(Edit: Removed Brace from the recommendations - their service wasn't working for me at all unfortunately. Will update if that situation changes - http://forms.brace.io/. For now, I'd suggest Formend, which is working perfectly.)
Russ R: Welcome to Econtalk, a library of economics and liberty. I'm your host Russ Roberts of Stanford University's Hoover Institution. Our website is econtalk.org where you can subscribe, comment on this podcast and find links and other information related to today's conversation. you'll also find our archives where you can listen to every episode we've ever done going back to 2006. Our email address is [email protected] We'd love to hear from you.
Today is May 1st, 2014 and my guest is Marc Andreessen, legendary entrepreneur and venture capitalist. He's the company-creative of Mosaic, the co-founder of Netscape, and more recently, the co-founder of the venture capital firm, Andreessen Horowitz. Marc, welcome to Econtalk.
Marc A: Russ it's great to be here, thank you.
Russ R: I want to start with your career. You were at the heart of the first browser war between Netscape and internet Explorer. That seems like hundreds of years ago. It's a little more recently than that but give us a quick thumbnail of what happened to Netscape and then how you escaped from that war.
Marc A: So, Netscape was founded actually 20 years ago this month. It's sort of a special time for me. Netscape is a company that grew incredibly quickly. We grew from zero 3500 employees in three years. [inaudible 00:01:25] Netscape was always thought of as this small little browser company and people used to come visit us on the campus and they would be completely flabbergasted at all the buildings and all the people. The reason we had all those people was because in addition to browsers, we built a very broad range of internet software and internet services at that time. We were one of the first companies that did internet ecommerce systems. We were one of the first companies that did internet publishing systems and software for a long time, powered actually a lot of the newspapers [inaudible 00:01:53] the Wall Street Journal.
Russ R: What happened to Marc Andreessen at the time of that sale. Did you go with that sale or did you leave?
Marc A: I actually became the Chief Technology Officer of America Online. AOL at the time was the most valuable internet company. AOL within a year after our sale, was a company that was worth 170 billion dollars on the stock market. That was a company that was in a dialup ISP business, if you recall that, at 25 million dial up ISPs. That was right before they went over the broadband, which completely destroyed the dialup ISP business. For those who believe in the efficient market hypotheses, one of the [inaudible 00:03:18] examples would be a dialup ISP worth 170 billion dollars went straight into broadband; which made no sense at all.
Russ R: It didn't last that long.
Marc A: It did not last that long. In fact, what had happened was, the management team at AOL actually figured that out. They realized it. What happened was, they traded their equity which they knew was going to collapse in value. They traded it for Time Warner equity and they bought Time Warner. That led to the famous AOL/Time Warner merger which is one of the great catastrophes in business history. I stayed for about a year. We have a tradition in the tech industry where when your company gets bought, you have a period of indentured servitude where you stay for at least a while to help make sure the integration happens. So I stayed for a year and then went off and started my second company.
Russ R: Which was?
Marc A: Which was actually the first cloud computing company. this now big trend of cloud computing was something we helped kick off in 1999. That was a company called Loud Cloud. That was the first company doing it, but today it's done with things like Amazon Web services. That company actually grew very fast. We grew from a standing start to quite a large business in the course of a year. Then we hit the dot com collapse, like running into a buzz saw. Half of our customers were dot coms; which all went bankrupt. Then the other half were big companies who were in a panic because they felt they had to compete with the dot coms by launching all kinds of new internet efforts. When the dot coms went bankrupt, most of those big companies said, I guess this internet thing isn't serious and it's going to go away; they shut down most of those efforts.
We almost lost that company. My partner, Ben Horowitz, recently written a book which is now a bestseller called, The Hard Thing About Hard Things. It's a business book and it tells the whole story of Loud Cloud where he was my business partner and he was the CEO. It's a dramatic story because we almost lost the company. We almost went bankrupt. Then through a series of miracles, we were able to do what's called a restart. We basically completely restarted the company as a public company into a completely different business. Ultimately grew it to be a successful company. Hewlett Packard bought that company in 2007 for about 1.6 billion dollars.
Russ R: Was that your main use of time in that 99 to 2000 to 2007 period?
Marc A: Yeah, I started those companies basically back to back. Netscape was basically a company in the middle of a boom, 1994 to 98 and sort of rode the upper momentum of the 90s tech boom. Loud cloud was started at the very end of the boom. We started September of '99. We had about six good months before everything caved in. Most of loud cloud which became opsware, most of that company was through the bust. We just kept all the way through the clash and kept slugging away through 2002 to 2004 when things got really miserable in the tech industry. We sold the company right as the industry was coming out the other end in 2007, heading straight into the credit crisis. In retrospect, that was probably good timing to sell that company as well.
Russ R: At that point, when did you start Andreessen-Horowitz, the venture capital firm?
Marc A: We started the planning for the firm a year and a half planning it through. We started that process in 2007. My partner Ben and I sort of spent nights and weekends writing the business plan and thinking through all the strategic things that we had to work out. Then we kicked off the fund-raising process in March of 2009. I remember that very distinctly because March of 2009 was the low of the stock market after the credit crisis. Nobody was raising an venture capital funds in the spring of 2009. It was not a time when investors wanted to hear about any venture capital fund. In fact, many of the large investors in venture capital and private equity were in a liquidity crisis and [inaudible 00:07:28]; including the big [inaudible 00:07:31] of dominance where they were having a real trouble meeting their commitments back to their sponsored organizations.
people have told us it was the most hostile time to raise any venture capital fund in 40 years. Of course, contrary or perverse, depending on how you look at it; we said, that's probably not going to be a very good time to raise venture capital because ...
Russ R: ... You'll be alone.
Marc A: Exactly, we were. There were only two venture capital funds that got raised in 2009 in the entire year. One was ours and the other was [inaudible 00:08:03] where I [inaudible 00:08:06]. He's one of the top venture capitalist in history of all time.
Russ R: Who is that again?
Marc A: It's a gentleman named Vinod Cosla. He runs his own firm called Cosla Ventures. He previously was partner at [inaudible 00:08:18]. He was one of the partner at Kline-Perkins who made them so successful in the 90s. Before that he was company-founder of a company called Sun Microsystems.
Russ R: I've heard of them.
Marc A: Which was a big successful technology company. He's one of the legendary tech entrepreneurs and investors of all time. I think it was straight forward for him to raise the fund. It was a bit harder for us but we were able to raise it. In fact, that turned out to be a very good time to raise a fund because it put us in the position to invest when a lot of other people had stopped investing.
Russ R: you've been around now. the firm's been around for five years. It's been five of the least pleasant years in the American economy that weren't called a recession. Officially the recession ended in 2009 but it's been a pretty mediocre run for the us economy since then, but the technology world has been okay. You've invested, according to the internet at least, of course the internet never lies. You've invested in Facebook, Twitter, LinkedIn, Foursquare, and so on; which are pretty successful companies. My question is, does that make you feel smart or is investing still a very humiliating and uncertain process?
