Google’s Garry Illyes has offered a detailed explanation of how RankBrain algorithm works. He has clarified many doubts related to the algorithm which have been expressed by others. One of the questions asked was whether RB utilizes UX signals of any type or not. In his reply, Illyes mentioned that RankBrain makes uses of search data in order to guess what a person may click while searching for something that hasn’t been entered on Google in the past.
Illyes also stated that RankBrain frequently resolves issues which Google used to face with conventional algorithms. RankBrain is more dependent on data acquired from user’s interactions with search results vs user’s interactions with content. He mentioned-
“It is a really cool piece of engineering that saved our butts countless times whenever traditional algos were like, e.g. “oh look a “not” in the query string! let’s ignore the hell out of it!”, but it’s generally just relying on (sometimes) months old data about what happened on the results page itself, not on the landing page.”
Illyes said that things like dwell time and CTR can make things more complicated and RankBrain makes use of on-page signals.