Reports have been circulating about a possible Google local search update early last month with people seeing higher than usual ranking fluctuations.
Turns out these were well founded. Google SearchLiaison confirmed the update in a series of tweets yesterday, saying they began using neural matching in early November for generating local search results.
What is neural matching?
Google’s Danny Sullivan has previously defined neural matching as an “AI method to better connect words to concepts. Super synonyms, in a way”.
Yesterday he gave the following example of how it works:
It is separate and unrelated to BERT and there is no change to BERT.
Google started using neural matching in organic search last year. It is new to local listings.
Further update information
Danny Sullivan gave more information about the neural matching update, replying to questions on Twitter.
It applies to all the ways Google gathers information about businesses.
It’s not spam-specific but it may help with spam “generally maybe helping businesses not feel pressure they have to shove every keyword into their name”.
Google said the launch was global, covering countries and languages worldwide.
It has now rolled out fully, though smaller updates happen all the time so things do change.
Danny Sullivan said the larger changes should now have settled down.
No specific action needed
Google says businesses don’t need to make any changes, adding that people should continue to follow Google’s guidelines on how to succeed in local rankings.
It seems this is the first time such specific feedback has been given about local search. Danny Sullivan said they would try and do more of this going forward.
