But Seriously
Thanks everyone who told me how much you enjoyed the Whiner Jerkins PowerPoint. It was a lot of fun to make. I think everyone needed a laugh for a moment.
Some people seemed to think I was trying to deliberately insult Sequoia or VCs in general. Not so. I think Sequoia’s advice is pretty logical, in fact, it’s largely the stuff you hear at Y Combinator. It’s all stuff my startup is doing anyway, and would be even if I took VC funding.
I have always viewed the "get popular now and worry about making money later" ethic with trepidation. There’s this pervasive belief that any company that gets loads of traffic will be able to become profitable at some point, and I’m just not sure that’s true. Maybe if you make the Alexa Top 100 or something, but that’s so hard to do as to be nearly impossible. There are hundreds or maybe even thousands of startups launched each year, and when you count the big corporations that aren’t going anywhere (Google, Yahoo, MSN, Myspace, Facebook, YouTube, Amazon, eBay, etc.) and the porn sites (with which you cannot compete) you have a lot of dogs fighting for very few scraps. Take a look, you’ll note there are very few recent non-porn startups there. I looked briefly and counted 3
On the other hand, build something like Draftmix or TicketStumbler and you can make a fortune from only 100,000 unique visitors a month which, as far as I can estimate, won’t even put you in the Alexa top 50,000.
It did get me thinking a bit about the typical VC-backed ethic though. If a VC’s advice is normally to be more aggressive in terms of growth in good markets, why is that? Just because money is cheaper doesn’t necessarily mean you should go for a land grab. There are very, very few businesses where that seems necessary. So why would that be their normal M.O.?
The reason is that venture capitalists profit more by investing in higher variance startups (all other things being equal) due to the simple fact that they can only lose the amount they put into your company, but they can make a theoretically unlimited amount. To put that another way, their downside is limited, but their upside is not. It’s the same reason people always warn against shorting stocks or selling calls.
Let’s suppose a VC invests $10 million in your company. Let’s further suppose the VCs mean return to be 3x, so in your case, $30 million. Let’s assume a standard deviation of $25 million as well, to allow for a small but realistic chance of an investment returning 10x ($100m). Here’s the graph of possible VC outcomes.
Now let’s assume we double the standard deviation to $50 million. Here’s the same graph:
(Sorry about the goofy stretching, but the graphing xls file I found didn’t allow me to alter the axis properly. The curve looks proper though.)
Put those together and we get:
Which curve looks better to you? (Of course, I’m not sure results follow a normal distribution, but I can’t guess what else they would look like.)
Founders, on the other hand, have the competing interest of reducing variation. They’d generally be happy to make a $30m mean return with zero variance. In fact, they’d often be happy with a much lower return and a much lower variance. Just something to be aware of when you take money, and it might explain why the typical advice isn’t frugality.
Also, a lot of people pointed out that VCs like Sequoia are now telling you that you’re going to have to take lower valuations going forward. And who is paying those lower valuations? VCs like Sequoia.
That doesn’t make them evil hypocrites, as many have suggested. In fact, they share interests with some of the companies they’ve funded who will be getting funding from elsewhere in the future. It’s just supply and demand. The supply of capital from limited partners is going to drop, and VC firms are going to go through a drought just like founders are in the upcoming year or two, which puts the ones that survive in a better bargaining position. If anything, they’re fairly kind for exposing their hand. That sort of openness is a lot of why they remain at the top.
Angels will go through a drought of their own. Wealthy individuals don’t generally leave many millions sitting in checking accounts. Many of them are diversified, so while the Dow dropping didn’t bankrupt them, in a lot of cases it probably knocked out a sizeable chunk of their net worth.
Also, despite my angel funding slide being meant mainly for humor, very much of the angel investment cash floating around The Valley these days comes from Google. Founders who sold their company to them, employees who got in earlier and have considerable stock vesting over time. A lot of them have lost a lot of money as the stock price has cratered. Hopefully many of them hedged by buying puts back when the share price was north of $600, but for some reason I doubt it.
I feel like if anything, we’re seeing the VC mindset and the founder mindset merge into one due to the rough times forecasted for the near future. That’s probably a good thing. Valuations were a little absurd there for awhile anyway.
As for Seesmic specifically, which some people asked me about, despite what I think of their product, I can’t understand why they had 20ish employees. Not to trivialize what it is they’re building, but Justin TV has built a product in the video market that’s probably more complex and has a lot more traffic, and they’ve done it with a fraction of the employees and what I’m guessing was also probably a fraction of the funding. That’s why I used them as my example.
