Innovation is almost insane by definition: most people view any truly innovative idea as stupid, because if it was a good idea, somebody would have already done it. So, the innovator is guaranteed to have more natural initial detractors than followers.

The three ingredients of 'founder's courage' to innovate: (1) comprehensive knowledge — an unreplicable pyramid built from knowing every hire, product decision, customer signal, and weakness in the codebase; (2) moral authority to throw out foundational assumptions because you made them in the first place; (3) total commitment to the long-term that exceeds any equity stake or bonus cycle.

The record labels missed the internet because by the time it arrived, all the original founders had been bought out, retired, or died. The professional CEOs who replaced them couldn't let go of the stranglehold-on-distribution assumptions that defined the business — the founders 'would have been nuts to continue believing an assumption that no longer makes sense.'

When Jobs killed the Mac clones and went radically vertical — adding applications, peripherals, and retail stores — most analysts thought he was insane. He was returning from NeXT, a marginal workstation company Apple bought for less than $500M. He had no benefit of the doubt; what he had was the founder's courage to innovate despite the doubters.

If you care about your bonus this year, you are directly incented not to make investments in new inventions — you incur the expense but reap no profits. Any serious innovation requires heavy investment, and beyond the up-front cash, costs may include lower growth, bad publicity, and internal grumbling as existing features atrophy.

Eric Schmidt succeeded at Google not by replacing the founders but by teaming with them to borrow their knowledge, moral authority, and long-term vision. 'Shared leadership and control are incredibly difficult to achieve. Doing so involves intense communication, deep humility, and some hard compromises. Almost nobody ever pulls it off.'

The two required characteristics to bet on a founder-CEO for the long haul: leadership, and desire — 'not necessarily the desire to be CEO, but the burning, irrepressible desire to build something great and the willingness to do whatever it takes to get there.'

General rule of investing: if you hire a professional CEO into a company that has found a large product cycle, they'll maximize it but likely won't find the next one. If you hire a professional to find the product cycle, 'get the jelly, because your company will soon be toast.'