Rocketspace Q&A with Vinod Khosla: Insights for Entrepreneurs

It's exciting to see so many people here. And I'm all fired up with my caffeine.

Now, I mentioned in my intro a few of the companies you invest in that have both succeeded and a few that have failed. I'm curious, before we get started into specific advice for investors, what was that aha moment for you where it shifted from failure just being just that, failure, to becoming a viable investment strategy?

You know, as many of you are entrepreneurs, you know that to be an entrepreneur, you have to be an optimist. So there wasn't an aha moment for me. It was almost just ignoring the failure part whenever that happened and just looking on the upside of things and sort of believing we can do, do what we need to do to succeed. One of my favorite quotes on entrepreneurship that I put on a slideshow I gave in 1986 was taken from hang gliding, which I used to do back then. It said, dedicated— it was a movie that was dedicated to those who dream the dreams and then are foolish enough to try and make them come true. And that was so true, because if you're totally sensible, you only do sensible things and you'll never be an entrepreneur. So being a little bit unreasonable is actually a key characteristic unreasonable, optimistic, maybe irrational. All that makes a pretty good entrepreneur.

What is it— why does it take company— why does it take failure to get companies to truly start solving problems?

I was explaining earlier that in the last 10 years I get so many requests to speak I almost I turn most of them down. About the only time in the last 10 years I've actually called somebody and said I'd like to speak was at the little conference here called FailCon, because failure is so key to success. My favorite quote, Harvard Business School wrote a case on our biofuel strategy a few years ago.. And at the top of the case, they put a quote from me saying, "My willingness to fail gives me the ability to succeed." Because unless you're willing to fail, you'll try safe things. And if you try safe things, you won't do anything interesting. It should be obvious that trying to not fail is a self-fulfilling prophecy to mediocrity. You can't do exceptional things by trying safe things, because safe and rational things are what everybody else does. That's why entrepreneurs try unreasonable things, and because of that, sometimes they fail. But they don't really fail. They just try again. And, you know, who remembers the times I failed? Everybody remembers all the successes. About the same time we started Sun Microsystems, we started another company. Same with— Scott McNealy and I started another company about 3 months before we started Sun called The Data Dump. We actually ran the 2 companies in parallel, but who remembers the failure? Anybody heard of the failure? It sort of becomes irrelevant if you take the right attitude to it. So I'm huge on trying things and not being afraid to fail. It's important in that to correct yourself quickly. When you fail, you can fail often as long as the consequences of each failure are small and you course correct. I sort of think of failure as just a way to iterate to the right answer. If you're trying new things, it's very, very unlikely that you're going to be able to tell where you really are going. It's nice to create the illusion that you have a plan and you know where you're going, but it is no more than an illusion because most of the time, most of what you do on your way to success is discover what works and what doesn't, and you have a series of hopefully small failures and don't stick with the bad ideas too long. If you're trying exceptional things, you also have to be very flexible And there's lots of names for it, pivoting is one, but to me it's about a series of small adjustments so you can adjust as you discover, 'cause anything that can be planned for completely is probably something other people have done already.

No one likes to think about that, about failing, and I know for me I have a little startup on the side and I'm always afraid, how will I know at what moment I should say, you know what, this isn't going to work. Because you keep wanting to try.

Vinod Khosla: You know, there's different kinds of this isn't going to work. Most of the time when people give up, it's because they decided on a path and stuck with it, as opposed to— so they failed in how and never got to test whether what they were trying to do was going to be successful or not. Just because you failed at how you're doing something doesn't mean that what you're trying to do is likely to fail. It may be you need different tactics to get to your goal. Your vision is still right. I mean, lots of people tried Facebook before Facebook. Where they failed was in the tactics or how they tried to do X or Y or Z. Not in the end goal of a social network. And so you have to keep in mind, if you believe in a vision— if you don't have a vision, you're unlikely to succeed. If you believe in a vision, you want to try all the possible ways and iterate and experiment as often as you can in all the possible ways to get to your vision. Sometimes you have to change your vision, sometimes you have to pivot, But more failure happens because people fail to try alternative ways to get to a— the vision was right and you discover 2 years later somebody else was successful with it because you didn't try different ways to get there.

Do you still think that anyone locked in a cage, if their life depended on it, they could solve any problem?

I believe people can solve most problems. It is absolutely— necessity is absolutely the mother of invention. And the more desperate you get, the more creative you get. So I've seldom seen a startup that hasn't run into critical, life-threatening problems that seemed insolvable. But when your life depends on solving that problem, you get very creative and you try lots of things. Most people succeed.

@GGibson out here has a question. If failure is so acceptable, why do VCs so often judge a company by their first hypothesis when deciding to invest?

