If you’ve kept a pulse on SEO industry trends over the last year you’ve probably seen the acronyms BERT and E-A-T. These industry buzzwords have been a hot topic over the last year and have been the focus of numerous articles and blog posts, as well as the heart of many debates. But, what are they really, and can you optimize for them? Here’s what you need to know.
What Is BERT?
BERT stands for Bidirectional Encoder Representations from Transformers. It’s a machine learning model that was announced by Google in October of 2019 and was designed to help Google improve its understanding of search queries.
With BERT, Google is able to better understand how a word relates to all of the other words in a query rather than just looking at each word individually. This helps the search engine to better understand the intent behind a search, and in particular, helps with more conversational queries where prepositions like “for” and “to” can have a significant impact on the meaning of a search.
One example Google used when announcing the BERT model was a search for “2019 brazil traveler to usa need a visa.” The word “to” and its relationship to the other words in the query help to identify that the search is specific to a Brazilian traveling to the U.S. In the past, Google didn’t understand this relationship between the words and would ignore the word “to” delivering results about U.S. citizens traveling to Brazil.
The BERT model is currently being applied to both web results and featured snippets and at the time of announcement was said to help Google search better understand one in 10 searches in the U.S. in English.
Can You Optimize for BERT?
While there are well over 1,000 articles in Google that suggest you can, BERT isn’t something you can optimize towards. BERT is simply one of the latest advancements in the science of language understanding that helps Google interpret the language nuances in a query and better serve results based on those nuances. At best, your well written content may show up for more relevant queries thanks to Google’s better understanding of those queries.
Speaking of well written content, let’s talk a bit about E-A-T…
What Is E-A-T?
While E-A-T has been getting a lot of attention in the last year or so, the concept of E-A-T has been around since at least 2014. E-A-T stands for Expertise, Authoritativeness, and Trustworthiness and was first introduced as part of Google’s Quality Rating Guidelines. These guidelines are used by Google’s third-party Search Quality Raters who offer feedback to Google on how to improve their overall search experience.
Oftentimes Search Quality Raters will be tasked with Page Quality rating, which evaluates how well a search result achieves its purpose. Once a page is identified as having a purpose, the next ask is to determine the amount of expertise, authoritativeness, and trustworthiness a page has. The page is then rated accordingly.
So why is something that was first introduced in 2014 still making waves today? You can thank a two year old algorithm update and “fake news” for that! First, an algorithm update in August of 2018, which the SEO industry would ultimately name “Medic,” appeared to have had a significant impact on sites offering health and medical advice (Hence the “Medic” name). It has since been speculated that these sites were impacted, in part, due to E-A-T and the lack of it on these sites.
Then, last year, Google published a paper that outlines how it fights disinformation and one of the topics highlighted in the paper was how Google’s algorithm assesses E-A-T. What was unveiled wasn’t anything new per se. The first item listed was PageRank and links, which Google has consistently used in its algorithms to correlate trustworthiness and authoritativeness of a website . Google then went on to outline how its Search Quality Raters use E-A-T.
Despite nothing new being introduced in either of these instances, the SEO industry for some reason latched on to E-A-T and it has become another highly touted buzzword that some claim is part of the algorithm and a ranking factor you can optimize for, but can you really?
Optimizing for E-A-T
First and foremost, Google’s algorithms aren’t necessarily looking for E-A-T. This was clarified by Google’s public search liaison Danny Sullivan back in February on Twitter. He stressed that E-A-T is really a tool for Quality Raters to determine if Google’s algorithms are ranking and displaying quality search results and not something the algorithms are looking for.
Our systems aren’t looking for EAT. Our raters are using that to see if our systems are working well to show good information. There are many different signals that, if we get it right, align with what a good human EAT assessment would be. See also: https://t.co/1fs2oJ9Gtl pic.twitter.com/GBbnYEjJUV
— Danny Sullivan (@dannysullivan) February 19, 2020
Where SEO’s seem to be getting hung up on the idea of E-A-T is that it is a single, specific ranking factor. Suggesting you can optimize toward it, are measured against it and are ranked accordingly, but that simply isn’t the case.
E-A-T is a human measurement, but if you want to understand how that translates algorithmically you don’t have to look too far beyond what Google has been telling webmasters and business owners to do all along. Create useful, relevant, quality content that your peers or other relevant sites on the internet will want to link to or share.
Google’s algorithms have been using various signals to identify and rank this kind of content for years, so if you want to “optimize for E-A-T” look no further than the SEO best practices that have been laid out for you for years.
The obsession and misinformation surrounding BERT and E-A-T are prime examples of how an industry can chase buzzwords and make businesses and marketers lose sight of what’s important. If you want to attract or retain customers you need to have a pulse of what those customers’ needs are. Building content that is useful, relevant and of quality can help satisfy those needs and in doing so, you also satisfy Google’s need to provide relevant, quality information to searchers in need of your content.