Enterprise software has a bad rap because it's such a jaded business; decisions almost never get made for the right reasons, and the level of politics and bs is absurd.

From my own experience: unless you're doing something really special (and even then it's not guaranteed), it's tough to convince a large firm to license your product if you're a startup or 2, 3 man firm.

The best way, actually, is through the proverbial "back door", i.e. make versions available on the web or downloadable, and get a critical mass of employees within the company to use it.

I absolutely agree. I used to work for a big enterprise software company, and it took me a while to get used to the fact that we're selling crappy software for millions of $ a piece. I think the problem is that large enterprises trust other large organizations to solve their problems. You can't really get ahead in a large enterprise and become a manager with a budget to spend on software solutions unless you have that kind of organizational mindset. You fear change and fear failure, and so you'd rather spend tons of cash on a crappy solution from someone "trustworthy" - so they can't blame you if the software you bought sucks, than on a small startup's product that no one else in your industry is using. BTW, Crossing the Chasm is a good book on the subject. I recommend it to anyone considering going into enterprise software.

Yeah, that's pretty much it. Asking really for clarification on Paul's comment about enterprise software in his speech. I sometimes think we've gone a bit too much B2C lately, and we're missing out on a lot of opportunities to Make Something People Want, with the People in this case working at big companies in traditional enterprise functions.