Long ago I used to do a lot of upstream tinkering, but quickly came to my own conclusion that in most cases differences in cables (ICs to be specific), input types (on some gear/applications these can be large differences for sure), software applications (when setup properly with the right gear), and others often render very small differences and in others no cases at all. You simply cannot generalize broadly here, acknowledging that fully, as in some cases there are major differences based on the use case and product you're talking about. But I found that you can spend a ton of time evaluating small scale differences, that like you mention, for me became more of a question of placebo vs. not; and/or, is this "difference" even better or do I perceive it to be because it's a change or because I paid for or invested time in it. (edit: tbc here I'm not including power conditioning here, which I do believe can be highly valuable to explore)
For me, the value proposition of spending a lot of time and money to explore the above just pales in comparison to the much greater and easily detectable differences in Headphones > Amps > DACs (generally in this order for me). So I choose to spend my time there. Far less worry about placebo, although bias still is at play to be aware of, and is less stressful to explore
for me.
I think ultimately a lot of people just like to tinker lol. And there is absolutely nothing wrong with that lol. I am not at all one of those people. Tubes are another area where I set and forget it--research tubes that are well liked, do a small level of A/B'ing and forget about it; as for me across the board mostly tubes have subtle differences (major exceptions here based on amp) and each comes with their own tradeoff. Guess you just have to figure out the philosophy that works for you.