That's a great question. You know, just as much as VCs judge your company on whether the first hypothesis is right or wrong, you should be judging VCs. And what do they do right and wrong? My personal view is people are much more important. Markets are much more important. Those are relatively invariant when you're trying to get— solve a problem or build a company. What the market hypothesis is, how you first plan to try something, those things can be changed. What at least we look for, and I know the kinds of VCs he's talking about, what we look for, I almost tell our guys, it doesn't really matter what the plan says. It's, are these the right people? Will they adjust, be open-minded, try different things, or are they too wedded to one thing? And the second part you can't change often, though it can be changed by pivots, is is the market. If the market's right and the people are right, almost certainly the plan doesn't matter. So I often talk about— and I have a paper on innovation on our website— that sometimes it's a disadvantage to have a plan, and I generally prefer not having plans, at least not rigid plans. Plans should be adjusted every few weeks or months. So what's a plan good for? So why do you ask for a plan? To me, what I get out of a plan is not what the plan is, but the quality of thinking of the entrepreneur. So I use the plan to judge the entrepreneur. Don't count on it as the way they're going to get to their goal. Now different VCs do different things. One thing we never do, we've made maybe 100 investments, I've never looked at an IRR calculation. And people look at me funny when you say you're an investor and you never do rate of return calculations, mostly because we don't believe you can come up with the numbers in your plan. Having been an entrepreneur, I can tell you I can cook the numbers also every single time to get to the answer you need. And Boree, bury the discrepancy in some assumption and then find some rational justification for that assumption. But plans are very important to do a couple of things. One, to look at the quality of thinking of the entrepreneur, to look at whether the entrepreneur has seriously looked at not only all the opportunities in the space but all the risks in the space. And if you looked at the risks objectively, then you think it's an entrepreneur who's thoughtful about what they're trying to do. And if you have a thoughtful entrepreneur, the rest will take care of themselves. There are entrepreneurs who just believe they'll be successful and don't want to question themselves, and those are generally the wrong kinds of entrepreneurs. We spend so much time working— so one thing I've never done is call ourselves or myself a venture capitalist. Either when I was at Kleiner Perkins or now, I just say my role is venture assistance, and if you go to our website, that's all it talks about. And our goal is to help entrepreneurs, challenge them, ask them hard questions so they can then discover. I don't think traditional approaches to a company work in the venture business. People have these views of boards of directors. I hate board of directors meetings. I seldom go to any. If we can, we'll have startups that don't have a board meeting for a year or two. And the reason is simple. You can't just report. What you really have to do is say, what are the 1 or 2 or 3 questions that are most important now till the next phase? And brainstorm around them. That's a very different model. So I can tell you, in 25 years on the— of being on various boards of directors, I have never once, not once, with one exception I'll tell you, voted against an entrepreneur or management team on any issue, and I discourage boards, I fight boards who want to vote on things. And the reason is very simple. VCs come in once every 6 weeks for a board meeting. They have dealt with 20 other companies, don't know enough about what you're doing while you're dealing with all the nuances that you can't explain in a slide. Your hunch is built up out of trying a thousand different little things. And no VC or board member should ever overrule that. Their goal should be to ask you hard questions, to push your thinking, maybe make you uncomfortable, but to leave the decision to you as the entrepreneur. If you believe in this iterative model of planning that I've talked about, then you want that kind of approach as opposed to a board that says, "Here's your milestones, here's the IRR we calculated, now go deliver and fix." So I sort of wandered off on that question, but at least I think that's a very, very important cultural difference between different types of VCs, and different VCs do very, very different things.

A lot of times what I find in terms of interviewing startups is that I feel as though none of the actual entrepreneurs feel like they have the power as they're going around and searching for funding. What power do these entrepreneurs have?

You know, that's a shame because entrepreneurs should feel like they have all the power. One of the whole hardest things, and I'm going to answer a different question first, and if I remember, I'll come back to your question. As an entrepreneur, one of the hardest things you do is to know what you know, to know what you don't know, and to know what you don't know. Most people fail because of the stuff they don't know they don't know. And beyond that, for the stuff they figured out they don't know, the hardest thing you do is decide whose advice you ask. Now, if you're a technical founder, which many of you are, you are used to coding, you have a great idea, and there's a marketing question. You have no way of not only judging the answer to that marketing question, you have no way of deciding If somebody is good at answering, who do you ask? I've seen people who ask their parents. They've never been in startup. Or ask the VP of marketing, but that VP of marketing may not be a good VP of marketing, maybe too traditional or only works in something else. So this question of whose opinion on what topic, and almost certainly you have to ask a different person based on the area or the topic. Trust me, it's the single most important decision you'll make is whose opinion to trust on what topic. My advice to entrepreneurs, and the other hard part of this is the same problem I was talking about with boards. Anybody you talk to spends 5 minutes on it, gives you advice. And in forums like this, I get advice often, and I always shiver because I don't want somebody to change their strategy based on a 30-second answer from me on a question where I didn't understand their business. So when you ask me what I should do, I'm very, very cautious because I've had people come back to me 4 years later and said, I did this because this was the answer to your question. It scares the shit out of me because— You know, I always try and say, here's the things you should think about, as opposed to guide them to an answer. But as an entrepreneur, it's important that you develop the skins— skills of judging who to ask for what advice. What I often tell people is ask 2 or 3 people, get their answers, hopefully very different people, and then decide which of those pieces of advice. So choosing between 3 choices, an entrepreneur knowing their business will do relatively well at, better than any expert in marketing or finance or whatever your area of question is. Am I being clear in this? This is sort of a fairly nuanced thing. 'Cause the hunches of the entrepreneur are generally better than the hunches of the experts. But the experts have domain knowledge, and if you try enough of them differently and ask them the same question and you get slightly different answers, you're probably pretty good at picking between those. So that'd be my advice. I know I didn't quite answer that question.

I think you did in a very indirect way. We have a couple questions from Gunshot Digital. Her first question is in regards to searching for VCs. Is money just money, or should you, you know, definitely go for the knowns versus the unknowns?

That's, that's another great question, important question. I love the name Gunshot Digital. I strongly believe that what you get from VCs, good VCs, has nothing to do with money. I fundamentally ask the following test: if somebody gave me 20% of their company for free, didn't take a penny from me, and they didn't make more money, then I failed at my job in helping them build a bigger, better company. I'd sort of say if after 3 or 4 years I haven't helped somebody build a company twice as big or 5 times as big as they would with just a source of money, then I've failed. And that's not in every case, because you can never do that. We make lots of mistakes. But probabilistically speaking, if I haven't extended the thinking, extended the range and boldness of the venture, then I haven't done my job. And so the source of money matters way more than things like valuation. I'll give you an example. I work with Jack Dorsey at Square. He had 5 offers, every major name on Sand Hill Road, every one. He took our offer, which was by far, far, far away, half the price of the highest valuation offer he had. Half. And yet he chose to go with us. And I don't think it was the lowest, but it was definitely half of some of the leading VC firms. And I think he'd say today he didn't make a mistake. He did the right choice because I'm always pushing him on how to be how to think differently about his business, how to think more broadly, how to characterize his risks sooner so he doesn't run into them, those kinds of things. So money, the source of money matters. If you aren't willing to work with a VC in a way where you're willing to give them equity for free, then they're probably just money.

I have another question from Amit.

By the way, it's sort of the single biggest source of frustration for me on my side. The number of times I see entrepreneurs who I feel like they could get so much great help, right? And they don't see it because they think they need this money. When, you know, even hiring 2 or 3 other key people is such a big deal and can double the value of a company 5 years later because you hired the 3 awesome people. I spend so much of my time recruiting for our companies.

And that actually is a great segue.

Okay.

I think you read my mind. Gunshot Digital once again. What's your proven method in hiring talent for a startup?

