4 ms·

The AH model isn't venture in the sense of betting on very early stage companies. AH seems to invest huge amounts in much later stage companies. The idea is th

The AH model isn't venture in the sense of betting on very early stage companies. AH seems to invest huge amounts in much later stage companies. The idea is that some of these companies would be public in the old days and all the growth is happening on the private side. (This is based on things @pmarca has said many times in many different forums). One could argue that the relevant comparison would be very volatile tech stocks. Comparing it with the few VC funds (the one VC fund?) that hit the great firms two funds in a row is not really fair. Edit: To put it yet another way, the volatility of the returns should be lower so we should be happier with a lower expected return.

rm_-rf_slash 10y ago
That's a good point, and it's reasonable to invest more in a few later stage (money-making) companies than a lot of little maybes. Seems fine and dandy to sprinkle 5,6,7 figure investments here and there if you have a lot of cash to play with, but those startups don't operate in a vacuum. They will need support from time to time, they will require work-hours from your firm that could be allocated elsewhere. Of course there's no one-size-fits-all, but every startup investment will in some way or another cost more than the check you write.
mathattack 10y ago
Very true. Seed versus growth versus pre-IPO versus PE have very different goals and volatility profiles. It may be more accurate to compare a16z to a moderately leveraged bet on an S&P, NASDAQ or tech index. It also can take a fund or two more to really tell. The strength of places like KPCB is their performance over time. One of the few asset classes where the best persist over long periods.
jasode 10y ago
>The AH model isn't venture in the sense of betting on very early stage companies. AH seems to invest huge amounts in much later stage companies. But that funding at later stages may have been happenstance rather than deliberate strategy. For example, a16z looked at AirBnb during the Series A fundraising. But they decided to pass on it, partly because Marc Andreesen didn't believe AirBnb's trust model for strangers in personal homes would work. When a16z looked at AirBnb again during their Series B fundraising, they invested. Marc has been on record saying they regret missing the earlier Series A round because it would have made them a lot more money. a16z also made a late 2010 investment in Facebook but some observers thought they did it for "logo shopping". In other words, observers said it wasn't a late investment timing on purpose -- it was because they were a new firm (2009) and they wanted to add immediate credibility to their portfolio. It seems they do want to catch companies early if they can.[1] On the other hand, I remember Ben Horowitz saying they deliberately held back on an Oculus Rift and let another VC take the lead (the risk). They later invested in the subsequent round. It doesn't seem like there's any rigid single strategy there with regards to timing. [1] scan for their "Seed" and "Series A" investments: https://www.crunchbase.com/organization/andreessen-horowitz/investments https://www.crunchbase.com/organization/andreessen-horowitz/...
- inputcoffee 10y ago
  You raise an interesting point, but I think the data actually makes my point. Firstly, I think you should count the "majority" of their investments as where the majority of their money goes. If they had one fund with a billion dollars, and $400 million went into Uber and $400 million went into a Facebook, but $200 million went into 90 tiny companies, I would argue that they are 80% (800/1000) late stage and not 2.2% (2/90) late stage. Secondly, what they call the round is kind of arbitrary. Consider that many (most? traditionally?) VC funds earmark around $5 million for all rounds in a company. They may put in $100k pre-product, $1 million after market fit and so on. AH put $25 million in Clinkle's "seed" round. Imagine how valuable Clinkle must have been at that point. Nonetheless, I think this shows that they are making much larger, much later investments than the "traditional" VC, and for reasons that Marc has talked about in public. Of course he wishes he had got into successful companies even earlier. Everyone wishes that. I wish I had got into AirBnB in the Series A. So do you, presumably. He fully concedes he is not as good as Peter Thiel (a, who is?, and b, this is his example) at going for the really early stage company and prefers to go in later. Its just really hard to tell which of 1500 companies is going to make it big, but it is much easier to guess that most of 20 big companies are going to get bigger.
  - jasode 10y ago
    >Of course he wishes he had got into successful companies even earlier. Everyone wishes that. Unfortunately, I made it sound like MA's thoughts about Airbnb was a generic "invest earlier means more money" cliche. I intended to show how AirBnb contrasted with Oculus VR. They passed on the early rounds for those 2 companies for different reasons. For AirBnb Series A, Andreesen didn't understand the value. It was a gap in knowledge/imagination about what AirBnb could be. It was not because of a "we're a late-stage not Series A investor so come back when others have already invested in you". For Oculus Rift Series A, a16z got the value of it, but they weren't sure about it gaining traction while the company was trying to solve the motion sickness issues. They let the other VCs take on that risk and they knew ahead of time they'd pay more to get in on the next fundraising round. Based on their actions and interviews, it's possible that AH's primary investment thesis is Series B or later but I'm not sure you can beat Sequoia and Benchmark with wait-&-see late-round investing at sky high valuations. Those other VCs hunt aggressively to get in on Series A. Part of the prestige for a VC firm is to be seen as a leader and not a follower.

