Why PDFs are different
A scanned PDF usually contains page images. OCR needs to render each page, recognize text, and optionally rebuild a searchable PDF layer.
PDF OCR guide
How scanned PDF OCR works
The live tool currently supports image OCR. This guide explains how PDF OCR should work and how to prepare scans.
Best preparation
Use straight pages, strong contrast, 300 DPI scans when possible, and avoid shadows or low-resolution phone photos.
Current workaround
Export a PDF page as JPG or PNG, then use the image OCR tool to extract editable text from that page.
Future server workflow
A production PDF worker should handle page rendering, OCR retries, file deletion, and searchable PDF output.