Google rolled out an update to its algorithm last week called BERT, short for Bidirectional Encoder Representations from Transformers. That’s a mouthful, so we’ll keep referring to it as BERT.
What is the Google BERT Update?
In laymen’s terms, BERT is a form of natural language processing for Google to better understand conversational queries. Essentially, it’s a focus on optimising long-tail keywords users type into Google. Not only does BERT allow Google to understand a query but also puts emphasis on the smallest of nuances present within the search phrase. This update is all part of the master plan to understand language and through that, understand its meaning. At the end of the day, a successful Google search is about getting you to the exact answer that you want. Even if you spell a word wrong or combine seemingly random words, Google’s true north star is all about giving you a SERP catered to exactly what you’re after and as such, BERT is all part of this process to decipher semantic search.
Google confirmed in a post that the BERT update would impact around 10% of searches. 10% is a lot in the search world.
Why didn’t tracking tools pick up the BERT effect?
Although BERT rolled out last week, MozCast, RankRanger and other tools barely caught wind of it. Primarily, these trackers keep watch on shorter keywords. These shorter queries do not require much natural language processing. As such, queries that were affected by BERT wouldn’t have been picked up by these trackers.
It’s important to distinguish that BERT will analyse queries and not web pages. In saying that though, on-page SEO becomes important.
Where is BERT rolled out currently?
At the time of publication, BERT hasn’t been rolled out uniformly across all countries and thus the influence is hard to judge. However, currently, BERT is influencing organic search results and featured snippets on the SERP.
- Organic search results: Currently, the BERT update has been rolled out to US organic search results on Google with one in ten results being affected. The update will be gradually activated across other countries however, there is no fixed timeline yet.
- Featured snippets: For these position zero results, the BERT update is already in full swing. BERT has been updated for featured snippet results in all 25 languages it is present in.
How will the Google BERT update affect content marketing and SEO?
This is a tricky one to answer because both Google and the SEO community have not been able to provide a concrete answer. With that said, however, webmasters should make room for changes. An important distinguishing feature to keep in mind is that content shouldn’t be published just for algorithms but for people, as well. Content creators can be less worried about writing for bots and instead, write conversational, informal (but informative) content for people who would be genuinely interested in reading them. In doing so, the content has a better chance of being matched to long-tail queries.
It’s important to remember that the BERT update doesn’t require long-form content. Yes, long-form posts are stellar additions to a website, but the key is to focus on the type of language used. BERT is just the tip of the iceberg for Google in its quest to understand and master the natural language and semantic search intent.
Another pro-tip is to minimise placing importance on keyword density. Yes, keywords are important and by the same token, so are short-tail keywords, however, it’s now not the be all end all of great SEO.
The takeaway for digital marketers
The BERT update isn’t necessarily about creating pages-long blog posts or on-page content in order to take advantage of long-tail keywords. Instead, the priority and focus should be creating quality content within a specified topic. Be the authoritative figure of content in your industry and make sure you answer search queries. Tools like AnswerThePublic or even Google suggestions are a great place to start when trying to understand how users are trying to find you and the answers you provide.
Yes, BERT may have only had a minuscule impact compared to other updates like Google Panda or even RankBrain but as we mentioned before, it’s only the starting point for Google to master the natural language and the conversational tone used. With the rise of voice-assistants like Ok, Google, search queries are slowly giving way to long-tail questions. If you’re already optimising for long-tail keywords, you’re ahead of the curve and for everyone else, it’s time to pick up the pace.
How is Digital Squad changing their practices?
As professional digital marketers, SEO is like second nature to us and we are at the forefront of change. We have already started making the change to be more semantic-friendly. Moving forward, our on-site content of our clients will continue to include both short and long-tail keywords.
At the end of the day, the best kind of content is one that answers a question. It can be long or can be short but if the user finds an answer, it’s a win in our books (and Google’s!). Our strategies will always focus on optimising content for ranking and this includes not just machine algorithms but for people who genuinely search the interwebs for answers.
If you want to know more about the BERT update or how you can implement a successful content plan to ride the wave of success with Google’s continuous algorithm changes, call the team at Digital Squad today. We are a top digital marketing agency Singapore and we can help you understand all the techspeak and deliver world-class SEO to beat your competitors and be ranked on the first page of a SERP.
Don’t believe us? We’ll let our case studies speak for themselves.