4 ms·

The problem with VCs is not that they're dumb. I'm a pretty severe VC-istan critic and I would never say that lack of IQ is their problem. Their problem is tha

The problem with VCs is not that they're dumb. I'm a pretty severe VC-istan critic and I would never say that lack of IQ is their problem. Their problem is that they're ill-equipped to make the judgments necessary to solve half the problem. They know business and acquisition structures and legal pitfalls, but in order to really get signal on the judgment of people-- I'm ignoring the judgment of ideas because I don't think anyone has that down-- they need to hire someone like me to vet technical choices, talent strategy, culture, etc. because an iPhone app is not going to give them that kind of data. How do you find that kind of person if you're not that kind of person? I think it's almost impossible. How are they going to tell the people like me from the many who claim they are? Likewise, I'd be incapable of vetting them for whether they're good at their jobs for the exact same reason. I'm not superior to them in any way; we just excel at different things and it's very hard for one side to judge the other. Perhaps the best way to start a venture fund is to have the selection made not by permanent VCs judging "the team" but by top-notch programmers who do the vetting part-time (no more than 10 hours per week). That way, they don't lose their technical touch and can tell based on technical choices whether a company has a future. (How would a VC know that a company running a typical enterprise Java stack is doomed? It's not his job.) By the way, the reason VC-istan sucks is not that VCs are evil or stupid people, because they're not. It's that humans are bad judges of character and VCs are no exception. When you can't judge expertise (and they can't evaluate technical expertise) you miss out on the associated proof-of-work and default to social polish, which means that you're going from a slightly positive correlation to what really matters (character) to a slightly negative one. VC-istan generates some awful startups, but not because VCs are bad people.

rdl 13y ago
VCs already do this. 0) VCs with technical/operating background. I'd put a lot of faith in e.g. pmarca's evaluation of a team. 1) EIRs 2) Outside advisors 3) Portfolio companies (you do some pre-screening, then have your existing portfolio CEOs meet with the founders of non-competitive new ventures. It's win/win, because the portfolio companies may have some useful business relationship, or if the founders can't raise, might be a good hire.
- michaelochurch 13y ago
  They're not doing it successfully. They could be. They pick advisors who know how to do what they already know how to do, to validate their decisions. That leads to some improvement in the process, but they're not picking people who complement their skill sets. They're picking less shiny versions of themselves.
  - rdl 13y ago
    The funds where I know EIRs and VPs with tech experience don't really fall into that; there are absolutely funds which are horrible and have horrible portfolio execs, eirs, and advisors, but I think if you talk to the really top ones, you'd be surprised by the quality of talent. The weakness is that a lot of funds can't/don't hyper-specialize in specific technology, but for instance there's a VC who has made a lot of security investments who has l0pht people on the team.
    - michaelochurch 13y ago
      Are you suggesting that someone like me should become EIR? I've seen the quality of people who get EIR roles. They're not slouches, but I could do better than most of them, and that includes the ones at top firms. I can actually detect things like what kind of programmers your tech stack will attract; non-technical executives are better with general-purpose marketing, but the 10X factor with regard to programming talent is black-magic to them. Even rarer than 10X programmers are people with enough experience and insight to know what brings "10X" into being and what smothers it. It's not just about the people; conditions and configuration play a gigantic role. I've avoided exploring the VC route (as in, becoming one) because I want to keep my technical skills intact. Also, I'd be +5 sigma great at one part of it (evaluating technical choices and judging talent) but I have no idea if I'd be any good at the other (also critical) aspects of the business. I also feel like if I committed full-time to nontechnical stuff, I'd lose that +5 sigma edge. Right now, I know that if you're using Clojure and your competition is setting up the Java/Maven/IDE environment, you'll almost certainly win. I might not be able to pick that sort of thing out, in 20 years, if I step away from using technology on a day-to-day basis. Right now, I could grep a codebase and tell you better than almost anyone out there if a company has the technical mettle it needs. ("Visitor" or "AbstractFactory" = bad, "mapcat" or "flatMap" = good.) I wouldn't have that if I sat out of the technical game for 20 years, because the signs and terms would be different.

rdl 13y ago

VCs already do this. 0) VCs with technical/operating background. I'd put a lot of faith in e.g. pmarca's evaluation of a team. 1) EIRs 2) Outside advisors 3) Portfolio companies (you do some pre-screening, then have your existing portfolio CEOs meet with the founders of non-competitive new ventures. It's win/win, because the portfolio companies may have some useful business relationship, or if the founders can't raise, might be a good hire.

michaelochurch 13y ago
They're not doing it successfully. They could be. They pick advisors who know how to do what they already know how to do, to validate their decisions. That leads to some improvement in the process, but they're not picking people who complement their skill sets. They're picking less shiny versions of themselves.
- rdl 13y ago
  The funds where I know EIRs and VPs with tech experience don't really fall into that; there are absolutely funds which are horrible and have horrible portfolio execs, eirs, and advisors, but I think if you talk to the really top ones, you'd be surprised by the quality of talent. The weakness is that a lot of funds can't/don't hyper-specialize in specific technology, but for instance there's a VC who has made a lot of security investments who has l0pht people on the team.
  - michaelochurch 13y ago
    Are you suggesting that someone like me should become EIR? I've seen the quality of people who get EIR roles. They're not slouches, but I could do better than most of them, and that includes the ones at top firms. I can actually detect things like what kind of programmers your tech stack will attract; non-technical executives are better with general-purpose marketing, but the 10X factor with regard to programming talent is black-magic to them. Even rarer than 10X programmers are people with enough experience and insight to know what brings "10X" into being and what smothers it. It's not just about the people; conditions and configuration play a gigantic role. I've avoided exploring the VC route (as in, becoming one) because I want to keep my technical skills intact. Also, I'd be +5 sigma great at one part of it (evaluating technical choices and judging talent) but I have no idea if I'd be any good at the other (also critical) aspects of the business. I also feel like if I committed full-time to nontechnical stuff, I'd lose that +5 sigma edge. Right now, I know that if you're using Clojure and your competition is setting up the Java/Maven/IDE environment, you'll almost certainly win. I might not be able to pick that sort of thing out, in 20 years, if I step away from using technology on a day-to-day basis. Right now, I could grep a codebase and tell you better than almost anyone out there if a company has the technical mettle it needs. ("Visitor" or "AbstractFactory" = bad, "mapcat" or "flatMap" = good.) I wouldn't have that if I sat out of the technical game for 20 years, because the signs and terms would be different.