It talks about the two approaches to science:

Inductive:
1) Create Hypothesis
2) Apply for Grant
3) Test Hypothesis
4) Alter Data to Fit Hypothesis
5) Publish

Deductive:
1) Create Hypothesis
2) Apply for Grant
3) Test Hypothesis
4) Alter Hypothesis to Fit Data
5) Publish

I think that what you're talking about is having the second approach. I think that's a universal truth (that we should pay more attention to what works than what we think should works), and isn't a matter of “top-down” or “bottom-up”.

I think that philsugar hit the nail on the head with “big picture” and “getting things done” as the top-down and bottom-up approach that you're discussing.

In any business (start-up or otherwise) there is a need to understand both why you need to do something, as well as how to accomplish it. And I think that's needed at every single step of the chain. The head of the company needs to understand why we need to do something, and have an idea of how to accomplish it at least enough to communicate it with people who can break the “how” down into smaller parts, and explain the “why” in the context of their group.

When I hear “top-down” and “bottom-up” I think of management styles more than anything. Most businesses have three elements: production (making the good), sales (selling the good), support (general management, IT, marketing, etc.). I believe that “bottom-up” means that the support groups take queues from the sales and production sides, while “top-down” means the support directs the efforts of the sales and marketing.

I believe that truly successful management is actually bottom-up to top-down to bottom-up. You listen to what your front-line people are telling you, figure out the bigger trends, set strategy based on your understanding, and sent it back down for feedback from the market.

But I think that's an entirely separate discussion altogether…

Mark – Can you provide more examples? thanks