Michael Arrington of TechCrunch has a really good article regarding the management and building of web applications. He makes some really good points about the common business sense that most developers overlook when trying to start a profitable web company.
First of all, as not to spawn confusion, I completely agree with the comments in the article, and believe when you start a web company (or any company) you should have a plan to generate revenue. At the same time, its kind of funny to look at the recently successful web startups (Facebook, Google, YouTube, Flickr and even TechCrunch itself), and see that most of them didn't grow out of a business plan at all, but rather grew into a business plan. None of these (Facebook and Google were started in academia, and although YouTube has an obvious model it could employ, it doesn't to this day) seem to have started with any revenue model.
It is interesting to see the obvious rule of finding a way to make profit first being broken by these high-profile startups. With Google, Yahoo and Microsoft on the prawl for more pagehits and advertising revenue, it seems that the recent key to success has been attracting and sustaining an audience more than having an initial revenue model. Perhaps there is a bigger rule sorrounding this: that the passion and effort to make a good application is where things must start. The revenue, although important, can be found if the idea is solid.