I was the first employee and lead developer for a startup in boom #1. It fell down, went boom. What went wrong: - Timing -- We were either ahead of our marke

I was the first employee and lead developer for a startup in boom #1. It fell down, went boom. What went wrong: - Timing -- We were either ahead of our market, so we had significant head wind making people realize that there was something there. Also, we hit the tail end of the boom, so round #2 of money was at a really bad time. - Focus -- We had three projects, we should have had only one. Plus assorted other revenue generating consulting. - Technical -- We made some unfortunate technical decisions, and because of that wound up duplicating a lot more infrastructure type code than we'd do today. We also wound up with version 2 looking just like version 1, but with a much better data store. The 9 months that that took would have been better spent making the user side better - Personnel -- Single founder, one under-experienced lead developer, and two other devs from completely different software cultures who never meshed well. Any better team members would have realized that this was mildly disguised anarchy and beat someone into shape. I know I probably should have been first in line. - Sales and Marketing -- We were selling to non-profits. The sales cycle was long, sometimes approaching a year. If you have lead times like that, you need to be well funded and in for the long haul, not trying to piece stuff together and sell individual features for dollars. Next time: - Fringe platforms should be looked upon with suspicion. I'm not saying that you should avoid them, but had we even started with LAMP (or linux/apache/postgres/p-language), we might have had a fighting chance. Initial time to market was great, but I'd say that within 3-6 months, it was a drag. - One founder is tough. He was way too split between fundraising, selling, and direction. Also, I mocked some of his code when I found holes in it, so he didn't really trust his technical ability as much after that. - Developer compatibility is important. Mixing a MS alum, a debian developer, and someone who grew up on a fringe commercial platform is a recipe for disaster. - Money -- Make it easy for people to pay you for stuff. Long billing cycles are awful. Long sales cycles are awful. One of my metrics for if a business has its shit together is if they can get invoices out fast and correct. - Know when to quit. -- I hung on too long, because it looked like there was the littlest bit of hope, and I could deal with missing a paycheck here and there. Now, I'd probably have called it off a lot faster and gotten on with the next thing. What now? I did consulting for a while till I found a job with the current company. I'm working for a small company, different set of issues. Technically, we're in a ton better shape. Financially, too. I'd do it again, but I'm not 100% on that, since I've got kids that are really my startup project right now.