So there is no one single method for hiring talent. Again, it's a very, very nuanced thing, but let me give you some specific actionable things. I have two papers on our website, and I spend an awful lot of time writing things for entrepreneurs to use, whether they're in our portfolio or not, because otherwise you're not a venture assistant. Let me describe two papers. One is about interviewing people, especially when you've never interviewed. You know, you're hiring a VP of marketing, you're hiring a CFO, you're hiring a CEO, most likely have no idea what you're looking for because you don't know the domain. So the first paper is about interviewing. It's about 15 pages long and it's half a dozen or a dozen tricks I use to evaluate people in areas I don't understand. So if I'm hiring somebody in advertising and I haven't worked in advertising, how do I know what questions to even ask? How to evaluate them? How to do a spec of what I should be looking for? So without going into the details of that, there's a 15-page paper on that. The second thing which everybody, every investor will tell you is hire great people. Now, that's like saying buy low and sell high and you'll do well. Right? It's stupid nonsense that people who don't understand and haven't done recruiting will tell you. I don't know of anybody who's come in to me and said, "Well, we are trying to hire bad people." So everybody has the right intent. The second paper is called Gene Pool Engineering. And I hope the latest version is on our site. My chief of staff is here and I'll make sure he makes sure the latest version is, because we just revised it. And basically talks about how do you engineer the gene pool of a company, the backgrounds of all the people in the company to maximize the opportunities you have and to minimize the impact of the risks in your business. And just doing gene pool engineering will change your view of what your problems are in your company and what your opportunities are, and force you to look more broadly at both opportunities, all the possibilities, and all the risks. Once you've done that, it'll change your thinking permanently, and more importantly, if you have a risk, think about hiring for that particular risk. Then you have to think about which are the companies that have solved that risk before. Then you have to think about who within those companies worked on solving those risks, and you follow this process. You end up with, you know, my goal is for every single risk a company has, of its top 5 risks, they should identify 5 to 10 people who worked on those kinds of risks before and hire 1 or 2, hopefully from very different areas. And that way you've engineered the gene pool both to avoid your risks and to maximize your opportunities. So it's a process, there's no magic answer, but it's a very, very actionable thing on the gene pool engineering side and on the recruiting side. How do you ask a question? What questions do you ask? How do I judge a person? One little trick, I'll often interview somebody, ask them a question, and then interrupt them and ask something else. If they don't come back to the original answer, it tells me something about their attention span. I'm purposely interrupting them or distracting them. There's 100 little tricks like that used to say, What are a person's strengths and what are their weaknesses? Because, you know, just like when you present your plan, you don't say you're doing bad hires. Anybody, candidate coming into you, will show off his best side. And so you have to find the truth behind what they're saying. And the interviewing paper is about specifically interview tricks, especially in the entrepreneurial startup context. Sorry, long answers to short questions. Short questions.

There was a lot of great stuff in there, and I think RocketSpace, you have the link, if you can put it out there right now with the hashtag, his gene pool paper, I know we have the link, that would be great, and then everyone can grab that. We have a question from the other FB. When would you invest in a company who is not a first mover in the space?

When they have awesome people, one. And two, they are not counting on the first mover failing. Here's what's important. Generally, a lot of startups look the same till you get into the details. So if a follower has first awesome people, and two, they're not trying to replicate what somebody else has done but approach the market in a different way with a different value proposition, or a different user acquisition strategy, or a different emotional appeal, something different. Then it's entirely possible to do somebody who's sort of a fast follower. But just chasing somebody else with the exact same strategy doesn't work, unless the leader has really bad people. And that's a dangerous thing to count on. But it does happen. You can out-execute.

Amit Gupta, 777. How quickly after a meeting, after meeting an entrepreneur, can you tell whether they have what it takes to succeed?

Usually, to be honest, after 5 minutes. Oh, wow. You know, being honest about it, you develop a whole set of biases in the first minutes, good biases and bad biases. And then you spend the rest of the time either confirming or denying those biases and if you're doing good due diligence, you know, if I formed a bias and say I don't buy this story, it's much harder for me to overcome my own sort of the priming of my brain. But we do spend the time, and sometimes if I really like the people, I'll say, here's my areas of concern, prove me wrong. And the entrepreneur responds well, that's great. Sometimes I decide I love this because these 3 reasons, and then I'll have somebody in our team say, hey, there's 3 reasons Why do you think these are not right? Let's go prove the other side. Disprove why I like it. So on both sides, both why I like things, why I don't like things, people tend to form opinions pretty quickly, but if they're doing a good job, they will then go test those assumptions, both negative biases they may have formed as well as positive biases they may have formed.

Tech bubble. Given how low cost it is to create a software startup these days, is early-stage venture capital headed for irrelevance?

I don't think so. There's two ways to look at this, right? Here's the great thing about the venture business. When you lose your money, you only lose one times your money. When you make money, you can make 50 times your money or 100 times. Usually, In a bubble environment, the problem isn't money, the problem is skilled people to work on problems. And if they get splintered across too many startups, then you don't have the talent to make these adjustments I kept talking about earlier. Execute well, but keep making the iterations and adjustments in your plan. And if you don't do that, then you're unlikely to succeed. If you don't have other great people in your startup who can challenge you. So you have a worldview. If you don't have a founder, co-founder, or two co-founders, or even some senior employees who you listen to, who keep challenging you, then you're not gonna question yourself and iterate. If people are there to do that, and you consider them peers and respect them in a way where you take their concerns, then I think you're much more likely to succeed. That skill becomes more splintered in a hyped environment because everybody can get— each one of those founders can get funded to do their own thing. I think creating a dynamic of questioning assumptions in your own startup, among your own team, in a way where people feel they can stand up and challenge you and argue argue with you is absolutely essential because without that you can't make the adjustments that you need to find the right path or the right business plan. So, and then there's this issue of if there's some visible failures, you will see the herd VCs drop away very, very quickly. And people will get squeezed. I mean, some good ideas won't get funded. My personal view is the more experiments there are, the more good ideas come to light. If you can do fundamental analysis as opposed to follow other VCs, more startups is a good thing, and it lets us pick from a larger variety of things to fund.

More startups is a good thing, but I think what many startups are struggling with is the high demand for really great engineers and developers, and there's a little bit of a scarcity. So how are you advising your portfolio companies in hiring?

You know, look, we do have an advantage. So just last week, we have a startup called NAS, Kori Race. Vinod Khosla called me and said, there's this fresh graduate out of Harvard. 3 people are trying to get him. He's really, really awesome. Can you spend half an hour with him? And I just said, sure. I sat down with him for half an hour. He signed up. We do have an advantage in recruiting against other— I'll tell you, there's a great—

Anyone want to use him for recruiting?

