Google Featured Snippets - text source different from image source

[Propools]

While doing some keyword research I ran across a Google results page that had a nice informational snippet. Interesting thing was that the textual information had a link to one site and the image which represented the keyword contained a link to another site. Has anyone else noticed this?

[not2easy]

If you mean Google's "Answer Boxes", then yes. Those are the selected results at the top of a search results page that are separated from the list of results.

Many of us have noticed that they take information from one site and hook it up with an image from some other site. It is one of the facets of this "feature" that is discussed here (2018): [webmasterworld.com...] though the same thing is discussed in several other threads - however, most are not recent.

[goodroi]

Nothing new, very common. Google's been doing since at least 2017.

[Robert Charlton]

Propools... Google may well find an image from a second site more usable than an image from the site it pulled the text. The original discussion on WebmasterWorld was in this thread, and cites an article by Barry in seroundtable....

Mashed Up Featured Snippet w/ Links to 2 Different Sites
April, 2017
https://www.webmasterworld.com/google/4844393.htm [webmasterworld.com]

From the thread ^^^ in relation to your comment...

Barry mentioned in the seroundtable article I'd cited above that this kind of thing might happen... and he predicted that some SEOs wouldn't be happy about it, but he also noted that this is how it has worked for years in "Top Stories or News box content", and everybody's survived. It could even be argued that it's giving two pages links up top rather than just one.

[MrSavage]

more usable than an image from the site it pulled the text

So is that a way to avoid using the word "scraping"? Why can't we call it what it is? This community has talked about scraping for how many years but if Google is doing it, it's suddenly not called scraping? The fact that Google can scrape content from two source to create the ultimate answer ultimately says they are the greatest scraper ever. Adding to the content creators difficulty, if they ever tried to create a page with such a succinct answer, be assured the thin content penalty would apply. It's not like we could create one-page/paragraph answers like Google does using people's content. They have the luxury of creating something that we cannot, unless we want thin content penalties. I think this is called a farce? They say create rich content pages but I suppose if we put multiple answers on one page it makes scraping less resource hungry? Just a theory.

[RareBit]

I always see 1 result that uses my image and someone else's site link - boils my pee

[lucy24]

Never mind disgruntled webmasters. Do human users know how it works? Do they realize that if they click here they’ll be taken to one site, while if they click there they’ll be taken to another--or, worse, to another search using different words? It’s especially confusing if your answer box has multiple images. (Random experimenting led to one with five images pointing to four different sites. Surprisingly, the doublet was not wikipedia.)

[engine]

Propools, people are choosing to make this information available to google. It's worth considering if it raises the profile. If one site doesn't, another will.
Quite often this data is used in answers on audio search.

[RareBit]

Scrap my comment from yesterday - its worked in my favor now! Good job Google:)