Quick answer here, I'll come back to this if there's interest:
I've heard many stories about value stream maps that were too big to be useful. If it takes days to create and the result is a map with 2000 steps, there's a lot of waste involved in an activity that's meant to reduce waste!
Adding detail and spending time is an act of diminishing returns, so it's important to start small and try to get value from as little effort as possible. You can always do another pass or spend more time where you see value.
Likewise, I think a lot of the traditional iconography and paradigms of Value Stream Mapping, such as factories, trucks, spectacles, boxes etc definitely get in the way of the 20% of VSM that drives 80% of the value, the steps and timing. Focusing on those two core components makes the process faster and simpler, making it easier, more useful, and more efficient. I want teams to map every quarter, so it should be easy and fast.
The other wasteful aspect is striving for too much accuracy. If you have 2 steps to measure, and one is an hour and the other is a week, it doesn't matter if the first is really 3 hours and 22 minutes, because it's clearly not the bottleneck. That's not a rule, but it's an easy guideline to help focus efforts where the problems are. Detail isn't the issue, visibility is the issue. If we're in a situation with zero information or information in separate brains, then detail isn't the solution, just getting some actionable information that people can share should be the goal.
We can always dig deeper!