In a recent podcast, Google’s John Mueller, Gary Illyes, Martin Splitt, and Duy Nguyen from the Google search quality team, discussed Google’s methods for ranking search results and preventing and dealing with spam content.
READ MORE: Google Confirms New “Spam” Algorithm Update
An interesting piece of information provided by Nguyen was that Google uses machine learning models to deal with “obvious” spam.
He explained that Google uses a “very effective and comprehensive machine-learning model that basically took care of most of the obvious spam.” This machine learning model enables the Google search quality team to focus on “more important work,” such as hacked spam, online scams, and other issues that machine learning models may not pick up on.
Google’s machine learning models are also constantly working on improving their spam prevention methods when it comes to search by analyzing years’ worth of data.
Insights into How Google Ranks Search Results
Mueller, Illyes, Splitt, and Nguyen also discussed how search rankings work, diving into Google’s methodology.
Here is a summary of what was discussed:
Google’s first step is to compile a shortlist of around 1,000 results for any given query. Google generates this list based on how topical and relevant the query and the content on a particular page is.
From this list, Google will apply ranking signals and factors to come up with an even shorter list. According to Illyes, this part is where “the magic” happens.
Google then “assigns a number and we calculate that number using the signals that we collected during indexing plus the other signals. And then essentially, what you see in the results is a reverse order based on those numbers that we assigned,” said Illyes.
Algorithms that are most commonly used are RankBrain and the HTTPS boost, however, Illyes explained that HTTPS doesn’t have the capability to rearrange search results.
So, there you have it. Are you at all surprised by these insights or have you always had a feeling that this is how Google does things?