I’ll sometimes filter based on “most popular” and the result will be that the top item is som expensive thing that makes me wonder if it’s really the most popular. It seems obvious that this might just be a way for the store to manipulate buyers. Does anyone have insight into how these work and is there any real function behind them?
You’re asking about “recommendation algorithms” if you want to do some more formal reading on the topic.
The specifics of how they function are often among a particular company’s more closely-guarded trade secrets. TikTok, for instance, is famous for the efficacy of it’s. Google is pretty good at it too, and while I haven’t used his properties in awhile, I suspect Zuck’s is also perfectly competent at doing what he wants it to do.
Reddit’s was always pretty weak, which I actually liked, as the content was more random instead of being more carefully selected.