Izzi Smith and Marcus Tandler in their 302 of a Kind weekly SEO video chat interviewed Microsoft's Fabrice Canel. In that interview Fabrice Canel explained that he and no one really knows the individual weights of the ranking signals used at Bing because they are all handled by machine learning.
We know Bing used a lot of machine learning. The interview sounded similar to the message from when I interview Fabrice as well back in August.
In short, almost all, if not all, the ranking signals are put through machine learning and machines decide on the weights of those ranking signals - not humans.
Glenn Gabe summed up the message of that part of the interview with these tweets:
This is from a great video with @izzionfire & @mediadonis where they spoke with @facan about all things Bing. Some great nuggets of info there. That point about machine learning is incredibly important to understand. They are not setting the weighting. The ML system does. :) https://t.co/DsIAPnD0fW
— Glenn Gabe (@glenngabe) November 9, 2020
So in short, while Bing know what signals they use - they really do not know the exact weight each of those signals are given.
Google does do a similar thing but it is not clear how widely Google uses machine learning. Remember last week when Gary Illyes from Google explained how ML is used for adjusting weights for selecting the canonical URL?
I don't seem to be able to embed the video here, but you can watch it on Facebook.
So if you ask a Bing rep how important an individual factor is, they really might not know.
Forum discussion at Twitter.