Loic Le Meur is clearly a one-man PR department, so they’ll probably get through it alright. But it’s clear that, rational or not, things are changing, and a lot of startups won’t, and that’s going to be both good and bad. You’ve got to pull the weeds if you want a beautiful garden.
October 15, 2008 at 10:13 am
The distribution would probably be a Chi^2, which is basically a normal distribution squared. This makes sense because the return can't be negative, if you consider the 10 mil to be a sunk cost.
I'm sure the VC's prefer higher variance, because they can get a better deal on higher variance projects. You said above that startup founders are willing to give up EV for lower variance. So it stands that the greater the variance, the greater EV they are willing to give up, which is great for VCs doing a high volume of deals. The same is true in most industries; farmers trade risk to options traders if they expect future variance, the more variance they expect, the better deal they give on the options.
October 15, 2008 at 10:13 am
The distribution would probably be a Chi^2, which is basically a normal distribution squared. This makes sense because the return can't be negative, if you consider the 10 mil to be a sunk cost.
I'm sure the VC's prefer higher variance, because they can get a better deal on higher variance projects. You said above that startup founders are willing to give up EV for lower variance. So it stands that the greater the variance, the greater EV they are willing to give up, which is great for VCs doing a high volume of deals. The same is true in most industries; farmers trade risk to options traders if they expect future variance, the more variance they expect, the better deal they give on the options.
October 15, 2008 at 10:15 am
Also, its good to see you have time to post regularly again. Good insightful writing.
October 15, 2008 at 1:40 pm
Of course you choose to link to us the day we decide to soft launch concerts
. Anyways, TicketStumbler is down for a bit as we complete this process.
Please read more here:
http://ticketstumbler.com/new-stuff/2008/10/15/...
http://ticketstumbler.com/new-stuff/2008/10/15/...
October 15, 2008 at 1:43 pm
Rocking dude. I would have used you guys to get better seats to the Black Keys last week (or at least look) but didn't feel like dicking around with StubHub and Razorgator myself.
October 15, 2008 at 6:29 pm
Matt
Again, the presentation was hilarious – but you knew that already. You like Seesmic that much huh?
October 15, 2008 at 8:41 pm
nice analysis of VC-entrepreneur [mis-]alignment. particularly like the observation that 50K high-value customers may be worth more than 1M low value ones. most folks tend to pay too much attention to uniques, and much less to that minor point (not to mention retention / conversion rates).
on the other hand, an operative assumption here is that VC picks are mostly random and can't be improved… i'd say with a little bit more rigor & metrics, they possibly could.
if you can improve your on-base percentage (weighted by bases / runs), it may make more sense to hit singles all day long than to keep swinging for the fences. i know i'm in the minority on this point, but hey i always was a little different (read: weird
October 15, 2008 at 8:44 pm
Well, a home run might pay 10x what a single does. Thus you'd way rather take a 20% shot at a home run than a 100% shot at a single, if you're a VC. If the single is $10m and you're a founder, that's not true.
October 15, 2008 at 9:59 pm
i might quibble about the #'s and the weighted average, but regardless for the FIRST-time entrepreneur, the weighted calculation doesn't make sense… the entrepreneur may take 1-4 years per at-bat, and they want to maximize on-base potential. in addition, GETTING on base the first time up DRAMATICALLY increases your chances next time (and all future times), since even after a minor win you are much more likely to get capital.
October 16, 2008 at 9:47 am
The distribution for the size of a company in a market is scale free. For every time you double the size of the company, you reduce the number of companies by a factor of X. X will be something like 7 but it varies.
So if our market is say ad-sense sites and X is 7 and we determine that there are 2401 sites worth $1 million, then there will be 343 sites worth $2 million, 49 worth $4 million, 7 worth $8 million, and 1 worth $16 million.
Clearly VC's are only interested in working the upper part of the curve and the higher you work on the curve the bigger the payout is for moving up one notch.
Basically as a investor, you need to look for people who can separate them selves from the masses quickly. There is a lot more luck to this than any one likes to admit so your best plan is to spread out your investments, push every one hard and hope you get lucky.
October 19, 2008 at 9:36 am
FYI we're back up and have soft launched concert tickets. Thanks for your patience everyone.
http://ticketstumbler.com/new-stuff/2008/10/18/...
http://ticketstumbler.com/
November 14, 2008 at 9:41 pm
as many have suggested. In fact, they share interests with some of the companies they’ve funded who will be getting funding from elsewhere in the future