If you work on multiple, unrelated projects, one of the biggest potential nightmares you'll see as you scale up is work flowing over into other work, blending together, getting mixed up, and otherwise ruining your ability to think without being stressed out.
The answer? Build fences.
Clearly separate out different types of work.
Obviously, separating out completely different categories and unrelated projects makes sense, and is even necessary as volume increases. I used to have one email address that all my mail went through. It was easy for a while, but eventually work that was imminently crucially important got mixed in with casual whenever emails, got mixed in with bills to pay, and all sorts of things. I broke things into three discrete inboxes, and that helps. If you have too much email volume on things of varying importance, multiple email addresses by topic -- kept separate -- could be an answer.
Then, letting deals blend together is no good. I just recently created a "Deals" folder on my desktop, with subfolders with everything I'm working on. It's a godsend. All my files are in the right places. I used to do this much more ad hoc, and this is so much better. Actually, I created folders for every major campaign I have... and so far, I haven't had a single thing that doesn't belong in one of them.
Very helpful. Consider doing it.
(My folders: Organization, Health, Company, Writing, Deals, Charity, Money, People, Books)
Even beyond separating out unrelated stuff -- I'd strongly recommend mentally separating and fencing apart different modes of work. In one major company/project I've got, I have three distinctly different hats -- one of them looks like a lot like an executive, one of them looks a lot like a contractor who has fixed spec tasks, and one looks a lot like an employee who should be punching in and out on the clock, getting performance reviews, and otherwise bringing some consistency to some important but repetitive tasks (sales, invoicing, followup, service, etc).
Often, I was blending these areas together, and inevitably one or another ball would get dropped. We just spent some time realizing that, setting some weekly reviews for management, and looking to define out exact requirements for "employee mode." It's a work in progress, but valuable.
Sometimes it's easy to change mental modes, but even when it's easy I think there's some cost to it -- you can't get into a pure flow state. And sometimes, it's not easy, as the different roles in your life compete and fight with each other for attention.
When that happens, build some intelligent fences. Figure out the different modes of work you do, schedule different times for them, and clearly understand what type of work you're doing at the time. Your sanity will surely thank you.
How much skill is the right amount to gain? When to keep going, and when to quit?
Some thoughts based on a few quite good comments on "The 1 to 10 Scale vs. The 1 to 10,000 Scale" --
Random: "Knowing when to STOP developing a skill is vital if one's goal is to become a generalist. A decathlete can't afford working solely on his javelin throw all year long... he has 9 other sports to get good at! Of course, for the 1-sport performer, obsession is the name of the game."
I think there's roughly four levels where you make big gains in an area, any one of which can be a natural point for stopping the amount of learning you do.
My post before this was a kind of therapy / Buddhism / personal growth kind of deal, but I also spend a lot of time thinking about how to run effective teams and to be a responsible, thoughtful manager of people. It is my work: I am a lead engineer at Bungie, an independent video game developer of about 300 employees (though not for long, we're growing.) There are some unique aspects to making videogames, and I'll use game development terminology here as I refer to, say, texture artists or sound designers or programmers, but when I talk to friends in different creative industries - film, industrial design, other software development - I find these themes are pretty universal.
If you're going to manage people, you're going to have a lot of conversations about employee performance. It's just bound to happen. Sometimes, like during reviews, it might seem excessive. You might wonder if's worth all the time it takes. It is. It's OK that you spend a bunch of time on this. As a manager, that is your job. It's your job to have well-formed opinions about how you evaluate people and how you work with them to help them grow. If you aren't spending time on that, then you may be succeeding as a leader, but probably not as a manager. Apples and oranges.
It is, however, important to spend this time well. During conversations about performance, everything you talk about should boil down to one thing: the value they contribute to the team. What is their value, and how can they become more valuable?
I find a lot of review conversations tend to focus on strengths, weaknesses, and specific work results. These seem like reasonable topics, and there's value there, but I also find this often leads to a review that looks like this: