syniriaCustomer
Service Agent
Start free

Why a scanned PDF is rejected, and how to fix it

Your agent can only learn from PDFs with selectable text; here's how to make one.

Updated

If you upload a PDF that's really a picture of a document, such as scanned pages, a photographed menu or an exported image, you see: "This PDF appears to be scanned or image-only, so we can't read any text from it." The file is rejected and nothing is stored.

How to tell before uploading: open the PDF and try to select a line of text. If you can highlight the words, it's fine. If your cursor draws a box over a picture, it will be rejected.

How to fix it:

  1. Export from the original. If the document started in Word, Pages, Google Docs or Canva, export a PDF straight from there instead of scanning a printout. The text comes out perfect.
  2. Run it through OCR. Most scanner apps, Adobe Acrobat and many free tools have a "make searchable" or OCR option. Upload the searchable copy.
  3. Type it in. For a short document, putting the content in your business description or FAQs is quicker and more accurate.

The check is generous: a PDF needs only about 20 characters of real text a page to pass. If it fails, there's essentially no text in the file.