This was a pretty interesting read. I think the important lesson from the demise of Friendster as well as the subsequent demise of MySpace (which some people try not to admit) is that engineering actually matters. Friendster failed because it did not work. It was too slow, sometimes pages would not load at all and it was too easy to spam.
So it got to the point where people had to take forever just to delete the spam from their folders, because it would take a couple of minutes to open each message and discover it is spam.
The article talks a lot about business decisions, but the business side does not really matter that much until you get your web site to work.
That's a great take-away I think. I think that is one of the reasons Facebook was so successful. I've been a member of Facebook since 2004 and I can't remember a single piece of spam I've ever seen through it.
In 2004 when I joined, it was a pretty simple, no-AJAX standard PHP site and worked very quickly. From there, it just took off and they were able to hire some of the best engineers and server architecture people out there.
I've gotten 5-10 pieces of spam through facebook in the last year or two, but it's all been from non-techie friends who had their account phished. So at least I was able to help them clean it up and (hopefully) learn how to avoid it in the future.
When your company has no revenue, the "obvious" answer is to sell as hard as you can until your cashflow positive. But it's not going to help unless the product actually works.
This seems to be the gist of the comments here, but I'd like to propose a third requirement: operations.
Ultimately, a business has to become self-sustaining, and that can't happen until the acts of building the product and selling the product become sufficiently well defined that they can be handed over to a professional staff. (See also: E-Myth)
So it got to the point where people had to take forever just to delete the spam from their folders, because it would take a couple of minutes to open each message and discover it is spam.
The article talks a lot about business decisions, but the business side does not really matter that much until you get your web site to work.