As someone who's spent some 17 years doing almost nothing but startups, allow me to add my $0.0296 ($0.02, adjusted for inflation1) Some combination of #1, #4,

As someone who's spent some 17 years doing almost nothing but startups, allow me to add my $0.0296 ($0.02, adjusted for inflation1) Some combination of #1, #4, and #6. #6 is the one where you don't truly respect the process of management works (in much the same way we all get frustrated at managers who don't understand how the process of software engineering works), and are oblivious to the challenges and concerns involved. This is an easy trap to fall into and most engineers I know have fallen into it at some point in their careers. Good management is often virtually invisible, but bad management sticks out like a sore thumb -- or so the conventional wisdom goes. Remember: 1. Management is a skill, just like coding. It's deep and complex and multi-faceted and you're only seeing the surface of it. 2. In startups, many entrepreneurs are getting their first taste of a completely different world. It's a learning process even for an experienced manager/executive. Imagine being a career web developer and say, cutting over and doing 3D game development on a console system in C++ for the first time. Only it's not a hobbyist thing, it's your job and people are counting on you. 2. Startups are not like normal environments. Inaction is usually worse than a bad action, because at least with a bad action you've learned something and narrowed the search space a bit. You should EXPECT that in a startup, seeing management trip up and make "mistakes" will be a common, everyday occurrence. Oftentimes the right decision will look, externally, like the wrong decision. If you want a well-oiled machine in a well-understood problem domain, go work at Cisco. Startups are about discovery. Product-market fit, paths of least resistance, etc. I'm personally inclined to this form of myopia myself, so as a rule, I only work within my personal network at this point because I have a group of entrepreneurs, executive-level programmers, and so forth who I trust implicitly. I may not understand a given decision, and any given decision may be a mistake but I'm entirely comfortable leaving their domain in their hands while focusing on doing my best in my domain. I find it's less stressful, and more rewarding whether the venture succeeds or fails. (1 - Yeah, I looked it up on WolframAlpha. I'm bored, what I can I say.)