How to make a scanned PDF searchable on Mac

Check the document before running OCR

1. 1

   Search for a distinctive word

   Open the PDF in Preview, press Command-F, and search for a word you can see on a page. Test more than one page: a document can contain both ordinary text pages and scanned pages.

2. 2

   Try copying a sentence

   Paste a selected sentence into a plain-text document and compare it with the page. Live text recognition in a viewer can help select text, so selection alone is not proof that the PDF contains a persistent text layer.

3. 3

   Prepare a readable source

   Rotate sideways pages and use a clear scan. Around 300 dpi is a good starting point for printed text. If the words are blurry, rescan the page rather than enlarging the image.

Create a searchable copy with NeatClick

The action supports unlocked PDFs up to 500 pages. Use a document you can open normally; password protection and an excessive page count need to be resolved before processing.

1. Select the PDF in Finder and choose NeatClick → Make Searchable. If the NeatClick menu is missing, enable its Finder extension using the setup instructions on the Support page.
2. Wait for the job to finish and open the generated PDF, rather than reopening the source. The recognized text is added as a searchable layer; the visible page still looks like the scan.
3. Use Command-F for a name near the start, a phrase in the middle, and a term near the end. Copy a paragraph into a text editor to inspect the recognition and reading order.

What to verify in a real scan

For an invoice, compare the invoice number, date, decimal points, and total against the page image. OCR can confuse 0 with O, 1 with l, or omit punctuation. A search for “1080” may fail if the recognized text says “1O80”.

For columns or tables, check that pasted text follows the intended order. Searchable text is useful for finding pages, but it is not a reliable reconstruction of a spreadsheet or an editable word-processing document.

When recognition is incomplete

If only some pages fail, inspect those pages for rotation, faint text, shadows, or small print. Rescan from the paper source if possible, then run OCR again from the improved source rather than repeatedly processing the same poor scan.

Handwriting and complex layouts may still need manual transcription. If Preview finds the words but Finder does not, the OCR may be fine: system indexing is a separate process and may not yet include that file or location.

If you only need the text

Use Copy Text or Save PDF Text when your goal is to paste or save the words. Use Make Searchable when you need to keep the document as a PDF with a text layer. OCR can still misread words and numbers. Making a PDF searchable also leaves any private information on the page visible.

Frequently asked questions