On October 25, Google announced that its latest algorithm update, nicknamed the BERT update, had been rolling out over the course of the previous week. 

Though Google continually makes changes to improve its algorithms and launches significant updates several times a year, the BERT update was labeled by Google as one of the biggest in years and is likely to be particularly impactful for a few reasons.



What is BERT?

BERT was first announced in November 2018 as Google’s latest advance in natural language processing (NLP) – i.e., how Google understands human language to provide the best search results for various queries. 

BERT, an acronym for Bidirectional Encoder Representations from Transformers, is more advanced than previous NLP techniques because it is the first model to use “deeply bidirectional, unsupervised language representation” according to Google. That means that BERT can derive meaning using surrounding context on both sides of a sentence or phrase’s primary keyword(s). 

Google gives this example to understand how bidirectional NLP works:

“…in the sentence “I accessed the bank account,” a unidirectional contextual model would represent “bank” based on “I accessed the” but not “account.” However, BERT represents “bank” using both its previous and next context — “I accessed the … account” — starting from the very bottom of a deep neural network, making it deeply bidirectional.”

The Google AI Blog, November 2018

What Does BERT Mean for Search?

The BERT Update enables Google to use this NLP technology to understand queries in a more natural, human way. 

In their announcement, Google provided the following example of before-and-after screenshots of how BERT would impact a query:

Example of BERT query set up

With BERT, Google is now able to understand the importance of the preposition “to” for understanding the meaning of the query. Rather than showing general queries about Brazil, visas, and travelers (the primary “meat” of the query), Google can now understand the specificity of the search and show content that fully answers the query, not just relates to it. 

BERT can also impact Featured Snippets by using contextual clues to provide better featured results based on bidirectional NLP.

How Is BERT Different Than Other Algorithm Updates?

Google has indicated that the significance of the BERT Update as it relates to query processing is major.

Google calls this update “the biggest leap forward in the past five years, and one of the biggest leaps forward in the history of search.” Google claims that BERT affects roughly 10% of all searches, meaning that there’s a good chance your website will be impacted in one way or another.

However, BERT doesn’t replace or repeal other algorithm updates, so there aren’t any changes to existing SEO best practices that webmasters need to be aware of.

In addition, brands shouldn’t worry about seeing major ranking drops across the board like they might have with algorithm updates that penalized low-quality sites or sites engaging in practices that are against Google’s guidelines. 

What Should Webmasters Do Now?

As with many algorithm updates, Google’s advice to SEOs and webmasters is to create great content that aligns with user expectations, rather than optimizing for a specific algorithm. The best strategy is to continue creating high-quality, trustworthy content that users can understand and engage with, and ensure that they’re placed on a site that gives search engines the best technical opportunities to discover and evaluate content.

In addition, there is always the potential for “aftershocks” following  updates of this scale as ranking fluctuations stabilize, so we recommend keeping a diligent eye on traffic and rankings data using Google Search Console and your analytics platform of choice.




Jacquelyn Green is an SEO Manager for Response Mine. With experience in technical SEO and content marketing, Jacquelyn loves working with clients to develop and execute unique organic search strategies that meet their business goals. When she isn’t studying SERPs or optimizing meta data, Jacquelyn enjoys painting, watching reality television, and hoping people will ask to see a picture of her dog.
4.7/5 (3 Reviews)

Subscribe to the Blog