Google RankBank Explained In Simple Language
Google is known for continuously refining its search algorithm. Each year it makes hundreds of minor changes but once in a while it comes out with major changes. Panda was released for the first time in February 2011, Penguin in April 2012, Hummingbird in August 2013 and 2014 saw the updates for Panda and Penguin. In October 2015, Google came up with a totally new way of ranking sites. It added a new factor in its search algorithm – the RankBrain. It has been revealed that RankBrain is the 3rd most important factor in its search algorithm. In the Bloomberg interview, Greg Corrado, a senior research scientist at Google, said:
“RankBrain is one of the “hundreds” of signals that go into an algorithm that determines what results appear on a Google search page and where they are ranked. In the few months it has been deployed, RankBrain has become the third-most important signal contributing to the result of a search query”.
It is widely believed that the first two factors remain content and links to the site.
What is RankBrain?
RankBrain is a machine learning system, which has become an essential part of Google Search algorithm. The only information that Google has given was in the video interview published on Bloomberg. Quoting Greg Corrado once again:
"All of the other hundreds of signals that are used are designed by Google's own engineers, whereas RankBrain is a learning system, it really is an Artificial Intelligence; it interprets language, it interprets your queries in a way that has some of the gut feeling and guessability of people".
RankBrain is not a new way in which Google sorts search engine results. It is in fact a part of the overall Google Search algorithm. Google has said that it has about 200 major signals and about 10,000 minor signals. Signals on a web page could be keywords in h1 or h2 tags, words which are in bold or keyword in domain, layout of the page, mobile friendliness, time taken to load page, keywords in alt tags, page ranking etc. The role of RankBrain will be to change the weightage, which is given to these signals.
How does RankBrain Work?
As far as we understand Google’s RankBrain works in conjunction with its Hummingbird. When a search query is made, the RankBrain first processes it through the machine learning system and passes the results to Hummingbird, which refines it, further. Thus, RankBrain is not a part of Hummingbird instead it complements it.
It is known that about 15 % of the search queries are made up of totally new search queries. It is challenging for Google to process these new search queries and display the most correct results. As a machine learning system, RankBrain teaches itself how best to process new search queries. If it sees a new search word or phrase, RankBrain can make a guess as to what the word could be related to, and filter the results accordingly.
RankBrain is useful for people who are not sure of what words to use for the search or use vague words, as it can guess what they are searching for with the help of similar meaning words. I quote what Jack Clark says about this on Bloomberg:
“Rankbrain uses artificial intelligence to embed vast amounts of written language into mathematical entities — called vectors — that the computer can understand. If RankBrain sees a word or phrase it isn’t familiar with, the machine can make a guess as to what words or phrases might have a similar meaning and filter the result accordingly, making it more effective at handling never-before-seen search queries.”
Is RankBrain Improving Itself?
The machine learning system is continuously learning and improving its results. How does it do this? It is said that Google feeds the system with past search data. The machine, then, analyzes the data for four important factors:
- Click Through Rate – If the top ranking results have poor click through rate, it obviously means that they are not popular.
- Bounce Rate – Did the visitor click ‘back’ from the page he was visiting?
- Google Analytics is used to analyze whether the visitor was satisfied with the page.
- Google also uses Google Chrome to find out the time spent on the page.
What does RankBrain do with all this feedback data?
It definitely tries to make sure that it does not show the sites, which have a poor click through rate or a high bounce rate or sites on which people spend very little time. Thus, the search results improve. These are then, released as new version of RankBrain and integrated into the Google Search Algorithm.
RankBrain And Its Effect?
RankBrain’s effect will most likely be felt by companies, which are trying to launch new products. The search for these new products may contain words, which have not been used before. In such a case, it is not sure how RankBrain may affect their ranking in existing search results.
RankBrain And Everyday SEO Practices?
RankBrain is not likely to affect everyday SEO practices. Any web page, which has good content and is following ‘white hat’ SEO is unlikely to be affected. Manipulative SEO, link farming, hiding keywords in same colored text will not work. The focus of SEO should be good content, no stuffing of keywords in content, incoming links, fast loading of page, popularity of the site on social media etc. One should basically follow the principle that ‘Google likes what users like’. So build your site for your visitors and not for search engines.