Marc A: Well, venture capital is kind of guaranteed to be humiliating because the most successful venture capitalists of all time typically tank about half their investments. By which I mean, if you taken any of the top performing venture capital firms over the last 40 or 50 years, if you get inside their portfolios and look at their portfolios. These are the top firms, the ones that return amazing returns over long periods of time. Typically they lose half their companies. In other words, half their companies go under and they either return nothing for the original investment or they return a fraction of the original investment.
It's a feast or famine business. In the same portfolio, you'll have both feast and famine. You'll have a company that gives you a 10 x , take AOL, a 100 x return and you'll have six other companies that are failing. The twist to how you spend your time is you spend most of your time actually dealing with your companies that are struggling and trying to help them. It's the companies that are struggling or failing that actually need the most help. The companies that are succeeding are generally doing just fine without you. The companies that are failing are the ones who really need help and support. A lot of what you end up doing on the job is supporting struggling entrepreneurs. It's kind of continuous [inaudible 00:10:44]. you're a troubleshooter. There's always something going wrong. We talk about this with partners, you have to be psychologically prepared for the opposite. It seems like it would be a life of glamour and excitement. It's more of a life of struggle and misery.
If you're okay with that, because it's part of the package, then the overall deal is pretty good.
Russ R: I still call it nice work if you can get ...
Marc A: ... it is. It beats all the jobs I had when I was a teenager.
Russ R: Exactly. What's interesting to me as an outsider, you learn things as you go forward. You still fail at least half the time. It's not like, well those five of the 10 were the losers; next time I'll miss those or won't invest in those. You still have to fail five out of 10. You don't know which five are going to be.
Marc A: The reason for that is we think you can draw a two by two matrix for venture capital. This is probably true for all investing but it's certainly true for venture capital. You can basically draw a two by two. On one axis you could say, consensus versus non-consensus. On the other axis, you can say successful or failure. Of course, you make all your money on successful and non-consensus. Definitionally the reason is because it's very hard to make money on successful and consensus because if something is already consensus, then money will have already flooded in and the profit opportunity is gone.
By definition in venture capital, if you're doing it right, you're continuously investing in things that are non-consensus at the time of investment. When we translate non-consensus to [inaudible 00:12:19] terms, it translates to crazy. You're investing in things that look like they're just nuts. Who would believe this PC thing would work at the time? When the VC's were investing in PC companies, the whole thing was considered to be a joke. When I started with internet startups, the one thing everybody knew for sure was that nobody would make any money with the internet. There would never be a business.
Russ R: Right, it's just a toy for communicating a little bit and interacting but there's no revenue stream that's going to come from it.
Marc A: Certainly not, obviously the one thing you know for sure is certainly nobody's going to buy anything from over the internet. That would be crazy because hackers might steal. You go on and on. EBay looked non-consensus right at the time venture capital firm called [inaudible 00:13:08] made EBay investment. A lot of people at eBay was like, that's crazy. Who is going to buy something from a seller somewhere around the world who they never met before? Like, that's just nuts.
Russ R: And it's used. It's not even a new thing. It's often a piece of ... it's garage sale material.
Marc A: Yea, exactly. That's exactly right. That continues through [inaudible 00:13:28]. More recently you have these things, AirBnB we're very involved in. AirBnB is this idea that you're going to rent out the back room in your house and a random stranger is going to show up and stay in the room. That's crazy. You would never do that. You would never rent out the room. Of course, you would never stay in a stranger's house. It turns out at AirBnB that people love this. It's growing like crazy. The revenue's exploding. The corollary to this is that crazy does not mean correct.
Russ R: That was my next thought.
Marc A: Crazy often just means crazy.
Russ R: Darn.
Marc A: Yea, exactly.
Russ R: I used to teach a business plan class in a business school. One of my students submitted a business plan that in the financials over the life of company made no for the money for the foreseeable future. I said, that's a bit discouraging to a potential investor. The student said, all the best companies don't make any money, so that's a plus not a minus. That's the challenge. Every crazy idea is not successful. It's a reverse causation confusion there. Despite all that, you've made some very successful bets presumably on non-consensus companies that at the time 140 characters as communication seems like a ridiculous idea. Are there some you missed that you regret that you want to talk about? Are there some lessons you learned that now you can't believe those were the mistakes you made back then?
Marc A: Sure, the corollary of this is the mistakes that we make in a field like venture capital; the mistakes generally aren't investing in something turns out doesn't work. This is something [inaudible 00:15:20] struggles to figure out. You put all this money in something and it failed [inaudible 00:15:23]. That's generally not the problem. The problem is what you decide. It's the big hits that you missed. Every venture capitalist that had opportunity to invest in Google and didn't just seems like an idiot. Every venture capitalist that had the opportunity to invest in Facebook and didn't feels like an idiot. The challenge in the field is all of the great VC's over the last 50 years. The thing that they have in common is they all failed to invest in most of the big winners.
This again is part of the humility to the profession. You literally, as an investor in any of these things, even if you've invested in some of the hits, you've made a long list of the hits you missed. Those are the ones that drive you crazy. I was not a professional investor prior to five years ago. We'll see in the fullness of time which ones I missed in this period. One of the things we do a lot is we try to back test our theories against history. We sit around all day long and talk about what are we looking for and how do we know if [inaudible 00:16:22]. We sort of [inaudible 00:16:26] criteria for investment all day long. One of the questions I always ask and I'm not sure of the answer is, would we have investing Google when they were raising venture capital?
I actually wonder if that's the case. the reason is, a good friend of mine was at another venture firm that passed on Google when they had the chance to invest in it. I said, why'd you pass on Google? This was back right around the same time that it happened so I know it's kind of accurate as opposed to just [inaudible 00:16:51] history. We passed on Google for three reasons. There are three reasons, number one, absolutely no business model. They had absolutely no idea how they were going to make money. Number two, the two most arrogant founders we'd ever met in our lives. Number three, very high premium valuation. Had there been only any two of those three problems, they would've still invested but all three problems together prevented them from doing it.
In venture capital, that's a 20 billion dollar mistake.
Russ R: But, if those are your three rules, you might save enough money on the other 30 billion that you might've thrown away. Well, 30 billion's hard to throw away. That's not a good answer maybe. I don't know. You see my point though.
Marc A: You identify the exact problem and then the problem with the problem; which is, the key characteristic of a venture capitalist is the returns are on a power [inaudible 00:17:54] distribution. Here's the way I think about in the math. There's about 4,000 startups a year that want to raise venture capital. Of those, maybe 400 will get funded by top venture capital firms. Of those, about 15 will be responsible for about 90% of profits for that entire year of companies.
Russ R: That says it all.
