Google says the thousands of updates it has made to Search every year since 2015 have helped it to reduce the number of irrelevant results shown to users by more than 50%.
In a new interview, Danny Sullivan, Google’s Public Liaison for Search, said the company’s never-ending quest to improve the quality and relevance of results has made the content shown much more useful, based on its own internal metrics.
Sullivan revealed that Google conducted a mind-boggling 800,000 experiments and made 5,000+ improvements to Search in 2021 alone.
The latter figure is an increase on the 4,487 changes it made to the year before and highlights how consistently Google rolls out updates, both big and small.
Sullivan said the scale of work and testing is perhaps difficult for people to comprehend, and that involves both quantitative and qualitative feedback.
Another interesting insight is that searchers are increasingly entering queries that are more closely aligned to how people normally write and speak.
“Since 2015, we’ve seen a more than 60% increase in natural language queries in Search,” Sullivan noted.
This suggests that rather than shorter queries that are more stunted in tone and delivery, searchers are entering more natural, longer terms and phrases.
Google reacted to this evolution in searches by launching its own neural network-based language algorithm called ‘BERT’ in 2019, which helps it to better understand queries.
Despite thousands upon thousands of changes already, Google is not resting on its laurels.
Sullivan revealed that it will next focus on the advances in artificial intelligence (AI) and how that will impact “what it means to search.”
