On October 26, Bloomberg Business published a spotlight on Google’s investment in artificial intelligence (AI) and machine learning — specifically RankBrain. RankBrain is an AI system that Google has been using for the past few months to process a large percentage of its daily search requests.
RankBrain represents Google’s ongoing quest to ensure the most relevant search results for Google users. There are no immediate actions for marketers to take in response; RankBrain simply reinforces the need to follow SEO best practices. Websites with quality content and good SEO will be positively impacted, and we can expect negative implications for those that still rely on spam tactics.
HOW DOES RANKBRAIN FACTOR INTO GOOGLE’S SEARCH ALGORITHM?
With machine learning, a computer teaches itself instead of being programmed by humans. (In Google’s case, it teaches itself to predict what people are searching for.) RankBrain can track what users click on and then interpret searches to find relevant content, even if the exact words that were searched are not found in the results (source). In short, RankBrain puts machine learning into play by recognizing patterns much faster than humans, allowing Google to serve up more pertinent results based on searchers’ queries.
Google’s search algorithm is already complicated enough, and now this new ranking signal — RankBrain — has been thrown into the mix. A ranking signal is a website characteristic (for example, mobile-friendliness) that is evaluated by search algorithms to determine where websites should be ranked within search results. According to Search Engine Land, Google has more than 200 major ranking signals and might also have as many as 10,000 variations or sub-signals, but this can’t be confirmed. Google has stated that RankBrain is the third-most important factor for ranking webpages. Although we don’t know what the first- or second-most important ranking signals are, knowing that RankBrain is number three gives us some insight on how important this AI system is.
HOW ARE SEARCH QUERIES AFFECTED?
- RankBrain refines searches — Google processes 3 billion searches per day and fifteen percent of those queries are ones that have never been searched on Google before. RankBrain enables Google to better interpret long-tail queries (i.e., extremely specific searches that include several words). Having a system like RankBrain to interpret and translate these queries cuts down on a lot of human legwork and helps Google more effectively understand future complex searches and their relationships to particular topics.
- RankBrain is not a new algorithm but rather a ranking signal — Hummingbird is still Google’s algorithm. RankBrain is a component within Google’s overall algorithm and “directly contribut[es] somehow to whether a page ranks.” Whether there is a RankBrain scoring system to evaluate content quality is unknown. Regardless, RankBrain appears to help Google more efficiently classify pages based on the content they contain.
- RankBrain likely enhances Knowledge Graph results — Google’s Knowledge Graph was launched in 2012 as a way to better connect search queries as “things, not strings.” In other words, it is a system that understands facts about people, places and things and how these entities are all connected. Given that RankBrain enhances the connections between search queries, intents and patterns, it is likely that it will also enhance Knowledge Graph results.
At the end of the day, Google is constantly trying to find new ways to perfect their algorithm and make sure they are serving up the most relevant and useful search results for their users. Overall, RankBrain helps users by giving them a better experience in being able to get better search results. With that, marketers and SEO professionals should still be focusing their efforts on SEO best practices, including:
- Implementing good, relevant content
- Focusing on content depth (i.e., semantically relevant topics, since Google’s algorithm has continually moved away from specific keywords)
- Evaluating your website’s content on a page-by-page basis (RankBrain allows for a more precise and customized approach to interpreting and ranking webpages. The more artificial intelligence is a focus, the less Google will be making sweeping, broad-stroke changes to its algorithm where several pages and sites are affected all at once.)
- Designing websites that load quickly and have an overall positive user experience, which includes making your site mobile-friendly
In the future, RankBrain’s machine learning technology will likely more positively affect sites with quality content and good SEO and negatively affect sites still relying on spam tactics. Like everything else in the search space, we can expect for Google to continually tweak its RankBrain technology. The Intouch Solutions’ inbound marketing team will keep an eye out on how this emerging technology continually affects the overall search landscape as well as the implementation of best practices.