It took almost two months, but now it is official news: right after testing phase back in the US, Google BERT has officially been released in order to interpret the language of queries in over 70 languages all over the world. A factor that closely concerns us, italian is among them!

Reminding you about Google BERT

Announced back in October, Google BERT is the sci-fi algorithm allowing Google’s machines to comprehend and interpret in the best way the language used by people in queries, so to make the search process of the most relevant answer among organic results easier.

Pandu Nayak, Google Search VP, chose truly binding words to introduce this tool, by defining Google BERT “the biggest step forward in the past five years and one of the biggest jumps ahead in the history of Research” and anticipating that it would have affected “1 on 10 queries”, allowing the search engine to “understand researches in the best way ever“.

As we were saying on a following topic insight, though, the first effects of Google BERT on SERPs were not that blatant, even because probably its feature can be applied to long tailed queries, in which the interpretation of the single words plus the general context becomes crucial not to provide unfocused answers.

Google BERT available in italian and other 70 languages

Anyway, testing phase in the United States must have been quite fruitful, because on December 9th the @searchliaison twitter account announced the launch of Google BERT in over 70 languages all over the world: alongside Italy, the new algorithm will be used to optimize researches performed in chinese, french, spanish, german, UK english, portuguese, among others. The complete list of languages can be found on twitter and Danny Sullivan has also announced that he will update users on possible new addings.

There is no optimization to perform on sites

There is an important aspect to highlight on the practical side: there is no on page SEO optimization work to do in order to accomodate to Google BERT. There are no particular activities, special techniques or arrangements to experiment, because this neural-net-based algorithm can only be used to better elaborate the natural language allowing Google to answer in a more accurate way to user researches.

Wanting to further specify, we can say that Google BERT will help Google to better understand the search intent of who is performing the query, and then the sites’ goal must be to realize contents perfectly in line with the topic their users are interested in. Quality articles, so to use the classic expression, perhaps even optimized thanks to SEOZoom’s writing tools, starting with the tool indentifying people’s original intent.

Improving comprehension and interpretation of queries for more relevant answers

The use of the newest technology has become inevitable given the ever-growing huge amount of questions people ask Google, more and more complex, that often require a sort of “interpretation” that succeed in going beyond the mere analysis of single words: the examples provided by Nayak during BERT’s announcement clarify what does it really mean “to facilitate the natural research and give back more relevant results”.

Not only effects on search results, but on in-SERP features as well

Google BERT not only helps classic researches, but also all of the features appearing on SERPs: in the US some analysts already verified some practical applications of this technology, on Top Stories Carousels in particular. As written by the Search Engine Journal, Google is currently using BERT models and other machine learning techniques to group correlated news articles inside the various carousels.

Google BERT applied to main news

Duncan Osborn, Google Search product manager, explains that “when you’re searching for information on a timely topic – a recent sports upset or the latest scientific breakthrough – you see a carousel of articles at the top of your Search results highlighting relevant news“. Thanks to the new technology, “when there are multiple stories related to your search, we’ll also organize the results by story so it’s easier to understand what’s most relevant and you can make a more informed decision on which specific articles to explore”.

In order to realize this experience, Google has “developed a new story-understanding technology to map the people, places and things involved in a news story, and then draw connections between them”. Basically, now the search engine is able to create specific and distinct groups of news articles, using “a variety of machine learning techniques including BERT models to examine the related articles and determine where one story ends and another begins”.

At the moment, these updates are only available for researches performed by mobile devices within the US for the english language, but the implementation of multiple languages and locations has already been anticipated starting from the next few months.