For me, aside from picking initially between kbin and Lemmy and then picking an instance (and the whole concept of instances), it was not having an algorithmically created feed. It took a bit to wrap my mind around since all of the social media apps and sites I was used to (and still use) provides this.

This is kind of an interesting one to me, not because I disagree or anything, but because at least personally, when I've tried to use corporate social media, I felt like I also had to do a lot of manual feed building/curation to get it to be worth anything. However, I do think where some of the algorithmic stuff helped a little was in the suggestions of similar or related pages/users, albeit somewhat rarely.

More than the algorithms it was simply the fact that it was a single platform where you knew they might be & so could search for them, so maybe it was a mixture of those details for you too?