On October 21, 2019, Google started to roll out the BERT algorithm in its search network. BERT is designed to better understand voice searches by utilizing machine learning and natural language processing (NLP). About a year ago, we published an article about voice search optimization and the future of voice search where we explained what voice search is, how widespread it is used, how to optimize your website for voice search, and where it’s headed in the future. Let’s investigate how natural language processing actually works and then apply it to SEO for a website.
A Great Analogy
Google uses neural matching and natural language processing to help understand conversational English. More specifically, when people talk, what are they talking about. I recently read a great analogy for how Google uses Natural Language Processing, or NLP, for language searches:
“…if someone is experiencing the soap opera effect on their TV. If you’ve ever seen a soap opera, you’ve noticed that they look kind of weird. Someone might be experiencing that, and not knowing what that’s called they can’t Google soap opera effect because they don’t know about it.
They might search something like, “Why does my TV look funny?” Neural matching helps Google understand that when somebody is searching “Why does my TV look funny?” one possible answer might be the soap opera effect. So, they can serve up that result, and people are happy.”
Salience: Core Component of NLP
So, how does NLP relate to websites? In order to understand how, we must introduce the term “salience.” Salience attempts to understand a voice search query, website content, and whether a web page’s content is a good example of the topic of the voice search query.
In our previous analogy the voice search is, “Why does my TV look funny?” The web page is about the soap opera effect. The content on the page might mention that it makes the image on the screen look “funny” or any other related synonym. This is how natural language processes help determine what results to provide in its search results.
NLP and SEO
In our article published last year, we discussed keeping your Google My Business listing updated, utilizing long-tail keywords, and having a mobile-friendly website. There are other important content features to consider when optimizing for voice search. First, understand the audiences needs and wants and provide it. A website should be a valuable audience resource that provides relevant information and answers questions. This will work well with the function of voice search and the BERT algorithm. Second, utilize Schema Markup. Available since 2011, only about 20% of websites use schema markup. It can help Google understand the contextual relevance of a website using microdata by creating a more enhanced description that can appear in search results. These are also known as “rich snippets” and can be used for business name, business type, address, hours, telephone, etc. Search engines are prone to process this type of data faster with the use of schema markup.
It is estimated that in the next couple of years, 75% of households will have some kind of smart speaker and 50% of online searches will be voice searches. For more information, feel free to contact us at 713-309-6380 or send us an email to grow@brandranchmedia.com.
Sources:
https://moz.com/blog/better-content-through-natural-language-processing
https://www.searchenginepeople.com/blog/10-natural-language-seo-tips-help-gain-edge-examples.html