Marc A: It's a feast or famine business. If you do invest in Google and you invest in a hundred losers you're a spectacular success. If you don't invest in Google and you invest in a hundred losers, you're a horrible failure. It turned out the only decision that mattered was did you invest in Google or not. Of course, the corollary to it is you never know when Google's going to show up. one of the weird things about Google was they showed up 1999. A couple things about google that would've made it hard at the time ... If you believe there was a bubble in 1999, which a lot of people think that they believe, at least in retrospect. That would've been the last time that you would invest in a money losing company run by arrogant founders at a premium valuation. Yet, that was the one to do. The second thing is, google was like the 36th search engine.
There was this whole thing about first to market and the innovator, the whole thing. There had been six years of search engines before google. Many of whom, at that point, were large public companies and were considered very successful companies; yahoo, Lycos, a whole bunch. Google was late to the game. Turns out they were late to the game with a fundamentally better product. My friend, peter keel, likes to say; he says, it's not first to market that matters. It's last to market. You want to be the last company because you want to be the company that basically [inaudible 00:19:28] all following competition.
Russ R: Yeah, if you can do that.
Marc A: Exactly, but what he says is, that's the key to getting all the investment returns. You want to be the one that basically is so good that it foreclosing all future competition. That often is not the first company. If you think bout the decision path, a lot of people think that if they saw google in 1999 [inaudible 00:19:50] they would've been smart enough to make the investment. The actual decision path that you've have to follow in your head to get all the way to investing in these guys with no business model at that time, in a field that had all types of competition was a good idea, that was a leap. It's no coincidence, by the way, that that investment was made by two of the smartest VC's of all time; John Door and Mike Morris.
That was a leap. That required real foresight, deep thought process, and a risk tolerance that most people simply don't have. So, on our best days, that's what we aspire to be like. We aspire to take the bets that other people won't make. We aspire to go way out on the edge of risk. Another friend of mine, said you'll never been able to work in any other area of financial services investing because every other area of investing is all about reducing risk whereas in venture capital, it's all about increasing risk. The big danger is you're not far enough out of the risk group.
Russ R: How much did Door and Morris put into Google at that point? Do you know?
Marc A: They put about 25 million dollars between the two of them. They bought, I believe 25 million bought somewhere between 15 and 20% of the company. That would have gotten diluted down over time. That would've translated to 10 to 12% of the company today. Google today is worth 340 billion dollars.
Russ R: It's a good deal.
Marc A: 25 million turned into ... if they held stock, which some of them didn't. If they held the stock all the way through, 15 years later, 25 million turns into somewhere between 25 and 50 billion.
Russ R: That'll pay for a lot of mistakes. you're suggesting that what made google better than ... I'm old enough, as many of our listeners are. That's what's great about the internet. I'm not talking about the Korean War now. I'm talking about the 1990's. I remember Lycos and Yahoo as search engines. People still use Yahoo sometimes as a search engine but you're suggesting that Google dominated that market because they had a quote; better product. What was better about it?
Marc A: My friend, Bill [inaudible 00:21:59] says, the key difference is that it had the it works feature. It had the feature where it worked. If you remember doing searches on [inaudible 00:22:11] back in those days on most [inaudible 00:22:12]. You would do the search and you would very often get back useless results. The programmers of those companies worked really hard to get you back quick results but a lot of the times you got back [inaudible 00:22:25]. What had actually happened was users had gotten trained to not expect much from their search engine. They had gotten trained that the results weren't going to be very good. Users didn't do that many searches and when they did, they would spend a lot of time going through it trying to find a lot of the results. [inaudible 00:22:44] or something like that.
The people who really understood this stuff; number one, you could use google early on if you had access to it. You could compare the results and could see that they were better. It was very visible and very visceral. Of course not only better results, it means people are going to be using it a lot more than they would use the other ones. If you were a computer scientist and you had access to these guys and you could talk to them; what you would find out was they had a completely different kind of approach than the previous search engines. They had this innovation that became famous called page rank where they basically had a computer science breakthrough; an actual technological breakthrough in how to do the scoring to get to the best search results.
It turns out those things are discoverable. This is one of the things we find. You had to be very close to the company to realize this. This was very hard to call from the outside. A lot of people who passed on Google, passed on Google without even getting to the point where they learn these details. It turned out the details really mattered because they really made a big difference.
Russ R: Right, the elevator pitch I have a better search engine is not that compelling. The question would be how much better and who cares? Then what? They answered all those questions. Like you say, from the outside that was probably very hard to discern. Especially the, and then what, because the ability to monetize that product was not obvious at all. Even if you [inaudible 00:24:11] the guts of the algorithm.
Marc A: Yeah, and then by the way even then, you would have to invest not knowing how they were going to make money. They didn't have that second part. They [inaudible 00:24:18] that second part out. One of the things I like to say is, we live in one of many parallel universes and we know how this one played out but there [inaudible 00:24:28] universes where [inaudible 00:24:29] played out. There are many other universes in which Google never figured out how to make money. If it crashed and burned, it'd be a cautionary tale today. There are just flips of faith that happened along the way where they were able to figure out this AdWords algorithm to be able to make money.
Had they not figured that out, and a lot of people didn't think they'd be able to figure that out, we would be having a very different conversation today. So, it's one of those things that goes back to what's humbling about what we do. It's limits of knowledge. There are real limits to what you can know. Which, by the way, means that if you're going to operate in this field and if your requirement in investing with something, backing something, or going to work for something is you're going to know for sure [inaudible 00:25:10] succeed; you're never going to do anything. There is no return without the risk.
Russ R: One of my favorite stories is how Fred Smith, supposedly ... It may not be true. After he got turned down for the last time by Chicago banks borrow money to keep FedEx afloat was in the Chicago airport, ready to return to Memphis to tell his employees that it was over; instead took a plane to Reno and put all of his money which included, I think, his sister's trust fund money on red or black or whatever and made enough money to cover his payroll that week. They made it. I always think what would've happened if he hadn't looked up at the board of departures and noticed a flight going to Reno.
If Door and Morris hadn't made that investment, would Larry Page and Sergey [inaudible 00:26:00] being doing something different right now or would they have eventually gotten to where they are? I don't know.
Marc A: The other thing was, nobody will admit today; but that company was very acquirable in the beginning before they had a business model. There were any number of other big technology companies that could have bought Google for small amounts of money. Larry [inaudible 00:26:20] could have gone on to be mid-level engineers at Yahoo. It's a twist of fate [inaudible 00:26:27].
Russ R: A fate worse than death.
Marc A: That's what some people would say.
Russ R: [inaudible 00:26:30] mid-level engineer at Yahoo but compared to what they became, that's all.
Marc A: That's exactly right. Again, that comes back to [inaudible 00:26:41]. How many great entrepreneurs are there who just haven't realized that they had a path that didn't involve [inaudible 00:26:46] company. How many of the business failures that we can all name were one step away from success and they just didn't figure out that one step? Those are the questions we'll never know the answer to. [inaudible 00:27:00]. They cause us to be ... We have this theory at our firm, it's a software term called [inaudible 00:27:06]. We [inaudible 00:27:09] buying decisions; which is to say basically, delay making decisions until as late as possible.
