How to make a PDF searchable or extract its text
Choose between text extraction and OCR recognition: a practical example for digital PDFs, scans and mixed documents.
Start with a word
Open the document in a PDF reader and try selecting and copying a word. If you get readable text, the PDF already contains a text layer: PDF to text can extract it without optical character recognition. A scan may already have an OCR layer, so its photographic appearance alone is not enough to tell.
How to make a PDF searchable
If you can't select the words, the page may be an image. PDF OCR analyzes the image and recognizes characters in the 12 available languages. Choose TXT for copyable text or searchable PDF to maintain the look of pages with a text layer.
The first use downloads the engine from the site; the scan remains on the device. Work on a few pages at a time, up to a limit of 20, especially on your phone.
An example with a mixed document
A document contains two digital pages and three scanned receipts. Use PDF to text on the digital pages and OCR only on the receipts to avoid recognizing text that is already available. The OCR result includes only the selected pages.
Check numbers and reading order
Compare the text with the original, especially dates, amounts, names and codes. OCR does not always recognize tables and columns correctly; TXT extraction does not preserve fonts or layout. For a visual copy of the page, use PDF to JPG and PNG.