How the count works
The tool reads the text layer of each page and counts words the way a word processor does: a run of letters or digits, with an apostrophe or hyphen inside a word keeping it as one. Other scripts, such as Cyrillic or accented Latin, are counted properly.
What is included
Everything in the text layer — headers, footers, page numbers and footnotes as well as body text. If you need the body only, expect your figure to be a little lower than the one shown here.
Scanned documents
A scan is a picture of text, so it has no text layer and the count will be zero. The tool says so rather than reporting a misleading number; the text would need OCR first.