Google is now using its RankBrain machine learning system to process every query that the search engine handles, and the system is changing the rankings of lots of queries.
The news emerged this week as part of Steven Levy’s Backchannel story about machine learning efforts at Google. From the story, in regard to RankBrain:
Google is characteristically fuzzy on exactly how it improves search (something to do with the long tail? Better interpretation of ambiguous requests?) but Dean says that RankBrain is “involved in every query,” and affects the actual rankings “probably not in every query but in a lot of queries.”
What’s more, it’s hugely effective. Of the hundreds of “signals” Google search uses when it calculates its rankings (a signal might be the user’s geographical location, or whether the headline on a page matches the text in the query), RankBrain is now rated as the third most useful.
We’ve already heard before that RankBrain is considered the most useful search ranking signal, behind content and links. But prior to this, Google had only said publicly last October that RankBrain was used to process a “large fraction” of 15 percent of the searches it had never seen before.
In short: Google’s clearly become so confident in RankBrain’s mysterious capabilities that it’s now used to help with every query that the search engine handles, more than two trillion per year.
What RankBrain may be doing
What’s not happening is that RankBrain actually changes the rankings of search results for all those queries, but rather for “a lot” of them, as stated. How can that be?
That fits in with what we’ve understood about RankBrain: it seems largely used as a query refinement tool. Google seems to be using it now for every search to better understand what that search is about. After that, another aspect of RankBrain might influence what results actually appear and in what order, but not always.
Imagine that RankBrain sees a search for “best flower shop in Los Angeles.” It might understand that this is similar to another search that’s perhaps more popular, such as “best LA flower shops.” If so, it might then simply translate the first search behind the scenes into the second one. It would do that because for a more popular search, Google has much more user data that helps it feel more confident about the quality of the results.
In the end, RankBrain did change the ranking of those results. But it did that simply because it triggered a different search, not because it used some special ranking factor to influence which exact listing appeared in what order.
Having said that, Google has said that RankBrain also is used as an actual ranking signal, repeatingthat yesterday at our SMX Advanced show.
For the SEO and search marketers worried about what they should do now that RankBrain has ramped up, the answer remains the same: nothing, but focus on great content. Even people at Google don’t quite understand how RankBrain does what it does, we’ve been told. Honest. But it’s ultimately designed to reward great content. So focus on that, which has always been the case with SEO, and you’re on the right track.
And to understand more about RankBrain, see our FAQ: All About The New Google RankBrain Algorithm article.