Unity to me seems like a team with a fantastic marketing department to contrast with a much weaker engineering department. I've tried to use Unity off and on for years and every time I spend time learning it, I regret it. I'm a full-time React developer, and the difference in engineering culture and best engineering practices between Unity and React is massive. The Unity experience is riddled with bugs, poorly-thought-out workflows, and half-finished or obsolete features. Simple tasks, more often than not, become an exercise in frustration. I've written a blog post about this[1], though it's incomplete (mostly because somewhere after writing the 2000th word I wondered why I was even using Unity at all) so I've never published it anywhere.
I've honestly wondered a lot why Unity is successful at all, and the only answer I can really come to is that their direct competition - Unreal Engine, and a few other studio-made engines like CryEngine - are even worse. (Which leads to another question: why is it so hard to make a good game engine? But that would be another thousand words, so let's not stray too far from the point...) Godot is a glimmer on the horizon, but it looks like it still needs a few years to catch up.
I also have a sneaking suspicion that Unity is successful because it is used by a lot of people as their introduction into programming, and they simply don't know that anything better is possible.
Still, I suspect that most developers who use Unity will eventually reach the same conclusion and move on, and this seems to be confirmed by the google trends graph, which has been sloping steadily downwards over the last 5 years or so[2].
[1]: https://www.notion.so/UnityIsFundamentallyBroken-e3b1474e7af...
[2]: https://trends.google.com/trends/explore?date=all&geo=US&q=%...