1. Delight is overblown, in my opinion. I think most of the people truly delighted by fancy animation are just other designers.
2. It's more useful to think about state when deciding when to animate. Could the user have trouble perceiving the change in state that just occurred? If so, then use an animation to help them visualize what happened. I believe this is the primary reason to use an animation - all others are vanity.
If (and that’s a big if) animation is used in moderation only when it actually communicates something and isn’t an active impedence (as demonstrated in the linked post), I think it has a significant effect for users. It’s just not the effect that many might expect.
Meaningful, unintrusive animations are one of the myriad puzzle pieces that come together to form a positive impression. They’re a sizeable chunk of that last 20% that separates “good” and “excellent” in users’ minds. They’re not strictly necessary, but between two equally good competitors they’ll help one pull ahead of the other, because users come away with a stronger impression of “solidness”. It’s not unlike how people tend to consider heft and resistance to flexing as markers of higher quality in physical products.
The problem is that since a decade or so ago, UI design as a whole has veered heavily in the direction of vibes, slideshow wow factor, and “branding value” (I felt a pang of nausea just writing that) and away from the volumes of well-researched best practices, and regard for good use of animation has been lost along with it. We’re well overdue for a correction that pushes UI design back in the direction of practical usability and away from Dribbble appeal.
This really is what UI polish of any kind is all about. You feel like you can trust it more, it feels more robust and reliable. Animation and gestures are a part of this, but it’s only the last mile after everything already feels robust.
Before you make it more glitzy you have to make it less glitchy.