We had a situation just recently where, you know, UberCab's a great company, doing really well, hot company. We had just agreed to fund Kaggle, which is a small 2-person startup, and they called me one night, it was Thursday night, And I was traveling actually Thursday and Friday. I said, "There's this great guy, but he has this great awesome offer from UberCab and they're so hot, I don't know how to compete. He won't even talk to us." So I just called the guy. By Saturday he had accepted the job, given up the COO job at UberCab to take the COO job at this 2-person startup. But that's our job. That's what I mean by money isn't money. I probably spend more time selling engineers, co-founders, key technologists on joining our companies than any other single thing I do.

I have a question from Abe Martinson. Why in the era of lean startups and scale or die, the VC model itself has remained non-scalable and extremely inefficient?

Well, I don't know what you mean by inefficient. You know, you'd have to sort of describe what you mean by the inefficiency. In the end, the VC business is not a replicable skill business. In some ways, if you ask me, why I might do better than somebody else in the VC business. I'll give you one simple answer. I probably screwed up in more ways than most VCs in the Valley. Over the course of 25, 30 years, first on the entrepreneurial side, going through a couple of companies as an entrepreneur, then on the VC side, and sort of done so many things where I look back a year later and said, "Duh, what was I thinking?" Hopefully, and I always say hopefully I don't repeat the same mistake twice, just find new ones to make, but that repertoire of mistakes is really, really valuable because in it is many of the ways your company can fail and somebody who's done this and has good judgment can help you. This is why the VC business is not replicable. The other problem is that most of the people, as they screw up, get more conservative. And there are a few exceptions. You know, Peter Thiel, you look at a guy like that, he doesn't care about the times he failed. He just looks like I'm talking about on the optimistic side. Most VCs don't want to do something they failed at before. When I would submit sometimes it's the best area to work in again because you know all the mistakes you made. You made them once. You hopefully won't make them the second time. Mm-hmm. So I think the natural tendency, people as they get more experienced, they get more conservative and they don't want to try new things. I got a lot of flak at the FailCon conference when I said people over the age of 45 basically die in thinking about new things. And what I mean by that is the following: the older people get, the less experimental they get, and I think that's counterproductive to successful startups. And that's why It's not that older people can't do startups. I'm 57. I'm happy to say I keep looking for new things to do. And I mentioned Peter. Peter is the same way. But most VCs don't want to change what's worked for them and don't want to do things that have failed for them before. And I think that's a huge mistake. Even if Friendster failed doesn't mean Facebook will fail. I mean, that's an obvious example. And you have to keep an open mind and dig behind the surface of we failed to why did we fail and can we do things different. And the older you get, the less people want to do that. You were asking— the question really was why is this not a scaling business? Because it's hard to find the people to do this right. To make lots of mistakes first and then take advantage of those mistakes as opposed to use them to develop biases that force them into more conservative plans.

In terms of Facebook, you mentioned Facebook, and I had a question that was asked of me before we started, and this person wanted me to ask the question. What do you think the impact of, or the outcome of Facebook's IPO will have on startups?

More angel funding. Higher house prices, higher rents. None of that is a surprise. But I think the good news is much more talent will free up also out of Facebook.

That's true.

And more people will be motivated. Yeah, the people you're really interested in, the people who say, "Hey, I can do the same thing Mark Zuckerberg did." Those are the real gems. The people within Facebook will take the risk, especially since they know they made enough money where at least they won't go hungry, you know, and then want to do something more impactful. One of the most interesting things I've seen is people who have been at successful companies, whether it's Facebook or Google or other places, and the ones that are most interesting to me are not the people who say, "Hey, I want to do a startup. I made a million dollars. I want to make 10." It's the people who say, "I made a million bucks. Now I'm free to do what I really want, what I'm passionate about. I want to do something much more impactful." And those people are so much more committed to their vision that they become almost missionaries. I found mercenaries are just a bad formula for successful startups because when, when a startup runs into problems like every one of you will, right, the mercenaries sort of hide, right, or go away or find a new job or find a new startup. The missionaries stay with their mission and find ways around their problems. I believe that sort of passion for a vision is an essential ingredient. You have to commit to to a religion that you believe in. And no matter what difficulties you run with it, you question yourself, you adjust, you make the kinds of adjustments you talked about, but you don't lose confidence in yourself. You know, Maya Angelou said, "I may face defeat, but I won't be defeated," or something like that. And I think that's really, really important. There was another favorite one of my quotes, "I'll fail and I'll try and fail, but not fail to try." So these things are essential ingredients for startups. I'd plaster these things on the wall in every startup, just as sort of the kind of culture you need to develop. I know they sent out some emails on Some of the talks I've given, much of those talks, especially the one I gave at MIT and then the one I gave at our CEO conference, the Vinod Khosla CEO Conference, which are all on our website also, were more about both at the high level and then at the very actionable level, the kind of culture you want to establish, the kind of process you want to establish, And it's not progress process in the rigorous sense, but sort of culturally how you behave, behavior patterns you want to establish. Those are absolutely key to successful startups. And the MIT one, for those of you who have the time to go look at it, was much more meant to be inspirational for 800 kids coming out of MIT. And the one at our CEO conference was meant to take the same things and translate them into tactical actionable things. How do you sort of make some of these high-level concepts actionable? I'm big on making everything actionable. Because most of the books and stuff I read are so high-level, I say, "What do I do with this?" You know, "How do I use it in my business?" So—

Malan Thakur. When brainstorming or exploring ideas, at what stage do you think an idea gains enough traction to devote your time entirely to it? How do you do it?

You know, that's a judgment call. And it's a complex judgment call. If you're at the stage where you're experimenting with 6 fundamentally different ideas, I wouldn't commit to one until you developed a real passion, a real belief in that vision. Because if you don't really believe in it, you'll keep adjusting away from it too quickly before you've tried all the ways. Keep remembering that things can fail because what you're trying to do is wrong, but they can also fail because how you're trying to do things are wrong and what you're trying to do is right. And I find people often think what they're trying to do is wrong when it is right, but what— how they're trying to do it is wrong, and they need to adjust their tactics, not their strategy. So think, you can have the right strategy and the wrong tactics and fail. You can also fail because you have the wrong strategy or the wrong end goal, but more often than not, it's just people need to adjust how they're tactically trying to approach something And those adjustments and that brainstorming in a good team will then get you to the right answer.

