Why not just check for text?

A document can have a text layer and still be mostly images. Checking both is more reliable.

It says partly scanned. What does that mean?

Some pages are generated and some are scanned, which is common in assembled documents.

What should I do about it?

Run OCR to make the scanned pages searchable.

inspect

Is My PDF a Scan

Tell whether a PDF is a real document or a picture of one.

Runs in your browser

How it works

Combines two signals: whether pages carry a text layer, and whether each page is a single large image. Together those distinguish a scan from a document reliably, including the awkward middle case of a document that is partly both.

Related tools

PDF Text Layer Detector

/pdf-text-layer-detector

OCR a PDF

/ocr-pdf

Compress PDF

/compress-pdf

Questions

Why not just check for text?: A document can have a text layer and still be mostly images. Checking both is more reliable.
It says partly scanned. What does that mean?: Some pages are generated and some are scanned, which is common in assembled documents.
What should I do about it?: Run OCR to make the scanned pages searchable.