In the latest Search Off the Record podcast, Gary Illyes from Google spoke about how Google Universal Search works and more importantly, how Google decides which vertical search feature to rank in what position in the main search results.
Gary Illyes explained why sometimes featured snippets show up at the top, or maybe news articles, or images, videos and so forth. In short, each of these vertical indexes bids for its position within the main web results. And how does Google decide to show images versus news versus videos? Illyes said that Google sees what users click on and if, for a specific query, more users click on more images, Google knows users want to see the images carousel instead of a news carousel.
One caveat: This does not mean Google ranks specific and individual documents, images, web results, etc. based on click data. However, which universal element Google may show in the web results, can be influenced by clicks. So, Google does not use click data to rank a specific image or a specific video or a specific web page, but it can use click data to, for example, show the image carousel box here or there on the web results page.
Indexes bid for positions in the search results
In the podcast, Illyes said that Google assigns a score to every result it finds for rankings. Then it will distribute that score to every index. By index, he means not just the web index but also, images, videos, news, etc.
Illyes said then Google “will take all the results that came up from all the different indexes and try to mix them together,” into what we know as Universal Search. Each of these elements or indexes “are bidding for their desired positions” in the search results page. So, featured snippets bid for its position, images carousels bid for its position, videos and so forth.
Some elements can “also say that I don’t want second position, or third position, or fourth position, or so on. I only want the first position,” Gary explained. Google also has “preferred positions for some things like, for example, the video results,” he explained.
Some features, like related results, always tend to show at the bottom. So those features might specify that they want the bottom position.
John Mueller of Google then asked Illyes, “So it’s almost like all of these different indexes, or kind of content have their own search engine and basically, they’re saying, ‘my result is like super relevant, or kind of relevant.’ And then, there’s like a super search engine on top of all of these search engines that mixes them all together?” “Technically, yes,” Illyes responded.
Clicks influence the features that get shown in the search results
Then Illyes explained that Google learns which feature to show for which query based on what searchers click on and want to see.
“How do you recognize if we should show images or videos? Or that? Is it just like video search thing?” Mueller asked Illyes.
“We learn it,” Illyes responded, “So, like when you search for something, something that normally doesn’t have images or videos, and you tap the images tab on the result page. Then, you are essentially teaching Google that there was this random person who wanted images for this particular query. And if there are enough users doing that, then you are essentially teaching Google that, that query might deserve images, or videos or whatever.”
Just an editorial reminder that this is not how Google ranks individual results in the search results but how Google determines what features (i.e., image carousel vs. news carousel) to show in what position in the search results interface.
You can listen to the full podcast over here.