Russ R: Sure.
Marc A: Because you really never know. There's a very high risk that your early decisions are going to be incorrect and you really want to delay making a decision until you get every piece of data you possibly can because the future's so hard to tell.
Russ R: In 2011, you wrote a piece for the Wall Street Journal called Why Software is Eating the World. Explain your argument and why the evidence continues to accumulate that you were right.
Marc A: Yea, the argument has to do with the evolution of the computer industry. The computer industry is one of the industries where it's trickled [inaudible 00:27:46] for a long time. Computers, historically, would get built for the biggest customers which were big companies and big government agencies. They would cost tons of money. Then 10 years, 20 years later, someone would package them up and do a cheaper form factor and make them available to smaller companies. Then 40 years into the computer industry, the PC came along which was the first thing that individuals could buy. 20 years after the PC, the smartphone came along. The smartphone is the big breakthrough because the smartphone is the first computer that is packaged and delivered in a form that everybody on the planet can have one.
The way I think about it is the two giant twin dominant stories of our era are; number one, the enormous rise of the developing world and the introduction of billions of people into what we would consider to be the modern economy. Then in parallel, intertwined with that, is the smartphone revolution which is everybody on the planet getting what is the equivalent of a super computer, from 20 years ago, in the form of a smartphone in their pocket.
Russ R: That they have with them all the time. Everybody has it and has it all the time.
Marc A: Have them all the time and they're all on the network. They're all connected. They're all on the internet. We've gone from a world where most people didn't have computers. The PC only ever got to about a billion people out of seven billion people. The smartphone is going to get [inaudible 00:29:12] billion. Two big things happening right now in real time, one is that in both India and Pakistan, the price of smartphones now has plummeted to $35. All of a sudden, it's a $35 consumer purchase which is within reach to a large number of people. Second is, even in the poorest parts of the world, [inaudible 00:29:32] considerable preferences. Even the poorest people in the world will choose smartphone and internet access; even over indoor plumbing and electricity given the choice.
You hear about people working in the field in the most poverty stricken parts of the world who would not [inaudible 00:29:49]. You really do have this universal computer for the first time. Everybody's going to have one. These things are shipping in the billions now. By the end of the decade, everybody on the planet's going to have one of these things. Everybody's going to have a computer. Everybody's going to be on the internet. That's a new world. That's a world that we never lived in before. We have no idea what that world is like. It's brand new.
One of the thing that you know is that, all of a sudden, [inaudible 00:30:15] conceivable way to take a product or a service. If you could take a conceivable way to deliver it through software, you can actually do that. I'll give you an example. Let's just talk about banking as one example. Which is, historically the idea of having an online only bank that was only delivered through software would have been considered lunacy because most people don't have computers so you need branches. You need tellers. You need ATM machines. You need this big physical footprint to build a bank. Today you can just make the simplifying assumption. You could say I'm only going to make the bank available online. It's only going to be available for people through their smartphones. It's just going to be software. There won't be anything else to it other than software.
All of a sudden, you could have a bank with an [inaudible 00:30:55] of seven billion customers entirely in software without any of the physical overhead of how today's existing banks operate. This is the software eating the world piece. We now, for the first time can basically go field by field, category by category, industry by industry, product by product; and we can say what would they be like if they were all software? Then, entrepreneurs in virtually every field we're talking about are attempting to do that. There are entrepreneurs attempting to do software only financial services, software only education, software healthcare, obviously then the media industry is being transformed with software, e-commerce. Retail is being transformed with software.
This is sort of where I disagree so much with people who are worried about sort of innovation slowing down. I think the opposite is happening. I think innovation is accelerating because the minute you can take something that was not software and make it software, you can change it much faster into the future. It's much easier to change software than it is to change something with the big, physical real world footprint.
Russ R: If only we could stay in digital hotels, because the biggest cost of hotels is they got to replace the furniture every once in a while. For so many other things, the furniture's digital so it's a piece of cake.
Marc A: Exactly, what's happened is all of a sudden, software professionals, not just in the US; all over the world, software professionals, software entrepreneurs are looking at industries that have not been tech driven, industries that have not been in that space. They're looking at those industries and they're saying, now is the time to build this software bank. Now is the time to build this software school. There are entrepreneurs all over the plant that have figured this out. They're going straight for it. Of course, our job as venture capitalists is to fund them. What are the consequences of this? Many things; number one, consumer welfare is on the rise way more than I see people are willing to give it credit for. I think that the universe of opportunity that opens up to you once you have a smartphone in terms of your ability to get to all these services, get information, get access to global markets, get access to education. From a consumer welfare standpoint, this is nirvana.
This is like, everybody has the magic box to which they could get access to all this software. That's amazing relative to all the physical limitations the way things used to work. Consumer welfare is on the rise very fast and on a much broader footing than people believe. Even in the poorest, rural villagers now have access to resources that the national security agency didn't have access to 10 years ago. That's an enormous consumer welfare change. Rate of evolution increases in a lot of industries because software can mutate much faster. Prices can come down very fast in a lot of industries. One of the things that I think is very interesting economically. I think price deflation across the economy is a much bigger factor than people think. you take a product that use to hardware and you make it software or it used to be a retail store and you take out huge parts of costs, which means prices fall. The final thing is entrepreneurship is on the rise, because everybody in the world with the right software, which is a large number of people, can now be an entrepreneur if they want to and can go after these opportunities in many different fields.
Russ R: So you've been very enthusiastic about Bitcoin. We recently did an episode about Bitcoin. You've actually compared Bitcoin to innovation such as the personal computer and the internet. Very bold claim; what's the source for that enthusiasm?
Marc A: This is not a claim that I have made about anything else in the last 20 years. This is the first time I've said this. That indicates the depth of seriousness which I take it. Bitcoin and the ideas underneath Bitcoin ... Bitcoin is this broad topic which in the computer science world is called crypto currency. [inaudible 00:34:55] this area of R&D that's been going on for 20 years. One of the things about Bitcoin that's important to understand is it's not just an overnight thing that somebody just dreamed up. [inaudible 00:35:02] 20 years of really hard work on the part of a lot of brilliant computer scientists. They finally catalyzed in the form of this Bitcoin thing. There's very deep, intellectual background behind Bitcoin.
The big breakthrough that's underneath of Bitcoin that is called distributed trust. The idea is take seven billion people, put them all online with their smartphones. In theory, you have the ability to do business with anybody on the planet but how do you know who to trust, right? How do you do trusted transactions? How do you send money from point A to point B, right? Knowing who's sending it, knowing who's receiving it, knowing that the money is digital money, that it's not being copied along the way which is called the double spending problem. How do you do a transaction with digital money in a way where everybody else around you is able to verify that that transaction actually happened?
