I don't think it's that hard at all:
There are numerous instances where what the user wants and what is most profitable for your startup don't perfectly line up.
I have no idea what Stage6 offered, but I find it hard to believe that they wouldn't be able to get (more) funding if they had passionate users and dramatic growth.
Making money is, of course, another can of worms. I could start a business of sending everyone $100 (for free) and I'd see a lot of love and growth. Maybe that's what Stage6 did? If they can't make the argument that scale can get their total cost down enough that $2.50 CPM could support it, then they are "buying love", no?
They got something that users want (I used their product, and I'm not alone), but they didn't cope well with success. And after serious political issues they closed. More info: http://news.ycombinator.com/item?id=124574
The startup I left in September has now been around for 7 years. They're profitable (obviously; otherwise they wouldn't still be around), but don't have product-market fit: they have fewer than a dozen customers and need to go through a laborious and often fruitless sales process for each new customer. IMHO, their problem is that they're too focused on "Don't stop" and not enough on "Build something people want." When one of their product ideas fails in the marketplace, they usually assign an engineer or two to maintain and update it, in the hopes that if they add a few more features or make it a little faster, someone will buy it. As a result, they had something like 7 products when I left, only three of which were bringing in revenue. They might get much farther by concentrating resources on really trying to get a breakthrough with one product than scattering their resources across 7.
It's harder to find examples for companies that made something people want and still failed, but they're out there. AllAdvantage.com comes to mind: they gave away free money, lots and lots of people wanted it, but they still failed because they weren't making more money than they were giving away. Friendster too, sorta: lots of people wanted them, enough that it killed their servers, so nobody could get to them, so everyone went to MySpace. Both of these were at least indirectly because they didn't view the startup as a long haul and work out a sustainable infrastructure.