Truth.
Truth.
I would suspect that poor management is a contributor to high-profile failures. But only a minor factor for startups in general. Not making something people want, and giving up too early, are far more critical problems for early-stage companies than mismanagement, and there are a lot more early-stage companies than later stage ones.
Most young startups fail because of founder disputes. Working closely with another person is a skill gained with time and experience, and has a lot in common with management.
I think management is important. I think interpersonal skills are important. I don't think that either of them are the most important reason that startups fail. I think that market is.
When I look at the many projects I've worked on, led, or had firsthand experience with, working at a big company, the biggest common denominator was that all the successes were things worth doing. And the failures, barring some glaringly mis-managed ones, were things not worth doing. When you're working on something worth doing, a pressing problem for the company or your users, all the wind is at your back. You naturally find lots of other talented people willing to work with you; they forgive your little screwups and move on; and you don't have trouble getting more resources once you can demonstrate some minor successes.
By contrast, I've known some incredibly talented managers - we're talking folks who have sold companies to Google, who have worked on products with userbases in the billions, who have managed hundreds of people successfully before - that crashed and burned because the project they were working on fundamentally did not need to be done. Good management makes a good project go more smoothly; it cannot transform a bad project into a good project.
That's probably the biggest thing I've taken away from my big-company experience, and the one that (should I choose to start a startup again) alters my thinking the most. With my last startup, I thought "Well, the idea isn't quite all the way there yet, but solid engineering and raw talent can take it across the finish line." This time, I'd be much more likely to insist on verifying it with real users and making sure that they really, really want it before getting other people (beyond a cofounder) involved or taking it to investors.
It takes—looping back to the main point here—education to know the best ways to take care of a dog. Sure, you can just practice by trial and error for a while, or you could use knowledge from people who have done The Right Thing for years and get ahead of many common mistakes you would have made out of "intuition."
That's really the minimum you should do with dogs or people.
Seriously? If you try to control people pain and reward system, most people see through it and hate you for it.
You want them to have a innate feeling of accomplishment. You completed something difficult? And that felt good. You don't need cookies from a master for that.
The number one thing you need to do, is set something up which they can measure themselves against. 90% of the time people will enjoy improving it, without any input.
People don't react to pain/rewards like dogs do. They see through it instantly.
You vastly overestimate the competency of people.