People can't say I was frauded or I never got the money or whatever. How can you do that in a way that doesn't require centralized institutions? How do you do that where it doesn't require a bank, credit card company, or a payment processor?
Russ R: or the Department of Justice; the police. it'd be great if you could avoid the police.
Marc A: That's exactly right. Let's talk about this for a second, the way contracts work in the US is you sign a piece of paper. Well, what happens if somebody forges a signature. Well, you call the cops. You take it to court. You have a lawsuit. You go through all this stuff. You're exactly right. You either are working with employees who are working with contract law and with courts. You immediately fall back on the centralized institutions. You hope they get you the right outcome. What if you had a digital contract that was unforgeable? Once you signed it, that was it. It was provable after you signed it and nobody else could've signed it.
Russ R: What you signed you had to keep. That there was no uncertainty or virtually no uncertainty that you'd keep the promise that your signature represented; because that's the other piece that's always uncertain.
Marc A: The first thing is just being able to interact. There are consequences to many of these things, but the first thing is to be able to interact. Let me give you a basic concept. [inaudible 00:37:15] cost of ownership; who owns what? Who has title to what? In the west, we take it for granted that we've got title. Real estate titles, we've got. We've got title agencies. We've got contract law around titles and all the rest of that. Ultimately, there's enforcement if somebody tries to squat at somebody else's house. It's unfortunate but we have very clear ways of determining what's what. Of course, ...
Russ R: ... they're very expensive those titles. When you buy a house, a frightening large amount of money goes to prove that you're actually buying and selling the house that you both have in mind. There's a big set of institutions around it.
Marc A: That's exactly right. In much of the rest of the world, in much of the developing world, there isn't clear title. [inaudible 00:37:55] written about his a lot. This used to be a problem in India. I don't know what the number is now, but there's some large percentage of real estate in India and it's just not clear who owns it. This is a generalized problem. If you can't have land, do you have the basis for capital and all the things economists talk about. Then you say, we need to have a system to which everybody in the world can establish ownership. Then have consensus too of who owns what and have a way to transfer ownership from one person to another in a way that can be validated and can be trusted without having to recreate these giant centralized institutions; which might either take decades to build in systems where the government's not strong enough to do that yet or might just be [inaudible 00:38:39] expensive. It may not be possible what we might consider to be a modern title system for land in the world where incomes are much lower.
It might be part of the development trap. It might just not be cost effective [inaudible 00:38:52] required to let people develop economically. What if we could just do that digitally? What if we could just do that on the internet for free? Well, we can send email back and forth for free. I can send messages back and forth. Why can't I send title back and forth? There were a set of breakthroughs that had to happen around trust and photography that had to happen that in fact happened over the last five or 10 years. The key breakthrough was the Bitcoin breakthrough by this anonymous inventor, [inaudible 00:39:22] who came up with this idea of the block chain, this trust model for establishing who owns what, who controls what, who has committed to what at different points in time. That's a really big breakthrough. You can think about that breakthrough. Bitcoin, people think of it as digital money. It is digital money but its deeper than that.
It's potential also digital contracts. It's potentially digital title, digital ownership, digital keys, digital assets, unique media files, which has always been a big problem on the internet, single copies of media could be done this way. Ultimately, digital stocks, bonds, loans, insurance contracts; you can kind of see where I'm heading with this. The distributed trust breakthrough is a a wedge that technology has now made possible. Now what's going to happen and is happening right now is hundreds and soon thousands of entrepreneurs starting companies to do software based contracts, software based keys, software based signatures, software based title, all these different categories using this underlying technology. That's what going to happen in the next five years and that's what we're funding.
Russ R: What could derail that? What do you think affects the viability, not of Bitcoin, but of this crypto currency distributed trust breakthrough. Is that breakthrough over? Is it over? we're done. It's solved. We don't have to worry about it anymore or are there things that you think are still uncertain about it?
Marc A: At the highest level it's hard to see it stopping. The reason it's hard to see it stopping is it's just math. It's just math. It's just bits. It's like stopping [inaudible 00:41:11]. For example, [inaudible 00:41:17] Turkey's going through this right now where something happens politically and they decide they don't want bad behavior on twitter or YouTube so they ban twitter. The next thing you know is hackers in that country find three different ways for people to still be on twitter or all the behavior on YouTube then they ban YouTube. Increasingly, they look like they're doing terrible things to their citizens. Then at some point they just turn the entire internet off. The problem is, if you shut the internet off you tend to drive everybody out into the streets. That's the last thing you want if you're trying to prevent a revolution.
If you're north Korea and you can prevent people from using the internet, you can stop all this stuff, but I don't know how modern countries can shut off the internet. I think we're past that point. Short of shutting off the internet, you'd have to take a very aggressive path to intercept these bits. It's like trying to prevent people from talking. You could try except there are an enormous number of people who want to make sure that the free [inaudible 00:42:18] of information doesn't stop. So ...
Russ R: ... you're talking about the government response, but what about the private hacking response? Do you think the ability to hack into someone's Bitcoin wallet, for example, is going to be a problem in the future or is that quote solved? Is that problem of duplicate money? I forget the technical term you used, the digital copy. I pay somebody and then they can buy two different rugs with the same coins. Are those technical problems over or solved or do you think there are other risks involved technologically?
Marc A: Those problems are solved in theory. This is a complicated question so I'll give a fairly complicated answer. Those problems are solved in theory. The reason we know they're solved in theory is because, like I said, this isn't just an overnight thing. This is something people have been working at for over 20 years. Every step of the way, they've been trying to break it at the same time they've been trying to build it. Bitcoin itself has been out for over five years and many of the best hackers in the world have spent five years trying to break it. There's been a huge financial incentive to break it. It has not been broken. Can I prove beyond a shadow of a doubt that it's never gets broken? No.
Do I know that many of the best hackers in the world have been trying hard for years and have been unable to break it? Yes. In the real world, how do you know that something's secure? At some point, you know something's secure because people have tried to break it and failed. How do I know vaults are secure? Well, people haven't been able to open it. Theoretically, it's in a very good place. [inaudible 00:43:51] by the way, we feel it's in a good enough place to [inaudible 00:43:54]. We watch. We're deep in this world. We're deep in the security world. We know a lot of hackers so we track this stuff very carefully and nobody's making any progress cracking it; so far so good.
Every year that goes by, you get more and more assurance. You have more and more evidence that it hasn't been broken. That's in theory and that's important. It's important that that be the case in theory.
Russ R: Counts.
Marc A: Then there's a practice, it has to get implemented in the real world to get used by normal people. There you get into things like your Bitcoin account is going to be protected by a password and if you pick the wrong password, and your password gets easily guessed by a hacker, then somebody could take all your Bitcoins. That's still the case, but in a sense, of course that's the case because that's the case with everything. If you do that with your email, people can read all your email. If you leave your car unlocked with the keys in the ignition in the middle of the street, come back in two hours, it probably won't still be there.
