
Check words against the page image
The output keeps the visible scan and adds recognised text. An image can therefore look correct while copied text contains a mistake. Search for a known word, select a line, and paste it into a text editor to inspect the recognition.
For example, compare a date, a reference number, and an amount with the original image. Characters such as 0 and O, or 1 and l, can be confused. Tables, handwriting, and tightly packed print also need particular care.
OCR makes text easier to find and copy; it does not translate the document or verify its facts. Keep the original scan for comparison.