Joel R. Davidson: Given the rise of numerous consumer apps, are the brightest engineering minds solving the right societal problems?

You know, it's too highfalutin to say what's a high— right societal problem. What I find is people do things that they have a passion for, what their vision says. And then everybody has their own view of what impact is. You know, you can have a lot of impact because you change healthcare with a great startup. Or change the world with an information engine like Google. But you can also have a great impact because you gave 100 people new jobs and did a video game site. Maybe that's not a societal problem, but you did provide a lot of people jobs. That's probably more meaningful to them than having access to Google. One of the most rewarding things of a successful startup is when people come to you 5 years later and say, "Boy, you You helped me buy my house, or, you know, I had this problem and it sort of went away. Those things— so reward can happen at many different levels, and it depends on your personality. I wouldn't be too judgmental or highfalutin. Now, if you came to me and said, here's a great gambling idea, we won't invest. I don't like to invest things I don't think have a positive impact. But that's a judgment call. If you came to me with a great video game idea, we invested in Rumble, we'll probably do it. Now you may say, well, that's not, that's just a way for people to waste time. But those are judgment calls each person makes. You know, we've turned down many a gambling idea, but we've done video games or other things that others might not consider impactful. By and large, the more impact you have, the more fun it is, the more committed you can get to the idea, more religious belief you can have in the idea.

Friends learn, what are the forcing functions that entrepreneurs can use to close investments with VCs? What are the best techniques to actually close a round?

Again, that's a complex question. There's no simple magic. First, let me again refer you to a presentation on our website called— I think it's called— somebody will correct me. You'll find it for sure. I don't know if you'll find it under this name. It's called How to Pitch a VC. And it's really about how you make a presentation. One of the things— so let me give you a few examples. There's some tactical things on our website on how you make a presentation. What's too much content on the slide? What's too little? What kind of font to use? All those kinds of little details, and I spend a fair amount of time talking about these and how to make a more effective presentation because I tend to get very tactical in helping entrepreneurs. But let me give you some high-level things that I also talk about. Very often people will tell me a full story. Here's the idea, here's the mission, here's the marketing, here's the finance, here's the— that's really not how VCs make decisions. So I tell entrepreneurs, if you're going to pitch a VC, figure out the 3, 4 things they'll fall in love with. Almost every time you hear a presentation, you go out, talk to one of your partners and say, Here's what was great about it. Here's what I liked. And you sort of have to say, if you answered the question, why should they invest? You're much better off than doing a normal business plan kind of thing you learn in business school. That's why I stay away from MBAs. I'm an MBA, so I can say that. I'm 57, so I can criticize people over 40. But it's not completeness in the sense of having a marketing section, those competitive section, all that. It's why should somebody invest? What's the 30-second elevator pitch, the 3 reasons they'll tell their partners? Focus on creating greed there. If you convince them those If those reasons would result in a big company, great. The other part of it, which entrepreneurs mostly miss, is the fact that if they don't address the reasons why investors might be afraid to invest, that somehow the investors will skip them. All that'll happen is they'll decide why they don't want to invest without discussing them with you because it's not on your slide deck. Because you tried to hide it, you won't get a chance to address it. So there's prototype presentations on our website that say here's the reasons to invest, here's the risks in the business, here's how we are addressing the risks. So investing is— investors only have two emotions, greed and fear, right? Fear of losing their money, greed in making a lot. And so if you sort of keep If you can keep that and sort of address why your opportunity is great for their greed and why the risks in your business can be addressed so you can address their fears, that's sort of the best overall thing I can tell you. If you don't address those, they'll make one of a couple of conclusions. One, that you're not realistic, you're naive because you don't know what your risks are or what your problems are and you didn't address them. Or they'll make up their own versions of those and you won't get a chance to discuss them. So in that sense, people also— investors also are very sensitive to pitches that are all hype. They hear enough pitches that they know what's hype. And so one of the important things We as investors judge is whether somebody is really genuine. Now, some we also care about, would I enjoy working with the person? How do they communicate? How honest they are? What kind of relationship? What the chemistry is? All that's important. So.

So you spend a lot of your time giving of yourself to others and giving other people advice. So N. Black wants to know, who do you look to for advice and who actually keeps you in check?

Well, so surprisingly, I spend an awful lot of time with people who are naive in lots of areas. But they're the source of fresh ideas. So I find my biggest failure mode is being too cocky. Being too arrogant about knowing everything or having seen everything. So I try and spend an awful lot of time with people with fresh new ideas and say, "Is there a new idea that I haven't heard of before? Is there something I haven't thought about before?" Now there's two kinds of things, two kinds of behavior patterns. If you look at most psychology, there's sort of Most decisions are made by what Dan Kahneman in his book Thinking Fast and Slow calls System 1. Basically, the primal human brain makes most decisions. The neocortex does all the thinking part. The primal brain, System 1 as he calls it, has an immediate response without justification to almost all the input. If you've— and the older you get, the more you fall into this pattern. You have an experience base, you have an immediate response, you go act on it, that's your bias, you won't change your mind. I try and spend a lot of time questioning those assumptions. Usually what happens in most humans in this, the System 2 or the neocortex, actually makes logical decisions that don't have an intuitive answer. Most of the time when your System 1 has made a decision based on your biases or previous experiences, System 2 spends all this time justifying that decision. Okay? And I'll give you some examples that are very, very funny. I'll come back to that. What you really want to do is not rely on sort of your primal response, your System 1. You want to get all the input and then say, "What are all the reasons why whatever I've decided can be wrong?" 90% of time is spent justifying why you're right, not why you're wrong. It's very, very important to think. And I use people who I've— whose perspective I've never had, like a young kid who's 23, listening to them on why I'm wrong or what have I missed, some new ideas, some new— and that sort of keeps me fresh and younger. Every time I give a talk, one of my daughters will send me a email about all the things I did wrong. I get that often.

The author of The Practical Genius says everyone needs to have a fat brain in their lives. Clash Scavenger: Can you talk about founders bringing in CEOs? At what point should they bring in a CEO?

