Paul Haahr, a lead search ranking engineer at Google, said on Twitter that there haven't been many changes to Google's algorithms around duplicate detection and clustering in the search results. He said "we make improvements to our code over time, including duplicate detection." "But that's been mostly stable for years," he added.
Here is his tweet:
@ajkohn @methode .@ajkohn @methode Obviously, we make improvements to our code over time, including dup detection. But that's been mostly stable for years.
— Paul Haahr (@haahr) March 30, 2017
We've covered many of the unconfirmed changes around clustering and duplicate detection within the Google search results since before 2007 and some confirmed ones including in 2012 and maybe in 2014.
I have not seen people talk much about how Google filters and clusters those results in a while. Which makes sense since Paul from Google said they really haven't changed much there in "years."
Gary Illyes from Google also happen to have posted a DYK explaining that it is rare to see two results from the same domain. But if you do, it is because the quality of the other sites are extremely low. Here is Gary's tweet:
DYK usually when you see more than 2 results from the same site in the SERP, that's because other results score much lower for the query? pic.twitter.com/Bz4DPaR8PC
— Gary Illyes ᕕ( ᐛ )ᕗ (@methode) March 30, 2017
Forum discussion at Twitter.