It sort of reducing down to this broader question of making sure that digital systems that we use are generally secure. There's all kinds of work into that throughout the industry using new kinds of authentication methods; using fingerprints, retinal scanners, all these different approaches to make this more secure. At some point, you do have the types of product issues, but it's certainly every bit as secure as an online banking system as an example. Well put it this way, it's more secure than [inaudible 00:45:21] online bank. It's more secure than the Obamacare website. It's more secure than most commercial websites. It's beyond the point where you usually worry about this in terms of building a business. You just need to work with your users to make sure they do the responsible things.
Russ R: If Bitcoin were widely used and accepted and it's on the road to being that payment system, would you be comfortable with large chunks of your wealth in your Bitcoin wallet?
Marc A: Yeah, definitely. [inaudible 00:45:53] very clever about Bitcoin. Money doesn't actually have to be online. One of the [inaudible 00:45:59] of Bitcoin is, you don't actually have the money in the wallet. Bitcoin is numbers. You can actually store your Bitcoin in a [inaudible 00:46:04] and print your Bitcoin on paper and store it in a safe deposit box. There's a variety, it's called cold storage. There's a variety of cold storage methods. There are actually companies that are going into this business of Bitcoin cold storage. It's almost like a safe deposit box business. The safe's to store your family's jewelry in a safe deposit box in a bank. Well, yeah, just buy the vault.
Russ R: More or less.
Marc A: By the way, the other thing is insurance. You can insure your jewelry. There will be a variety of different insurance methods through Bitcoin. Bitcoin wallet companies themselves are going to have insurance and then you yourself will be able to buy insurance against these kinds of risks. It'll be like anything in the real world. It'll be culmination of you don't carry all your money in your back pocket with you in the form of cash every day. You store some of it in a bank account. You maybe store some of it in gold in a safe deposit box. Maybe you store[inaudible 00:47:00]. Maybe if you're really paranoid you bury it under your front porch. It's kind of the same thing with Bitcoin. You put it on paper and bury it in the proverbial [inaudible 00:47:09] under your front porch if you want to.
Russ R: Well, let's shift gears. Let's talk about the news business. You've written about that recently. Most journalists are very pessimistic or worried about journalism. they're the people who work for newspapers that I talk to. Other journalists are less worried. You're optimistic about the future of journalism, why?
Marc A: Yea, I take a very different perspective on this. It's a perspective based entirely on business and economics. I'll kind of describe why. If you study the history, the news business is actually an old business. It's been around for about 500 years. [inaudible 00:47:45] through the first 450 years of the newspaper business, it was a brutally competitive business. one of my favorite books is a book called, Infamous Scribblers which was the history of the news business in colonial America. It's sort of a slice in time thing of what it was like to be in the newspaper business in colonial Philadelphia, 1770. Which was like when Ben Franklin was [inaudible 00:48:10] business.
What you realize is it was a brutally competitive business. Any given city would have 15 different newspapers. They would all have a different subjective point of view. Some of them would be political attacks [inaudible 00:48:25]. People were writing under pseudonyms. I think Franklin himself had a dozen different pseudonyms he would write under to [inaudible 00:48:33] different agendas. It was kind of this rolling free for all of activity, information, news, efficacy, and politics. It was kind of this zoo.
It ran that way for part of 50 years. By the way, it created very large empires in the process. The first empire [inaudible 00:48:49] were very big businesses based on that kind of approach. What happened, I think was, especially in the US after World War two, the news business consolidated into oligopoly structure. In particular in the newspaper biz, the consolidated into local monopolies. This was due to scale economics per city. In any given major city in the US, over the course of a couple decades, you went from 15 newspapers to five newspapers ultimately down to one newspaper. You had the Chicago Tribune in Chicago. You had the LA Times in LA. It was only the largest cities would have a few newspapers, but even there you'd have a dominate one like the New York Times.
Then 50 years passed where, if you were a journalist, by definition you were working for a monopoly. Those of us that have worked with companies over the years, there's a big difference in interacting or working for a monopoly versus working with a company that's in a competitive business. It's the difference between dealing with a company that has to compete everyday versus a company that doesn't have to compete. Every monopoly has the same model, which is we don't care because we don't have to.
You have these companies, they were business like any other and they had problems. They can act like monopolies. They had generations of managers; two generations, three generations passed. At a certain point, you only had people going to work for these companies who wanted to go work for a monopoly. By the way, it worked. These businesses got better and better. The newspaper business didn't actually start to collapse until after 2005. [inaudible 00:50:30] grew tremendously even though the initial phases of the internet boom. They had very big margins so the business executives certainly made a ton of money; [inaudible 00:50:42] expense account, big fancy buildings, lots of long lunches, the whole thing.
Then what happens of course, the distribution technology changes. The reason these all centralized in monopolies is because of the scale of economics distribution; having the printing press and the [inaudible 00:50:56] trucks that would actually get the newspaper out. You had a scale advantage if you were the sole provider. The internet striped the monopoly status on the distribution up from all these companies.
Russ R: Then they took the revenue stream away from them too. You had to [inaudible 00:51:11] in the classifieds.
Marc A: They did something very specific there. It wasn't so much that you had new online newspapers that did the same thing that the offline newspaper did. It was the product got unbundled. In the old days, people had this sort of romantic view of the newspaper was like this [inaudible 00:51:29] of democracy that the journalists liked to talk about. The reality was most of the newspaper was the grocery store ads, the car dealership ads, the want ads, the classified ads, the sports scores, the stock quotes, the funny pages, the horoscopes, Dear Abby, entertainment news, TV listings. It was this bundle of information and it made sense for all that information to be in a bundle. It made sense for all the revenue streams associated with all those components to be in a bundle because of the cost of distribution.
It didn't make any sense to have 20 different specialized newspapers covering all those different topics. It made sense to have one everything was funneled in. You put that on the internet then all of a sudden, to your point, Craigslist takes all the classifieds. Yahoo movies takes all the movie listings. Yahoo finance takes all the stock quotes. ESPN.com takes all the sports scores; and on and on. The product got unbundled. The distribution monopoly fell away. The competitive battle started immediately.
The competitive battle has been really fascinating. You had newspapers competing with newspapers who didn't used to compete. The New York Times and the LA Times never used to compete and today they do because users on the web can go to either one equally. The New York Times would have never viewed CNN as a competitor in the old days, but in the new world, New York Times.com or CNN.com; two different websites. You have competition across media channels. You have this great unbundling taking place and the revenue vanishing. It felt like the perfect storm.
If you're inside one of these media companies searching for a news publisher, it's like the entire thing is collapsing. There's a lot of truth to that, but the root cause is you were used to being a monopoly. You weren't used to competing. What's the answer to this entire thing going to be? It's going to be to compete. It's going to be a take a stance of every business in the world that actually has to compete for a living and figure out fundamental questions. What's my differentiation? What's my competitive advantage? What's the appropriate cost structure? Who are my customers? What do they want? Where am I unique? Where am I not unique?
