May 24, 2021

Understanding Google RankBrain

Google-RankBrain

Google SEO is a skill that has been in demand for a long time now. Companies and agencies are investing more and more money to stay relevant. SEO has immense business potential, especially when considering that 75% of users never even browse the second page of search results.

It is essential for any business to appear on the first page – and preferably on the top – to simply get found and have a chance at the potential profits.

However, for everything that is beautiful and wonderful on the outside, there is an intricate machinery operating underneath. Let’s look at it from the perspective of an average person searching on Google. Have you ever wondered what happens when you search for something on Google?

An elaborate machinery kicks in and starts searching for the results that would be of most value to you. This machinery is driven by algorithms, two of the most important ones of which are HummingBird and RankBrain.

Jump to a section…

How does Google Hummingbird work?
What is Google RankBrain?
What is Machine Learning?
How Does RankBrain Work?
Rankbrain and Hummingbird
The Benefits of RankBrain
How Google RankBrain Influences the Way We Do SEO
BERT vs Rankbrain
How to Optimize for RankBrain?
Conclusion

How does Google Hummingbird work?

When you search for something on Google, an algorithm at the backend reads your search query and tries to make sense of what you are looking for. Hummingbird handles this job today.

Hummingbird algorithm is different from the algorithms in the past because it is conversational in nature. When Hummingbird faces your search query, it reads the entire query and tries to make sense of it as a complete sentence. It is commonly referred to as Semantic Search, in contrast to the earlier approach of trying to make sense of individual segments of the search query.

What is Google RankBrain?

Rankbrain was designed to improve search results for ambiguous or poorly worded long-tail queries and new queries that Google has never seen before.

Greg Corrado, a senior research scientist at Google, was the first to reveal existence of the RankBrain program in an interview to Bloomberg in October 2015.

From the story:

“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.”

Based on machine learning, RankBrain is an impressive imitation of the human cognitive ability. It creates and analyzes user profiles to understand what kind of content would be most relevant to a user. The following are some of the aspects of Google RankBrain you might be interested in learning more about.

What is Machine Learning?

One of the most profound applications of Artificial Intelligence, machine learning involves empowering the machine to analyze patterns and results so that it can teach itself to perform certain tasks more efficiently. The algorithm works harder with new data and search queries and becomes smarter in the process, eliminating the need for human intervention.

How Does RankBrain Work?

RankBrain analyzes your search patterns, location, history and how you interact with the results of search queries. It then uses this profiling to recognize what each user is looking for when they type a search query. It attempts to sort results based on the relevance to a user.

rankbrain flow explicited

Image credit: Damian Fanish

As an application of machine learning, Google RankBrain improves itself by doing three essential steps offline:

  • Processed historical data is used to understand how to make better predictions.
  • The resulting predictions of this experiment are put to test by RankBrain to check relevance.
  • If the predictions prove substantial and accurate, RankBrain updates itself online with the required changes to become more efficient in sorting results.

Rankbrain and Hummingbird

Google RankBrain builds an extensive knowledge-base by profiling users. It analyzes the meaning and intent of search queries for the user and then uses its understanding to sort through the results of Hummingbird’s Semantic Search, to deliver the most relevant content.

The Benefits of RankBrain

Rankbrain likely brought solid improvements in several fields:

  1. Improved tolerance to complex long-tail keywords – Users searching for unusual and complex queries are likely to find satisfactory search results more quickly (without the need to run consecutive queries).
  2. A better ability to handle Voice Searches – Voice search technologies like Siri, Google Now, and Cortana should return more accurate results thanks to Rankbrain.

Google has officially acknowledged that it uses almost 200 factors or signals to decide how relevant a result is for a user. RankBrain is ranked 3rd out of the 200 signals.

Google estimated that around 3 billion search queries are encountered every day, out of which 15% have never been seen before. RankBrain actively helps in processing these.

An interesting example is the phrase ‘why are pdfs so weak’:

google rankbrain results before and after

Image credit: stonetemple.com

Looking into January’s SERPs (right screenshot), it seems Rankbrain succeeded to interpret the ‘weak’ adjective in the context of security encryption and returned results accordingly.

In a test to determine what is more effective in understanding the relevance of content, RankBrain scored 80% against the engineering team’s 70%. RankBrain managed to beat the engineers who designed it!

How Google RankBrain Influences the Way We Do SEO

RankBrain has shown proven results for effectiveness when compared to previous algorithms. By far, its most significant victory has been reading and understanding what is more relevant and helpful to a user. That is precisely where the focus of SEO strategies should be now.

The need of the hour is to make SEO value- and audience-centric. Focus on what the user thinks of your content.
If your content is what the user interacts more with and if it provides them the value they seek, you have got the winning formula. It’s that simple.

BERT vs Rankbrain

BERT became the next big Google update launched in October 2019 since Rankbrain. BERT stands for Bidirectional Encoder Representations from Transformers and it is Google’s neural network-based technique for natural language processing (NLP) pre-training. BERT and Rankbrain have the common goal to better understand the intent behind the query and serve better search results according to that intent.

Rankbrain looks both at users’ queries and web pages’ content to interpret search queries and makes a guess by finding similar queries that performed well before, so the resulting page may not even contain the exact phrase that was used for search.

BERT enables Google to better understand natural language: longer texts, conversational queries, the nuances in a context of words, and connections between the words. BERT and Rankbrain are complimentary algorithms and can be used separately or combined for the best possible query phrase interpretation.

How to Optimize for RankBrain?

The answer is also simple: provide value to the users. Let’s look at a few ways to achieve this goal:

Freshness

You are looking for tips to improve your Photoshop skills. You do the search, and here are two typical results:

One dated March 10th, 2010

Another dated April 1st, 2018

Obviously, the more recent one would be more relevant to a user, considering that software and technology keep upgrading year after year. So, try to write new content, or change your old content and make it relevant!

Keywords

Do not use keywords that sound weird and unnatural. Shift your focus to building a set of natural keywords. As usual, long-form ones are more conversational in nature and tend to be picked up more easily by Hummingbird (and RankBrain, by extension). Try to create variations of such keywords and use the entire set to make content that feels natural.

Conversational Writing

Do you know what kind of content is the most engaging for an average user?

Content that is clever, occasionally spiced with humor (not dry), and is human in tone.

Create content that engages users and keeps them on your page. When your content speaks to the user, they might even speak back (in the form of comments). Promote dialogue with your content.

Conclusion

Nowadays, RankBrain is a major factor dictating the SEO game and it is here to stay. While it has made SEO work a bit different than how it used to be, there is definite value in it. Google’s role in serving consumers with the most relevant and helpful content has been revolutionary.

RankBrain is in-line with that role and ensures that with the ever-increasing data, users continue to find content that is most relevant to them and resolves their needs. It is time for the players to change their SEO strategies and benefit from Google RankBrain.

If you have questions on RankBrain or other SEO matters, feel free to contact our SEO team

BACK TO BLOGBACK TO BLOG
Let's talk Marketing
Get in touch!