5 entries
Mark Williams-Cook, Candour / AlsoAsked
Google assigns site quality scores at subdomain level, sorts almost all queries into 8 semantic categories, and scores passages for agreement with consensus.
Rand Fishkin and Michael King (source: Erfan Azimi), SparkToro / iPullRank
Google stores far more than it publicly discusses, including click-based measures, site authority values, and per-document quality signals.
Pandu Nayak (testimony), US Department of Justice (proceedings), Google / US DOJ
Navboost re-ranks results using aggregated click data over a rolling 13-month window, and internal documents describe it as one of Google's strongest signals.
Pandu Nayak, US DOJ v. Google
Nayak agreed RankBrain "looks at the top 20 or 30 documents and may adjust their initial score", that it is "certainly more expensive than some of our other ranking components", and that it is "too expensive to run on hundreds or thousands of results". On DeepRank he testified that "training transformer models is more expensive than training straightforward feedforward networks like Google Brain or RankBrain. Those are cheaper to train."
Google Search Central
The guide opens: "If your site doesn't have a large number of pages that change rapidly, or if your pages seem to be crawled the same day that they are published, you don't need to read this guide." It targets "Large sites (1 million+ unique pages) with content that changes moderately often (once a week)" and "Medium or larger sites (10,000+ unique pages) with very rapidly changing content (daily)", calling the numbers "a rough estimate" rather than exact thresholds.