It's a time when we need proprietors for news organizations that are like the proprietors from the [inaudible 00:53:48]. We need proprietors, owners, and managers who full on capitalists, full on aggressive business people who are very good at spicing up markets and identifying revenue opportunities, very good at rationalizing cost structures and all the things that you have to do in a normal business. Of course, this causes the journalists to freak out. They're like, my God. The whole point of objective journalism is you have this separation church and state and the business isn't supposed to affect the journalism. How are you going to fit all the business people [inaudible 00:54:15]? It'll ruin the whole thing.
I think the exact opposite answer, which is the way to guarantee high quality journalism is to have it be a business. If you don't have a successful business, you have a charity or in worst case scenario, bankruptcy. You have to solve the problem as a business. The good news is that the market is much larger than it used to be. This is something we think of a lot about in tech. We think that in venture capital. We think about what's the market size of what you're going after. The internet has caused you all these problems but the internet has given you a gift which is a much larger [inaudible 00:54:48] market.
Even the New York Times, historically its total market size for people it could distribute a physical newspaper to in New York. Today its [inaudible 00:54:58] market is the entire world. Hundreds and hundreds of million people around the world who need to know what's going on. These businesses need to be reconstituted around market segments that have a need for differentiated products, but many of those market segments are [inaudible 00:55:13] than it used to be.
Business and finance news will be the leader in the recovery because the rise of the number of people globally who need business and finance news is growing very fast. Of course, those people have expense accounts. They can easily afford to pay for information that valuable to them. I think the Wall Street Journal from here is going to do really well. I think that the Financial Times probably [inaudible 00:55:37] under a new owner, at some point is going to do really well. Then there's a variety of new startups in business and finance news that are growing very fast that are going to do very well. General news and other categories of news will follow.
Russ R: It's an inspiring story. I happen to agree with it. I think you're right. There's a lot of romance about being a journalist that I think isn't even true. The objectivity part and supporting democracy, we do need journalism to ... is journalism as useful for exposing tyranny and corruption? I think it will continue to do that. I'm like you, an optimist. How about ... We're low on time but I want to hear your thoughts about two areas that everybody cares about which are healthcare and education. We talked a lot about [inaudible 00:56:22] on this program. Do you think they're real or overblown? Are they going to revolutionize education in the ways that we've been talking about? Is software going to change healthcare in the next five to 10 years? I think it will but I'm curious what you think. You're smarter than I am.
Marc A: [inaudible 00:56:39] I think the answer to education is a broader question. I think the real question is a broader question. The existing education, especially at the college university level, actually works pretty well. I always tell people if you can go to Stanford, go to Stanford. My friend Peter [inaudible 00:57:00] was like, don't go to college. Go to Stanford, go to Harvard. I went to the University of Illinois; great school. Go to Purdue. Go to university of Washington. These are great. By the way, there are also great private high schools or public high schools that are very good. It's great, if you can get there that's great.
There's a very large number of people in the world and a large number of kids coming up who are never going to go to any of those places. We know that because the numbers just don't work. Run the math on how many people are there in the world today ages five and below who are going to hit the high school and college market in terms of age, five, 10, 15 years out. What I'm saying we know for sure is they're not going to be on a physical ivy league campus or a physical university state school campus. They're either not going to get education, which would be a disaster for the entire planet or they're going to have to get educated in a different way. The only way to scale the education system to meet the needs of everybody on the planet that needs to be education is through software.
It's a ridiculous cost prohibitive exercise to try to figure out how to replicate the campus model for the number of kids that's going to ned to be educated. You were asking about it over a 20 year period, we have to solve this problem of software. A large part of the future of the planet depends on solving this problem of software. It becomes a moral issue quite quickly. We have to figure out how to do that. The argument that's almost a complete waste of time is would you rather have a muke or boat sitting in the lecture hall at Harvard. That [inaudible 00:58:38].
Russ R: I couldn't agree more. People say, face to face is so much better. I said, yeah, if you have a good teacher. How many people have a good teacher? It's much better to have a great teacher on the internet than an awful teacher face to face. There's no comparison.
Marc A: Or no teacher face to face.
Russ R: They don't show up in certain parts of the world.
Marc A: Exactly, that's the big thing that has to be tackled. The other thing is I think there's two kinds of students. The thing I hear from American educators, especially at top-tier institutions is, you don't understand. You're naïve. Kids don't want to get educated. They only got to class reluctantly and if you're not on their tails all the time, they won't even do their homework. They're out partying all the time. To me that's like a sense of people who are [inaudible 00:59:22] lazy. There's a much larger number of people in the world who do not have that problem and whose parents do not have that problem. The difference between educated and not educated is fundamental differences in quality of life. It's absolutely fundamental and everybody knows it. There's no [inaudible 00:59:38].
This whole thing where people lack motivation for an education. Yeah, rich [inaudible 00:59:44] might like motivation but like everybody else in the world he's got to have the motivation in spades [inaudible 00:59:49].
Russ R: Yeah, they're kind of [inaudible 00:59:51].
Marc A: We collectively have to figure out. This is a challenge. Companies play a role here. Governments play a role. Non-profits are playing a really big role. [inaudible 00:59:59] fan. A lot of people are. What he wants to do [inaudible 01:00:07] is going to be just tremendous. By the way, teachers play a huge role. One of the huge opportunities, you're obviously an example of this is, but the best teachers in the world are going to be able to have a much bigger impact as a result of this.
Russ R: Make more money.
Marc A: You need to make more money and [inaudible 01:00:19]. Exactly, this is one of those things where a lot of the best teachers are going to get really fired up by what this means for them and the impact this can have on the world. They're going to play a big role. That plays out. [inaudible 01:00:33]
Russ R: Healthcare.
Marc A: Healthcare, the challenge in healthcare is also the opportunity in healthcare which is to fix the incentives. The pessimistic view is that it's sort of an unfixable situation because the incentives are badly aligned and if you're not paying for your own healthcare and don't understand what you're buying, then you don’t care what it costs. That's when you get this out of control thing. Even if in theory you could break the cost disease and bring productivity [inaudible 01:00:59]. Nobody has the right incentives to do it so it'll never happen. The two really positive things that are happening right now in the US that we're looking at very carefully are; number one, this dramatic rise in high deductible health insurance. A lot of this is being sparked as a consequence of ... This may be one of the big positive benefits of the Affordable Care Act. A lot of people now are going to be operating under healthcare plans that have two thousand dollar deductibles. All of a sudden, a pretty high percentage of the actual health events that you deal with, you're going to be paying significant amount of money out of pocket.
Russ R:That'll change behavior.
