As long as I've been doing this, 100% of the time I had problems implementing, it was always for the same reason: not enough software. Whether analysis wasn't sufficient, features were missing, testing wasn't complete, whatever, we were always deploying before we were ready. I understand that there's a school of thought that says to push like this, and naturally, I know how important the non-coding items are, but still...
I never had project difficulty because an invoice wasn't mailed, or a meeting was missed, or a prospect wasn't engaged; it was always, always, always because we simply needed more time/people/resources to have more software.
Imagine swimming the English channel and realizing you had other things to do. Would you stop swimming? Of course not. That's how I feel about coding.
For a software endeavor, coding is the critical path. Do other stuff, but don't stop coding.
Ummm...Because most of us can't code while working on something else. I think each of us is just taking a different view of the words "stop coding". I think it means "for a few hours so you can get all of the other stuff done", while you seem to think it means "and never fire up vim/emacs again in your life". Obviously, I'm not telling anyone to stop coding forever (or even for more than a few days). If you're a tech entrepreneur you have to build things and keep building new things, or you won't be able to retain customers in the face of competitors that are always building new things.
(And I'm not among those who voted you down. I merely wanted to chime in with a bit of what I felt was important about the subject...not really in disagreement with you.)