It seems to me that the moral of the story for early-stage founders is this: Don't (over)-optimize for valuation.
Naturally you want to raise at the highest valuation you can, but it's not quite as simple as supply-meets-demand economics. There are different parties that bring different sets of skill and enthusiasm to the table, and all not only have to justify their investments to LPs, but are actually hoping to make a return on the investment.
If I were to raise a seed round again (and we were oversubscribed - not best-of-YC-oversubscribed but oversubscribed), I would optimize purely for investors that would be most helpful, and let the valuation be an afterthought. My guess is 90% of founders who have gone through it would say the same. Doubly so if they had a really bad set of investors.
If you are a YC founder and you want to start a bidding war, can you find some sucker wiling to give you money at a an absurd valuation? Probably. But when Series A/B time comes you have to justify that absurd valuation. The day of reckoning is ~1 year away. That's a hell of a lot of pressure on a short time-frame for a company just barely getting off the ground.
To a certain extent (and I recognize this is a very flawed analogy), your seed round valuation is like getting somebody to bet on what you'll score when test time truly comes. So if you want to say, "I'm going to score X" and you get somebody to believe you, that's great, but now you have to fulfill on that promise or nastiness comes. That could mean a lot of dilution, a down-round, CEO firings... not fun stuff.
That being said, every company is different, and a lot of the time I see people saying, "Look at that valuation, it's 1999!" while knowing absolutely nothing about the company, its founders, the market, the trajectory, the metrics, the revenue, etc. There are relatively few metrics someone outside of the deal could use to determine whether or not we're in a bubble, so we mostly look at how many zeroes are behind a valuation and determine according to that if it's "a bubble." Pretty difficult, but everyone will have their say.
Personally I wouldn't mind things being deflated a little bit. We didn't raise the highest valuation possible, mostly because we wanted to go with people we trusted and get back to work. I'd like to think taking a little bit of the valuation edge off would make hiring less competitive, rent in Silicon Valley cheaper, etc, and I'd be OK with it being less hot than it is now. Obviously I hope a 2000-like scenario never happens again, but I'll be around no matter what happens, so let what may come.
Naturally you want to raise at the highest valuation you can, but it's not quite as simple as supply-meets-demand economics. There are different parties that bring different sets of skill and enthusiasm to the table, and all not only have to justify their investments to LPs, but are actually hoping to make a return on the investment.
If I were to raise a seed round again (and we were oversubscribed - not best-of-YC-oversubscribed but oversubscribed), I would optimize purely for investors that would be most helpful, and let the valuation be an afterthought. My guess is 90% of founders who have gone through it would say the same. Doubly so if they had a really bad set of investors.
If you are a YC founder and you want to start a bidding war, can you find some sucker wiling to give you money at a an absurd valuation? Probably. But when Series A/B time comes you have to justify that absurd valuation. The day of reckoning is ~1 year away. That's a hell of a lot of pressure on a short time-frame for a company just barely getting off the ground.
To a certain extent (and I recognize this is a very flawed analogy), your seed round valuation is like getting somebody to bet on what you'll score when test time truly comes. So if you want to say, "I'm going to score X" and you get somebody to believe you, that's great, but now you have to fulfill on that promise or nastiness comes. That could mean a lot of dilution, a down-round, CEO firings... not fun stuff.
That being said, every company is different, and a lot of the time I see people saying, "Look at that valuation, it's 1999!" while knowing absolutely nothing about the company, its founders, the market, the trajectory, the metrics, the revenue, etc. There are relatively few metrics someone outside of the deal could use to determine whether or not we're in a bubble, so we mostly look at how many zeroes are behind a valuation and determine according to that if it's "a bubble." Pretty difficult, but everyone will have their say.
Personally I wouldn't mind things being deflated a little bit. We didn't raise the highest valuation possible, mostly because we wanted to go with people we trusted and get back to work. I'd like to think taking a little bit of the valuation edge off would make hiring less competitive, rent in Silicon Valley cheaper, etc, and I'd be OK with it being less hot than it is now. Obviously I hope a 2000-like scenario never happens again, but I'll be around no matter what happens, so let what may come.