There's a grain of truth to what you say, but it's misleading to attach it specifically to Google.
Whenever you're doing something for somebody else, they're going to want to see two things from you: 1.) that you understand their problems and concerns and 2.) that you're doing your due diligence to address them. It doesn't matter whether it's Google, some other big corporation, your own startup, or a consultancy. If it's your startup, your customers will want to make sure that you understand their problems and do your due diligence before they fork over their money. If it's your consultancy, your clients will want the same.
I've found that teams that understand this, grok Marissa's way of thinking about the world, and are willing to put in the legwork to iterate on their ideas and make them as good as possible have quite a bit of latitude to do what they want. Those that just spit some s onto the page and come unprepared tend to be the ones that gripe about how Marissa blocks them from getting anything done. Many of the teams I've seen that had problems with Marissa also had terrible UIs, and I'd really hate to inflict them on the user.
In many ways, this is just another way of saying the same thing you did - if you want to be successful, you need one of the decision-makers to back your play. But that's the case anywhere. I've been fond of saying that entrepreneurship is just trading one boss for many bosses - it's great if you hate your boss, not so great if you hate your bosses. Which is a good reason to pick a company or market whose bosses you fundamentally agree with.
Also, speaking as someone who let his startup die because pretty much everyone with social influence over me expressed doubt - if you give up because of what other people think, it'll severely limit your ability to get anything done. If senior executives don't like your project, the best thing to do is listen to them, figure out if they have a point, and if they don't, go do it anyway. The worst thing that happens is that they fire you, and then you're in exactly the same boat as if you quit to found a startup, except that you've been learning about the problem domain while drawing a paycheck the whole time.