Boy, tough question. Let me first talk about the pros and cons. Of bringing in the CEO. As a founder, you have a lot of vision about what you're trying to achieve. There's a lot of things you're missing about running a process. How do you run a company? How do you make people get the most out of people? What are the nuances in the market you're going into? What traditional behavior will you run into that you haven't yet thought about? Bringing all that expertise, and I talk about a lot of this in my gene pool engineering paper, is important to create this dynamic of challenging all your assumptions. There's also a separate paper I have on innovation that's also on my website. It's too long, it's about 30 pages long. I write a lot. Mostly to try and be helpful to entrepreneurs. Sometimes I write for myself to clarify why I think a certain way because I often do conflicting things. I'm not consistent with myself and then I'll write them down and try and dig deeper of why was I inconsistent with my own self. You know, and it's amazing how often I'm inconsistent with my own beliefs. I just try and catch them and then examine them deeper. It usually leads to some new insight. I'm pretty analytical about those things. But back to your question. If you can create this dynamic of all the different experiences you don't have and how to manage them well, that's a great thing. Now, do you need a CEO to do that? Sometimes you do and sometimes you don't, and mostly it depends on your personality. So I'll give you an example. A few, 2, 3 years ago, we invested in a company called Lookout. They do mobile security. They had all of 1,500 downloads, not 15 million or however many they have today, right? And the CEO was a young founder in his early 20s and, you know, we talked about this and I said, Hey, pretty unlikely you'll grow based on what I've seen to be a CEO. I was telling my partners. And then over 18 months, he was so good at listening and bringing in expertise he didn't have and supplementing his team that I said it'd be a shame to get a CEO here. And I've spent all my time trying to help him and coach him so he doesn't have to get a CEO. And I'm almost certain now he won't need a CEO. But you, let me give you another very specific example. I mentioned Square earlier. Jack Dorsey bought Keith Rabois as COO. I think the company would be worth half as much without Keith. But Jack was able to focus, decide what he did really well and what he couldn't do well. And who to bring in. And because he did, first, the company scaled way more than Jack could have scaled it by himself because he bought Increeth. Two, Jack actually has empowered himself to make many more strategic decisions, spend a lot more time on strategy and on product, which is his strengths. So the situation is different. Some entrepreneurs abroad, some are not. Some CEOs want to take control, some CEOs want to empower the founder. So let me give you another great example. Scott Crook, when he was building Intuit, brought in Bill Campbell as a CEO. And Scott would say it was one of the best decisions he made. Even though Bill was the CEO, Bill never did any strategy or product and left all the critical functions that determine the success of introduced to Scott, he could do it full-time. It was the same relationship Jack and Keith have. But the titles became almost irrelevant. In Intuit's case, the new guy had the CEO title and Scott had the chairman title, but I would say he did 90% of the key decisions on product and strategy. And Bill did 90% of the decisions on operations. Same thing's happening at Square, but Jack, um, retained the CEO title. So it's, can you engineer this environment? And that's why I again refer you back to this paper on gene pool engineering. The key thing is what are you missing in skill sets and experiences, even biases. Uh, you want different biases in sort of managed conflict within your company on each of those biases to get the best out of and not be too swung in one direction by the bias of one person, which is mostly the experiences of one person. You want an engineer that's kind of, I sometimes call it managed conflict, sometimes I call it organized chaos. There's sort of hard thing to describe. But you want to engineer that inside a company. And then if you have to give up the CEO title, it's largely irrelevant if you have the right CEO. And if you have the right board, if you end up hiring this wrong CEO, they'll get rid of the CEO, not the founder. The intuitive response of VCs is to just get rid of the founder and keep the CEO because he's a better manager. But if you had to pick between great management and great vision and intuition, I'd every time pick great intuition and great vision over great management because there's other ways to do management. It's very hard to replace intuition and— so Google's another example. I worked a lot with Larry and Sergey when they were were starting out. They got a CEO, but the CEO didn't take vision away from Larry and Sergey. The title almost irrelevant. Each did the things they did well. Sometimes it's not possible. Sometimes the CEO coming in say, "I won't do it and I need the CEO title." So each situation is different. It is a hard question for you to decide Early in a company, it's a bad idea to bring in a CEO, especially till when you're still iterating and using mostly your intuition to iterate on tactics to get to your vision. A good manager will mess up that process. Almost always the ideal situation is if there's a great CEO with a vision, a founder CEO, and a team under which, which is very different than the founder CEO that can execute, has other skills that the founder is missing. Almost always that's the preferred situation, but sometimes that doesn't work because the characteristics of the founder, and founders have to be realistic about that.

Todd Eldsberg, given current economic trends, how aggressive Should a company be around growth versus hedging for downward risk?

I get asked this question often. To be a good startup, you have to be schizophrenic. You have to have a very optimistic view of the world, have a large vision. But you have to be paranoid, and I often recommend people read Andy Grove's book, which was written in the '80s, called Only the Paranoid Survive. So you have to have paranoia, and you have to have sort of vision and optimism things. And so that's why I say you really have to be schizophrenic. And a good founder is nuanced on when to be schizo— when to be when to be optimistic and believe in their vision and passion. And a good board will help. But let me get much more tactical. So I often ask people, what are all the things that could go wrong? Now what does your business plan or your financial plan look like then? And ask, if you have to take advantage of your upside, what's the best plan to put in place, press on the gas? And I sort of force companies I'm working with to go through that exercise at every critical juncture. Sometimes just monthly new business plans or financial plans. And both the upside plan and downside plan. And then I ask, how do you switch from Plan A to Plan B? If you're on this aggressive step on the gas plan, you have to have done it in a way where with 60 days' notice you can downscale to 70% less burn rate. So for example, if you have to pay twice as much for something you need done, whether it's renting space month to month instead of on a 2-year lease, or hiring a contractor to do something, but it's much cheaper to hire an employee. You keep as much flexibility as possible and almost think of your mind oscillating between the upside plan and the downside plan, always keeping, when you're on the upside plan, the ability to ramp down in case things go wrong. I learned that lesson the hard way very early in Sun's history. Almost nobody thinks Sun ran into trouble. There was a time we were shipping these monitors— those days computers came with monitors— and we got a new supplier, started shipping them. Within sort of 2 or 3 weeks, we discovered that the average life of our monitors in the field was 30 to 60 days. I think we worked at least 20 hours a day for 2 months trying to find the problem. We couldn't, and we were sure we were going to go bankrupt. And we did everything possible to ramp down every penny. And I think the earliest I left the office was 3 AM during that 90-day period. And I was always in by 8 AM. It was just our life depended on solving this problem and extending our runway. So I do think one of the mistakes entrepreneurs make is when they build a plan, they believe it's going to happen. Always wondering what the worst case is. Is absolutely critical, and sometimes people say I'm too aggressive and sometimes people say it's too conservative because I'm trying to oscillate between these in my guidance to an entrepreneur I'm working with. And when to do each is absolutely critical. And there are no rules about it. It's judgment calls. For example, if you don't have competitive pressure, you can be a lot more competitive— conservative in your financial risk. If there aren't huge network effects, then somebody else catching up isn't going to make a difference. If there's huge network effects, somebody gets those network effects, you're going to lose. Then you have to take the risk. So think about this issue completely differently. The job of founders and CEOs in a startup is not about building the company. It is, but the more important job is managing risk. Let me give some examples. You hire 5 more engineers, you have a lot less schedule risk, right? But you have a lot more financial risk. You say, "I'm going to add these 3 features to my product." You've reduced your marketing risk, but you've increased your financial risk, or you've reduced your comparative— you've increased your comparative risk because it's going to take you longer to launch the product and competitors may catch up. So almost everything you do increases some risks and decreases others. And people never think about it this way. If I hire 5 more people, my financial risk is much higher. If I do minimum viable product, which I'm a big fan of Derek Reece's methodology, I am increasing my market risk. It may take me longer, it may take me— I may have wasted the product, so if you sort of think of one of these spider charts with risks and say, what risk does every action increase and decrease, that's a good way to think about tactical decisions as you make them.

