In wait of (and working towards) that glorious day when I will gladly hand the recording chores to someone more akin to them (I've heard there are even nutjobs who like this stuff), I'm obliged to split my personality in different "hats" or personas in search of expressing my musical creativity into the objective world.
I've found making such division clearcut and well defined, "playing well" the game of roles, brings great benefits. In the end it is all about allowing the musician to be on his own with his music, without detours or distractions.
The first persona that appeared out of my workflow was the producer. But short after, I've found the need of taking part of the producer part out and creating the role of the engineer, too.
The engineer is in service of both the producer and musician, and its role is providing them with the shortest, most Sesame Street instructions to speed up the tech stuff and allow them to go back to their craft. The engineer sweats that stuff so that the producer and the musician don't have to; so that they can each focus their whole energy on creativity.
The first job for my inner ingeneer has been to create a quick system for the musician to record vocal tracks. I've just noticed that, for a long time, I've been dragged by a certain resistance to record new takes, simply because the process to record was not documented enough. That's where an engineer role becomes handy.
The engineer defined the "process loop" of what goes into recording a vocal track. It will look simple here, with that simplicity that takes time to achieve, because you have had to take away everything non-essential.
Basically, the musician is the next client in the chain. The engineer must guarantee that his client can:
1) Record voice sounds + listening to them at the same time (together with the rest of instruments)
2) Listen to the resulting whole at comfortable volumes
The outcome of 2) will be an OK take, which can be kept, or wrong take, to be discarded. From there, the musician can decide to go back to 1) for another process loop, or decide he has enough stuff.
Once defined those 2 states, the engineer went down one level of granularity and defined for each one the tasks to be executed by the musician. It all went in one single page, like I say with the focus on speed and "technical transparency". Imagining how it works when there are really 2 different persons is handy: if the musician says "could you add more reverb to my track", and the engineer takes say 20 seconds, the process gets interrupted.
This "loop page" has worked very well for the vocals. It took like 2 hours to successfully close the first loop, but it's a job that is done forever. Now I'm working on doing the same with guitar tracks; again, defining the states, then going down on the actions to take. I'll end up having several of these loop pages for guitar, because sometimes I use a distortion pedal, others amp simulation... and other factors...
The big aha thing here, perhaps, is taking this 2-layered perspective on the recording process. If I only had a list that defined the 1) and 2) states, it would be clear but not actionable; I would have to think each step time after time.
If I only had the actions list, on the contrary, I would have a brainless script that would liberate my broadband for creativity, but as the purpose that those tasks aim for is not exactly described, that creates a dangerous indetermination factor there.