There are definitely things to be said about cost-benefit analysis acting as a Great Filter for interesting art, but the real reason "kill your darlings", as advice, should fucking go is:
quick, what does it actually mean
no, like, specifically
what actionable thing is it telling you to do
It's a neat-sounding phrase, isn't it? Clever. I bet someone was really pleased with themself for it, and as advice it's bad because it sounds like it's telling you to do something, but the second you ask what that is you're on your own. It's beautiful and actively bad at its job
making it, ironically, a darling that should have been killed
This is why I’m a much, much bigger proponent of Julia Drawfee’s “delete your art”, the full version of which is “look sometimes you’ve gotta delete your art. I did it by accident that time but we’re live on camera so I have to make sure it’s not a big deal, but also sometimes you’ve gotta just delete your art. Sometimes you get stuck on a thing and you keep trying to make it work, or sometimes you just don’t like how it’s going and think you could do better, or sometimes you just delete your art. You know what it’s an order now, whatever you’re working on right now, delete your art”.
That’s a paraphrase because the whole thing actually had me wheezing too much to really remember it but also spoke to me in ways that “kill your darlings” never did. And it’s actionable too! Delete your art.