rm_-rf_slash 10y ago

That's a good point, and it's reasonable to invest more in a few later stage (money-making) companies than a lot of little maybes. Seems fine and dandy to sprinkle 5,6,7 figure investments here and there if you have a lot of cash to play with, but those startups don't operate in a vacuum. They will need support from time to time, they will require work-hours from your firm that could be allocated elsewhere. Of course there's no one-size-fits-all, but every startup investment will in some way or another cost more than the check you write.

mathattack 10y ago

Very true. Seed versus growth versus pre-IPO versus PE have very different goals and volatility profiles. It may be more accurate to compare a16z to a moderately leveraged bet on an S&P, NASDAQ or tech index. It also can take a fund or two more to really tell. The strength of places like KPCB is their performance over time. One of the few asset classes where the best persist over long periods.

jasode 10y ago

>The AH model isn't venture in the sense of betting on very early stage companies. AH seems to invest huge amounts in much later stage companies. But that funding at later stages may have been happenstance rather than deliberate strategy. For example, a16z looked at AirBnb during the Series A fundraising. But they decided to pass on it, partly because Marc Andreesen didn't believe AirBnb's trust model for strangers in personal homes would work. When a16z looked at AirBnb again during their Series B fundraising, they invested. Marc has been on record saying they regret missing the earlier Series A round because it would have made them a lot more money. a16z also made a late 2010 investment in Facebook but some observers thought they did it for "logo shopping". In other words, observers said it wasn't a late investment timing on purpose -- it was because they were a new firm (2009) and they wanted to add immediate credibility to their portfolio. It seems they do want to catch companies early if they can.[1] On the other hand, I remember Ben Horowitz saying they deliberately held back on an Oculus Rift and let another VC take the lead (the risk). They later invested in the subsequent round. It doesn't seem like there's any rigid single strategy there with regards to timing. [1] scan for their "Seed" and "Series A" investments: https://www.crunchbase.com/organization/andreessen-horowitz/investments https://www.crunchbase.com/organization/andreessen-horowitz/...

inputcoffee 10y ago
You raise an interesting point, but I think the data actually makes my point. Firstly, I think you should count the "majority" of their investments as where the majority of their money goes. If they had one fund with a billion dollars, and $400 million went into Uber and $400 million went into a Facebook, but $200 million went into 90 tiny companies, I would argue that they are 80% (800/1000) late stage and not 2.2% (2/90) late stage. Secondly, what they call the round is kind of arbitrary. Consider that many (most? traditionally?) VC funds earmark around $5 million for all rounds in a company. They may put in $100k pre-product, $1 million after market fit and so on. AH put $25 million in Clinkle's "seed" round. Imagine how valuable Clinkle must have been at that point. Nonetheless, I think this shows that they are making much larger, much later investments than the "traditional" VC, and for reasons that Marc has talked about in public. Of course he wishes he had got into successful companies even earlier. Everyone wishes that. I wish I had got into AirBnB in the Series A. So do you, presumably. He fully concedes he is not as good as Peter Thiel (a, who is?, and b, this is his example) at going for the really early stage company and prefers to go in later. Its just really hard to tell which of 1500 companies is going to make it big, but it is much easier to guess that most of 20 big companies are going to get bigger.
- jasode 10y ago
  >Of course he wishes he had got into successful companies even earlier. Everyone wishes that. Unfortunately, I made it sound like MA's thoughts about Airbnb was a generic "invest earlier means more money" cliche. I intended to show how AirBnb contrasted with Oculus VR. They passed on the early rounds for those 2 companies for different reasons. For AirBnb Series A, Andreesen didn't understand the value. It was a gap in knowledge/imagination about what AirBnb could be. It was not because of a "we're a late-stage not Series A investor so come back when others have already invested in you". For Oculus Rift Series A, a16z got the value of it, but they weren't sure about it gaining traction while the company was trying to solve the motion sickness issues. They let the other VCs take on that risk and they knew ahead of time they'd pay more to get in on the next fundraising round. Based on their actions and interviews, it's possible that AH's primary investment thesis is Series B or later but I'm not sure you can beat Sequoia and Benchmark with wait-&-see late-round investing at sky high valuations. Those other VCs hunt aggressively to get in on Series A. Part of the prestige for a VC firm is to be seen as a leader and not a follower.