In an era in which the digital transformation is gaining increasing importance, it is essential to know and understand the different updates that Google has made in its search algorithm in order to use these updates in our favor and thus generate more visits to our page Web. Google has implemented more than 3000 updates to improve in one way or another the search results that users get when making a query; The most recent of them (published in 2018) considered one of the biggest updates of the Google algorithm is BERT.
BERT is the acronym for Bidirectional Encoder Representations Transformer, it consists of an open source neural network based on non-supervised natural language processing (NLP). Although these terms can be very confusing and technical, what is quite clear is that Google BERT is a system based on artificial intelligence (AI) that interprets much better what users want to say when searching and, therefore, offers more accurate results to the billions of users who search online content daily.
If so far Google analyzed word by word the expressions entered in its search box, with BERT the search engine is able to relate each and every word to properly understand the context in which they appear, and in that way can offer results more relevant It is also worth mentioning that Google BERT takes into account the interpretation of prepositions and other connectors, which until now considered orphan words of meaning for themselves in searches and, is also quite useful for voice searches.
According to media such as The Verge and Wired, Google searches after BERT have improved up to 10%, according to the results presented by users. This is something absolutely vital if we take into account that 15% of the daily searches carried out on Google are totally new; there is no prior information on whether the results shown match the user’s search intent.
Now, to understand the functionality of Google BERT, in this article we will describe its features, the benefits it provides and what you can do about SEO to get the most out of this Google update.
Google BERT features
The characteristics that make Google BERT represent a great innovation when compared to other algorithms such as RankBrain are the following:
Bidirectionality refers to the analysis of a phrase or sentence in two directions. This means that Google BERT analyzes the words that are located both to the left and to the right of a keyword, which allows you to understand in greater depth the context and theme of the entire phrase that a user enters for Google search compared to one-way language models.
In the following example: “I’d rather go to a cool place. I don’t like high temperatures”. In the first sentence the words “cool place” can refer to both a cold place and a fashion site. But, when analyzing the second sentence “I don’t like high temperatures”, it gives the necessary context so that the algorithm easily knows that the phrase is talking about temperature and not trends or fashion.
When analyzing the context in both directions, Google Bert understands the prepositions by showing results that were very well adapted to the needs of the users. A very common example is the case of the “2019 Brazil traveler to USA need a visa” search where a significant change is observed before and after BERT.
Before BERT, Google showed a Washington Post result for US travelers who wanted to travel to Brazil. The algorithm totally ignored the “to” connector, which in this case is essential to understand who travels to where. After BERT, when analyzing the context of the search we see that the result is much more rigorous; Google already understands that the traveler wants to go to the United States and, therefore, shows the result of the United States Embassy in Brazil.
In this way, Google is able to better understand the queries. BERT, seeks to process language more humanely and not as a robot as has been usual. And this is very advantageous in terms of SEO in the network.
Another of the innovations that characterizes BERT is a property called Transformers. It is a neural network that was initially used to improve automatic translations and consists of a mechanism for assigning scores that allows identifying the most correct meaning of a word within a particular phrase, in the case of words that have several meanings such as the word bank which can refer to a bank or urban furniture. These scores are assigned by analyzing, in a bidirectional way, the words located to the left and to the right of each of the terms.
Transformers also focus their attention on those words that make it difficult to understand the context, such as links or pronouns. In this way, they can understand references to pronouns, direct and even indirect complements. For example, in the following sentence: “My sister’s car is broken. I should pick her up”, BERT understands that “her” refers to “My sister” since she focuses on the pronoun and looks for the relationship between the two sentences.
Benefits provided by Google BERT:
Greater precision in the results. Thanks to BERT’s analysis of queries, Google’s results pages are much better suited to user interests. This means that keywords cease to be as relevant as individual elements, because from now on Google gives higher priority to search intentions instead of words that, by themselves, are decontextualized and fail to convey a specific idea.
Eliminates the worry of being penalized by the algorithm. An issue that is very beneficial for websites is that this new way of analyzing searches eliminates the concern of being penalized by the algorithm. Now you can work the SEO of your content and your website using a more natural language, instead of adapting your content to SEO that results in an artificial language too often.
Better understanding of the queries made in the search engines. The technology used by BERT allows Google to better understand all queries, since it uses context to understand phrases as a whole and not as a series of single words. With BERT Google aims to better understand how words are related to each other.
It offers better user experience. When users do a search on the Internet, they often don’t know what terms to use. Most of the time they build long sentences about the information they need. However, the results are not able to show a completely satisfactory response. This update means offering a better user experience to people. What it is about is to better understand the language used by people regardless of how the words are spelled or combined to find out what exactly the user is looking for.
For example, difficult queries on a topic that also does not have enough data: title of the book in which the protagonist travels to space with his dog, will be easier to interpret by BERT to offer an answer that fits the context.
Lets you understand prepositions better. Usually, prepositions alter the total meaning of the sentences. Therefore, BERT constitutes an attempt by Google to better understand users’ search intentions. For digital marketing specialists this also brings advantages. They can create content and use a more human language to position it. The searches from now on will be more comfortable and more natural.
Analyze featured fragments. The highlighted fragments are the answer boxes or the direct answers from Google that appear on the results page as soon as you make a query. While it is true that the operation of the featured fragments is prior to BERT, it is also true that with this new update the generated fragments will begin to be much more efficient, accurate and applicable for a larger number of queries.
How to get the most out of Google BERT
First of all, it should be noted that BERT is not a positioning factor, so we cannot optimize our website for BERT. When optimizing our website we must follow the same processes as always: make sure we have the right structure, that is fast, that does not duplicate content, that we use all the goals and labels correctly, among others.
BERT’s function will be to send us higher quality traffic by positioning our website when it meets the user’s needs even if the search keyword is not in our content. Unlike other algorithms, BERT does not penalize for violating the Google guidelines or reward for further optimization of the web. Its mission is to become a more human algorithm, understanding our language better and overcoming both grammatical and contextual limitations. In this way, it manages to respond to our needs in a much more rigorous and higher quality way.
Although it is not a classification factor in itself, we can apply certain changes in the way of writing the contents with the aim of positioning ourselves better in Google. In general, the main recommendation is that you start to develop much more conversational content, as if you were talking to your readers face to face. Do not focus on filling the text with rigorous keywords, which can cause you to lose your naturalness by leaving out prepositions, connectors and other linguistic elements.
Understanding the operation of an algorithm like Google BERT is crucial. It’s important to realize that, although it represents a great challenge, BERT is not a problem for SEO and, instead, it is an opportunity to achieve an optimal and sustainable organic positioning.
This Google update brings us information about the importance of developing truly human content, messages and interaction flows, which are positively valued by this artificial intelligence system focused on contextualization, which is able to identify the user’s intention. Contact us at Rocket SEO in Miami, we will ensure that your site is better positioned after this update and will run a FREE analysis for you.