That's often a lot harder to do than it sounds. Engineers don't feel this pain as much, but marketing and sales both see explosive increases in complexity at scale. If you've never managed a direct sales team before or worked for a long time on a well-managed direct sales team, you're probably not going to be any good at managing one yourself. It's hard.
The nature of a lot of engineering (bog standard business logic and UI implementation) is such that a good engineer can adapt to shifting needs and focuses. For example, an engineering team can weather a shift from consumer to enterprise much more easily than sales/marketing. But non-engineering positions are also more monetarily and professionally rewarding in the case of success - a textbook example of high risk, high reward.
This is also related to how engineering is often viewed (correctly or not) as a young man's game. The most commonly cited reason for that is how there's always new stuff to learn, but even if there wasn't so much churn in the relevant knowledge, the fact that younger employees can grow and "catch up" professionally much faster than in other business functions plays a role.
You don't necessarily get fired (if you're an engineer) when this happens, but you probably do get socially brutalized. The founders/investors inevitably begin bringing in "experienced managers" -- people whose primary qualifications are a few years in low-level management at a big-name company. They might even be the same age as you, if not younger. Important decisions begin to bypass you in favor of the new management structure. You get less and less authority in roles that you helped define. It's a really crappy experience.
The problem is that you're a known quantity, and the founders/investors see you as a great worker -- in a certain role. They perceive some problem, and since you're part of the system with the problem, you are seen as incapable of fixing the problem. Better to hire Magic Mike from Facebook or Brilliant Bob from Google to come in and make everything better.
I've seen this happen so many times. If you're an engineer, your choices are generally a) swallow your pride, ride out your vest and don't advance in your career, or b) quit, lose the money, and hopefully move into that more senior role somewhere else, so that you can be Magic Mike next time.
Yeah, that definitely happens. If being a founder is out of the question (which, as others have said, is the only situation in which you should value options as nonzero) yet you want to work in startups, then your focus should really be on becoming Magic Mike. Actually getting a chance to "grow with the company" is the exception, not the rule, even if it occasionally happens in engineering.
The idea that just hiring a team lead from Amagoofacesoft will solve all your engineering management problems is deeply embedded in the Valley ("nobody ever got fired for going with IBM"). Never mind that you'll be scraping the bottom of the middle management barrel at those companies, since the really good ones aren't going to leave.
