Monthly Archives: May 2010

Bill Gurley comments on group buying

Benchmark is obviously in plenty of hot companies, too, including Twitter and Zipcar [which neither KP or Sequoia has backed]. Yet Groupon is another “hot” company that you’re not involved with. Why?

Right or wrong, in both group buying and in the membership-buying space, we had strong concerns about the barriers to entry. That may have been a wrong perspective. We met with a lot of these guys. But that’s why we stayed out.

What do you think now that Groupon ostensibly has a billion-dollar plus valuation?

I’ve heard incredible things about the company. I wouldn’t have anticipated that they would do this well. Long term, I wonder whether the margin on a deal can be maintained when 10 [copycat] companies are calling on that same small business.

via peHUB » Bill Gurley on Rivalry in the Valley, Facebook’s Future, and What He Makes of Digital Sky Technologies.

No surprises- Another way of saying VC does not scale


If you were to look at the performance of large funds (those greater than or equal to the vintage year median size) for venture funds between 1983 and 2003, just 2% of the large funds returned more than 2x contributed capital. And 92% of the funds returned less than 1.5x capital. But if you were 24-logo-1to look at the performance of the small funds (those less than the vintage year median size) for those same years, the performance is much better. Indeed, 48% of those funds returned 2x — or, put another way, small funds were 24 times more likely to produce returns above 2x than large funds. And just 36% of small funds returned less than 1.5x capital. Wow.

via Redeye VC: Size Matters (at least for venture funds).

Hal Varian: Moore’s Law is also a coordination device

I heard Hal Varian say this at Berkeley a few months before he left it as a comment here:

But here’s another angle on Moore’s Law: it is to some degree a coordination device. A computer system involves CPU, memory, disk, display, etc., and all the components have to improve at the rate to avoid a bottleneck. The system is only as powerful as its weakest link. If the semiconductor industry tries to double the performance every 2 years, then the disk, RAM, screens, etc can design to that benchmark. By this view, Moore’s law could be 2 years, or 3 years, or whatever —- what matters is the common expectation for component parts manufacture.

via The Technium: Was Moore’s Law Inevitable?.