I'd like to open it up actually, if you have a microphone out here. I've been asking actually all of your questions, but there are quite a few that I haven't been able to get to. And while we're passing the microphone around, I really want to get to some of the more specific industries like energy, big data, as well as advertising. There's some questions there. And one question here is, "Venture capital the right way?" So there's a question back there. You have one already?

Yep. There's a— you can stand up and maybe we'll try and repeat the question.

Who is it?

Hi. I'm interested in your perspective on the shale gas. Revolution, and there's interesting gas-to-liquids technology that I'm wondering if you have invested in any of that, or if you're interested, and what your thoughts are generally on sort of the shale gas revolution and how it's going to relate to the energy industry. So I'll get to that answer in just a minute, and one of the things I'm going to try and do is generalize most of the questions so people who aren't interested in shale gas still get something out of the answer. But one of my favorite things that I told you I'd come back to is this idea of experts. You'll get advice from a lot of experts. And Professor Tetlock at UC Berkeley did a big study, 20-year study. He took 250 of the best experts he could find in the press, most credible experts. Followed them for 20 years across 80,000 forecasts. Now these are guys like Henry Kissinger's view of China or the Middle East. I mean, I'm talking about premium experts across 20 years, 80,000 forecasts. The rough accuracy of these premium experts was about the same as dart-throwing monkeys. And so I think Disrespect for expert opinion is an important part of being an entrepreneur. Irreverence, and it's in my MIT slides, is an essential characteristic. If anybody believes something, I always want to be a troll and prove them wrong. And that's good attitude to have, to have a real questioning attitude and not believe anything you hear. Take all the input, examine why they believe those things, what part of it is right or wrong. Being analytical about it is very, very good. You don't want to reject it out of hand, but believing it out of hand is more dangerous than rejecting it out of hand. So bias yourself to assuming anybody who's an expert is probably wrong. Why it relates to shale gas is the following. About 5 years ago, you know, the energy world, nothing changes over 30 years, that's what every expert tells you. I just gave the keynote at a Navy planning session where they were planning the Navy in 2040. And I asked them how the Navy's mission has changed since 2005 and it was completely different, because China's role is different, The hotspots are different, the Mideast has changed, there were so many tactical things that had changed. They said, "Well, it's completely different." And I said, "Why are you trying to plan 2040? Is the rest of the world going to stay the same?" This was my keynote talk to the 150 most senior admirals in the Navy. Energy is the same way. 5 years ago, the hottest investment area in natural gas in this country was building LNG terminals so we could import natural gas. And then there was a breakthrough in two technologies, one called fracking, one called horizontal drilling. Fracking's been around for a while, but that combined with something called horizontal drilling, instead of drilling down you drill sideways, they discovered they can extract gas from shale. Couple of popular shale sites, Pennsylvania, Marcellus Shale, couple in New York, other places. Suddenly, each LNG terminal was about $5 billion. They were being set up to import gas. Suddenly, there was no shortage of gas and you didn't need to import gas. Now people are pivoting to saying, can they use these to export gas because there's so much gas? So 5 years ago, everybody thought there was a shortage. Now everybody thinks there's an excess of gas. Technology does cause large shifts. Because now there's an excess of gas, literally in 2008 the price of gas was $14 per MBTU or per unit. Today it's $2.50. All the assumptions and spreadsheets built on these by all the experts have now changed. So the question was, can you convert all this excess natural gas into liquids and run them like gasoline or diesel? And yes, that's a very, very interesting area. There are people doing it, we are spending time looking at all those, but I don't want to go into the details because it's probably not relevant to everybody. There's a question here. You can just stand up and— Oh. What area are you investing in that you're most excited about, that keeps you up at night and that you just love thinking about? I'll again give you a piece of reading to do. Turned out I blew my ACL while skiing over Christmas, and so I had time to write. One day, the day after I blew my ACL, since I couldn't ski, I wrote about 10,000 words. That turned into 4 long blogs on TechCrunch. So early January, I published 3 blogs. Part 1 was called "The Surprising Path of Artificial Intelligence," and it was about machine learning in general and data analytics. Part 2 was called "Do We Need Doctors?" I'm convinced everything about medicine will change. You know, 10 years ago, if you said to anybody the internet will replace telecommunications, those voice calls, and something called ATM technologies, everybody would have said, "Never." And if you went to a place like India and said mobile phones will be relevant, people would say, "You're crazy." That was only 10 years ago. Today, they're adding 20 million new subscribers a month. A month. So every few months they create a new Verizon. So healthcare will be upended, almost certainly not driven by the health system, but driven by consumer health. So the second one was, will we need doctors? I think 80% of what doctors do will go away and 80% of doctors will go away. And frankly, who wants that? A computer will do much better than your average doctor in most things. Not everything, but in most things. And whenever I give talks like this, I get misquoted, so I have to be fairly precise. I don't want the average doctor. I want the best doctor. And I want to give him the tools to replicate himself 20 times over. Right? And that's what will happen. They will be leveraged in ways that gives me and everybody who can't afford the best doctor or is not in the right location, is in Jackson, Mississippi, can still get the best doctor for whatever their specific problem is. That's a very interesting area. I suspect just like everybody has a blood pressure meter at home, they'll have an EKG machine at home. We just signed a term sheet this week to do that. Machine that would cost a few tens of dollars would give you a medical grade EKG in your home as often as you want. And if you don't have a doctor to look at your EKG, the software will tell you or help your doctor look at it and tell him when they need to pay attention. Then I wrote a blog, my third one was called "Do We Need Teachers?" because I actually think we'll get rid of— you know, most teachers are bad teachers. Because they're there for a paycheck, not because they have a passion for teaching. People get upset at me when I say that, but the fact is 20%, 30% of teachers are really passionate about what they do. They love their job, they love teaching, they love kids. Those are the teachers you want to leverage. People who just are bad teachers should just go away. They're there for a paycheck, they do the minimum, and When they have to choose between helping a student or watching a movie or going to the beach, guess what choice they make? Right? And so I do think teaching will change and education will change at all levels. The fourth blog I wrote was— included those two areas, but actually is probably the thing I'd refer it to. It was in sometime in mid-February called "The Dozen Unhyped Areas in internet and mobile. Big data was part of it. But there were some surprising things people don't talk about. I defined— obviously I defined education as a category and consumer-driven health as a category, but nobody talks about emotion as a category.

