Ah, a question I can answer. This used to be a huge problem for me, but it's become much less of an issue. It's a bit weird, but I think what helped the most

Ah, a question I can answer. This used to be a huge problem for me, but it's become much less of an issue. It's a bit weird, but I think what helped the most was that I really got into the concept of "thin-slicing" as my default approach at work: "what is the least amount of work I can do that gets me closer to where I want to go?". I'm really ruthless with it now, like - do I need to design my whole db schema right away? hell no, I need one table - do I really even need the db layer to get to prove out what I'm trying to see? actually no, I don't - do I need a UI? yes, but I only really need these elements, I don't need auth, I don't need X, I don't need Y, I only need Z. It's really easy to fall into trying to optimize multiple things at the same time and thinking about second and third-order consequences of choices that you make. My new ethos, as bad as it sounds, is "focus on making something shitty. when i've got something shitty, then i'll make it less shitty" and repeat as needed - the compounding effect of "less shitty" is "eventually awesome". That doesn't always fly at work, but it's IMO the best way to get things done. I'd say 9 times out of 10, you learn so much more from making the shit thing than you do prematurely optimizing something in your head.