>There are ALWAYS business reasons that cause technical projects to fail.

I agree with this concept and what's worse is when a project is technically sound but still fails, or even when a complete high-performance accomplishment fails to be deployed.

But I'll bend the terminology of "always" and "business" because it applies even wider than that.

If you've ever worked for a 19th century company that is an actual bureau by definition, and has had literally over a century to develop from those roots into a much more resilient bureaucracy than could be accomplished in less time, you know what I mean.

What if it's not always a business reason for failure but a bureaucratic component that rises above a tolerable level?

i.e. a non-business reason for "businessmen" to fail.

In a pure bureaucracy that exists solely to maintain standards of some kind, the focus can not be on making money, or the standards could be compromised.

Others will fall by the wayside and only the most successful bureaucrats will prevail in their efforts. Handsomely rewarded sometimes through fees and taxes paid by the real money-makers whom the bureau has evolved to serve.

Yes, rewarded for their efforts, none of which are business-like at all and without any internal focus on making money whatsoever.

These organizations can be some of the most stable and long-evolved of all, plus set the most consistent example of political hierarchy that people in all kinds of places can tend to emulate when they don't have any better ideas.

So when bureaucracy creeps into a business where it has not yet made an incursion, it has to do so under the radar because it's the opposite of trying to make money.

People get good at this and move up in the hierarchy, and eventually there's nobody who's even good enough at actually trying to make money any more. It's a full-time effort just building & maintaining the bureaucracy.

You end up with people that "look" like businessmen, act the way they think businessmen should act, golf like businessmen, etc.

But haven't got a clue how to make a dollar from a functional technical success that's a complete no-brainer :\