Huh?

So if you look at something like food spotting, it's all about emotion. I was looking at a startup yesterday and I said, "Your startup is all about emotion. It has nothing to do with the utility of what you're trying to do." Zynga is about emotion. It's manipulating your emotions. There's no— nothing else involved. So emotion is a category that's heavily underused, underutilized. Pinterest is about— is an emotional product. It's principally an emotion product. So go read that blog. It'll tell you the dozen areas I think, and I suspect I've missed more than I included because I always miss lots of things.

Actually, sorry, I had the mic. Ajay here. Question about emerging markets. There's a lot of wealthy people now who have money, and the markets are not sophisticated enough to drive that market to a VC fund or an angel fund. Like, for example, I'm having a difficult time right now, like, convincing my dad, who has a lot of money, to invest in a VC fund or an angel fund in India. And I'm wondering, are VCs doing anything about educating people in emerging markets to drive their money from gold and bank account to actually invest in funds that will actually create— will fund ideas and create more jobs? Because, I mean, I don't think even, like, you know, talking about Europe, you know, Merkel and Sarkozy can argue all they want on fiscal and financial policy, but at the end of the day, it's really startups and that will create jobs and innovation that will create jobs. So I'm just wondering your perspective on— Yeah.

So I don't think traditional education works. You know, what really works is creating role models. So if there's a couple of— first, I do think emerging markets are ignored in the startup world a lot. And, but once there's one or two big successes, everybody will start investing in VC firms. So, the key in any of these things is to create a role model. Nobody thought social was important, and then Facebook created role model, and suddenly social is the fashion. Microblogging, both Tumblr and Pinterest created sort of this idea of a different kind of blogging. And Twitter created their version of microblogging. I think role models is the way things change. And if somebody makes a lot of money, people get greedy, and then suddenly it looks like all roses, and then you'll get overinvestment, then you'll get a few people losing money, and then you'll get the fear part appearing again. So we'll just keep oscillating between fear and greed and driven by role models, both good models for when it excites greed and bad models like Solyndra when it excites fear. Yes. I'm curious about your perspective on energy poverty and how there's nearly 1.5 billion people in this world who don't have access to electricity and so this is a huge global problem that really hasn't scaled. In terms of finding a solution. What are your— what's your perspective on finding a solution that might be more scalable in bringing these people energy? Again, let me give you a quick answer. I'm generally pretty positive and optimistic about finding solutions to those problems. If you can move a problem from the domain of big, large companies that want to maintain the status quo into the entrepreneurial world you can guess I'm a fan of the entrepreneurial world, you'll solve the problem. Almost no problem, almost, not 100%, but probably 99% of the problems are solvable if you get entrepreneurial talent working on it with their own ideas. I call it the power of ideas fueled by entrepreneurial energy. There are a couple of details. So, one of the things you have to do is get a lot of experimentation going. Lots of experimental— experimentation gets going if you have a few role models of where people made money and then everybody gets greedy and they'll fund every new experiment just like there are a lot of Facebook-like experiments today. So, you have to create this environment of role models that incite some greed so People will fund lots of ideas at the entrepreneurial level. Almost certainly no large company will solve any of those problems ever. Whether it's healthcare, whether it's education, whether it's energy, all of them will be solved by driving entrepreneurial energy, a lot of experimentation, and frankly tolerance for failures because failures come from experimentation. That's the nature of experimentation. If you get that ecosystem going, and I think we've started on it, it'll take a while, that's why I'm optimistic. So last question.

Shaheen Qazi.

Oh, there's one question there. We'll go in the back and stop there. Yes, I'll keep it short.

Okay. Thanks. Shaheen Qazi. And my question is, right now the pendulum has kind of swung towards centralized intelligence with cloud computing and with big data. And at the same time, we're seeing some new technologies on the horizon like AI and machine learning sensors that really are paving the way for distribution across everything. So my question is, are we going to kind of see a pendulum swing again, or is— are we moving towards some kind of balance in the network?

I think that's a false dichotomy. You have lots of sensors, lots of distributed data, a click on your mobile phone app, all of that leveraging cloud services to do machine intelligence, learning systems, big data analytics. All those are really exciting areas. I wouldn't classify them as either distributed or centralized. They're just good intelligence systems. So why don't we go to the last question in the back there? Uh, okay. Yeah. Just which was the problem with these monitors at, uh, Sun Microsystems, which almost brought down the company? I'm sorry. I didn't hear the question. Yes. Uh, if you can hold the mic closer to, okay. Which was the problem with these monitors, you know, at Sun Microsystems, which almost brought down the company where you worked 20 hours a day for 3 months? We had sourced monitors from Philips, and they made an engineering change to their production line, which was not supposed to change anything, but Murphy's Law always works. Thank you all for coming. I hope this was useful.