Also, Google didn't make the kind of money that Google does when it IPO'd. Companies' revenue numbers don't stay static forever. (Or if they do, the company is likely dying.)
Not the current $6b+ yearly profits, sure, but they were still hugely profitable. Google IPOd towards the end of 2004, a year in which they made $400m profit on $3.2b in revenue: http://investor.google.com/fin_data2004.html
(But I agree that you don't need $400m profit to IPO.)
As a former investment banker who has worked on numerous IPOs and follow-on offerings, there is only one thing you need to have a successful IPO:
Institutional buyers who want to buy the stock at a price acceptable to the company.
Back in 1999, they'd buy anything with .com in the name; nowadays, they want to see a pretty clear path to sustainable profitability. That doesn't mean you actually have to be wildly profitable; just that you're of a sufficient size and have implemented a clear business model. Facebook, LinkedIn, Zynga, and countless others could IPO tomorrow if they wanted. Twitter's issue isn't profitability; it's size. $25 million in revenue just isn't big enough to get out just yet.
