From the readings
What comes out of actually measuring a lot of sites, including the results that did not fit the theory.
Findings pulled out of the published readings dataset rather than out of best practice.
Seventeen sites, one thermometer, and three results that did not fit
The first run of the published readings dataset produced three results that argue against the instrument that produced them. Hacker News at 47. MDN at 29 for entity strength. The site that publishes the federation specification at 74, on 255 words.
The page that scored zero, and why that is the right answer
Example.com returns 21 words and scores zero for Content and Entity Strength. It is also a perfectly working page doing exactly its job. Both of those are true and the scale has to hold them at the same time.
Why a good site from 2019 reads hot for search and cold for AI
The most common shape in the readings is a site scoring in the eighties for Search Visibility and the forties for AI Discoverability. Nothing it did was wrong when it was built. That is the whole problem.