It’s never that much of a surprise when Google comes out with some new feature or update for its search algorithm. In fact, the company has made a name for itself in always striving for a better, more amazing product. When Google announced recently that it was introducing a new feature to its search algorithm called RankBrain, not many people outside the SEO community seemed to care. Even after Google pointed out that the new RankBrain system had been watching, tweaking, and changing search results on its own for more than a month, few people noticed anything different in their search results. Accordingly, there hasn’t been much in the way of public discussion.
Despite this lack of public awareness, RankBrain is one of the most significant advancements to Google’s algorithm since the Panda update, and if you want to preserve and improve your SEO strategy for the coming years, you’ll need to understand its full potential.
Before I get too deep into the tangible effects of the RankBrain system, let me explain the basics. RankBrain is a machine learning algorithm that works in conjunction with the Hummingbird Update to give better search results for user queries. The Hummingbird update included the then-new feature of “semantic search,” so rather than focusing on individual keywords in a user query, Google would be able to look at the entire phrase and user intention behind it. RankBrain takes this a step further by analyzing ambiguous, unclear, or otherwise indecipherable semantic user queries, learning from the experience, and applying that experience to future, similar queries.
For example, if you search for something like “what is the executive leader of the United States called?” you might get results about the position of “President.” On the other hand, if you search for something more ambiguous like “who is it that leads America in politics?” Google might struggle. With RankBrain, Google would be able to learn that the latter search query is actually just a less clear way of rephrasing the former search query, and would gradually shift search results to match. This is impressive because the majority of Google’s search algorithm updates, including Hummingbird, have been deliberately and painstakingly pre-programmed by human beings. RankBrain is going to learn, posit, and execute updates to itself over time, free of any human intervention.
Though RankBrain is only a few weeks old at this point, there are a handful of key misconceptions about what it is and how it fits into our current understanding of search.
First, understand that RankBrain isn’t a formal algorithm update. Unlike Panda, Penguin, or other landmark algorithm changes, RankBrain isn’t shaking up the ranking factors that Google considers when sorting out which sites to list first in the SERPS—instead, RankBrain is something of a ranking signal of its own. It’s working in conjunction with the Hummingbird update (which is an algorithm update) to produce a better understanding of queries—not a different selection of results. Think of it as a query translator.
Second, know that RankBrain is similar to, but distinct from, the Knowledge Graph. Because RankBrain and the Knowledge Graph are both forms of query service that analyze user queries and improve via machine learning over time, it’s easy to mix them up or assume that they work in the same way. However, RankBrain is focused on understanding queries, while the Knowledge Graph is focused on providing the best direct answers to certain queries. Think of it this way; RankBrain could serve to better understand your query, then pass it off to the Knowledge Graph for a proper and complete answer.
Users will continue using Google the way they always have, without any visual or experiential clues to suggest that anything different is happening behind the scenes. The end result is going to be better results for a greater number of queries, which is important, but shouldn’t provide any disruption to usual processes.
The nature of RankBrain means that the average user isn’t going to notice much of a difference. The changes it makes will be tiny, gradual, and will only exist for long-form user queries. Still, you might notice yourself getting better results in certain areas—apparently, RankBrain has already affected millions of queries (which isn’t that impressive for a search engine that processes more than 40,000 queries every second).
If you’re worried about what RankBrain will mean for SEO, don’t be. Because RankBrain is more focused on analyzing and mapping out user queries than it is sorting out sites for potential rank, you won’t have to change much about your current strategy. Best practices are still best practices, and there aren’t any strange new ranking factors to learn and implement. Still, don’t be surprised if you see some ranking shakeups coming out gradually in the next few months as RankBrain scales upward. Because ambiguous long-tail phrases are the main targets here, if your strategy revolves around long-tail keywords, you might see a small hit in overall ranks (or a small boost, depending on the queries). But it’s nothing to be concerned about, and it doesn’t demand any significant changes to your existing approach.
Ultimately, RankBrain is an impressive and significant addition to Google’s algorithm, but it isn’t going to revolutionize the world of SEO the way algorithm updates like Panda and Penguin have. However, this is a major step for Google in the introduction of machine learning to its central processes. Keep watch for other, similar machine learning segments of its algorithm to be introduced, and stay on your toes to react to those changes. You never know what Google will come up with next.