Marc A: Exactly, right. All of a sudden, it goes right back to the smartphone. If you give people the right kind of smartphone app, where instead just going to whatever doctor, paying for it and getting like you would in the old days, without of control costs; what if you could get the answer through your smartphone or could find it cheaper? For example, one of the big things in healthcare is a lot of people go to see doctors [inaudible 01:02:00]. Today's healthcare system, you have no reason to not go to the doctor. In the future, it's an easily diagnosed and resolved problem that a nurse can help you with. You'll be able to just do that.
By the way, you'll be able to do things like [inaudible 01:02:12] scoring on the providers so you could find a really good nurse or nurse practitioner to deal with whatever your minor problem is and pay a lot less money. That's one big opportunity. The other giant underestimated thing that's happening is more big employers are self-insuring their employee [inaudible 01:02:29]. These really big multi-nationals, they're actually footing the bill for their employee's insurance more and more. There you have very professional management teams, finance departments, HR departments that all of a sudden have a huge incentive to drive down costs. They can't drive down costs in a way that's going to damage their employees because that would be counter-productive.
They'll drive away the talent, but they have a big incentive to optimize [inaudible 01:02:54] straight to their corporate bottom line. We're seeing big companies as much more interesting adopters of new technology on behalf of their employees. There's a big opportunity there. Obviously this is a big one. This will take a long time, but you can envision a culmination of changing economic incentives combined with new technology, combined with increased role of individuals and big employers combined with technological advances. it certainly gives you wedges to come in and start having an impact.
I would say there's reason for mild optimism. Then hopefully over the course of the next 10 to 20 years we can start going after some of the really big [inaudible 01:03:36].
Russ R: Last question, are you worried about anything remotely like the singularity or the fear that technology is going to put everybody out of work except for the handful of people who can create the new software that's going to make our lives better? What's your feeling about that? Optimist or pessimist?
Marc A: Very optimistic as follows; we live in such a cynical and pessimistic time that nobody wants to think about the other side of this so I'll put it out there. We are, with the smartphone, with the internet, with google, with open source [inaudible 01:04:14], with all these things we're talking about; we're putting unbelievable tools in the hands of everybody on the planet to be able to do things in fundamentally better ways, get access to information, communicate, access global market, access up to date pricing for goods and services, open online storefronts, educate their kids, on and on. It's like the Swiss army knife of all time. You give somebody a smartphone all of a sudden the entire world opens up.
It's hard for me to believe where another three, four or five billion people are going to be able to be able to purchase a [inaudible 01:04:54] in a market economy for the first time ever in a fundamental integrated way and that's not going to lead to enormous economic growth. How could that possibly be the case? We basically have to believe that all these people are going to have all these tools, all these opportunities and they're going to do nothing with it, right? Which seems to me [inaudible 01:05:17]
Russ R: All the good jobs will be taken.
Marc A: Exactly.
Russ R: There won't be any news. Somehow there won't be any new ones.
Marc A: Right at the point of maximum opportunity, co-creation will come to a complete halt. I'm on the other side of it. I think we are on the verge of what could be globally a much faster rate of growth than we're anticipating because so much more talent is going to get unleashed. There's so many people all over the world who have not been able to participate in the economy at their full potential, who are now going to have the opportunity increasingly to do so. That means potentially a flowering of human creativity, innovation, entrepreneurship. You've got global spread of the entrepreneurship culture. You've got just what's happening in China alone with entrepreneurship is spectacular. You've got the global spread of venture capital. You've got the creation of all these new businesses, industries.
You've got fundamental breakthroughs happening in science and research; you've got all of these incredibly powerful changes happening and really human potential being unleashed. I think, the big questions are around institutions and leadership; which is, are the governments, the government's systems, and the economic systems going to be setup, reformed, structured in the right way so people can actually take advantage of these tools?
Are you going to be able to raise money for your company? Are you going to be able to keep money for yourself and your company or is it all going to be taken in taxes? Are you going to be able to hire employees or are you going to have European style labor laws everywhere where you can never fire anybody so you can't hire anybody? The sort of really big policy questions become even more important because they affect a lot more people. Of course, there are many countries in the world that have extremely ambitious and hungry, talented populations where the governments systems are still, at best, described as [inaudible 01:07:12]. They need to be reformed.
We need modern market-based economies. We need modern democracies. We need those ideas to spread as fast as possible. I think it will be a time of great political turmoil is probably my prediction. I think the number of government changes, government turnover, literally revolutions in the streets seems like it's on its way up. People are becoming less willing to accept a bad state of existence because they're becoming more aware of the alternative. I think they'll be a lot more political change, pressure, and strife. I think the stakes are going up.
Russ R: My guest today has been Marc Andreessen. Marc, thanks for being part of Econtalk.
Marc A: Great, thank you so much Russ.
Russ R: This is Econtalk, part of the Library of Economics and Liberty. For more Econtalk go to Econtalk.org where you can also comment on today's podcast and find links and readings related to today's conversation. The sound engineer for Econtalk is Rich [inaudible 01:08:17]. I'm your host Russ Roberts, thanks for listening; talk to you on Monday.
Coming soon: Book recommendations from Dustin Moskovitz. Follow me on Twitter for first notification!
Please excuse the typos - I pay for these transcriptions and haven't combed through the errors on this one.
Hacking AngelList syndicates with cofounder Naval Ravikant
Speaker 1: Today’s episode of this week in Startups is brought to you by audible.com, a leading provider of spoken audio information and entertainment. Listen to audiobooks whenever and wherever you want. Visit audible.com/twist for your free audiobook. By snapterms, online legal protection made simple. Visit snapterms.com and enter the code TWIST to receive 10% off. By Amazon web services, get the resources you need to easily get started with AWS activate, a new global program for Startups including AWS credits, training, developer support, a Startup community forum and special offers from third parties. Learn more and sign up at AWS.amazon.com/activate.
Jason: Hey everybody! Hey everybody! It’s Jason Calacanis and this is this week in Startups to show what we learn how to build companies and board that we have a doozy for you today. This is my exclusive interview with Naval from AngelList. We sat down at the Launch Hackathon and we had an incredible conversation about what matters in terms of building companies, raising money for those companies from Angel investors and the future of Angel investing.
Now that we have this crowdfunding syndicates, it’s an incredible episode. It is required listening for any entrepreneur in 2013 and 2014. Get in there and listen to this amazing episode.
Hey what a great amazing episode. This was with Naval. Stay tuned for our friend James Altucher on the audible author series coming right up.
Please join me and welcome Naval Ravikant. All right. Welcome Naval.
Naval: Thanks. Thanks for having me.
Jason: I appreciate you’re giving up your Sunday to come to talk to all of us.
Naval: I think these people are doing the real work. I’m just sitting here and talking.
Jason: They are. They are.
Naval: Which I think it’s still cold.
Jason: You started as an entrepreneur. You can pull up slide number one here. You had two companies I think of note. Tell us about them.
Naval: Yeah. Epinions was a company started in the first [bubble 00:02:31], the .com boom and it was a product review engine. It went public as far as shopping.com. Then vast.com is a white label classified big data analyst company. It’s still private, doing pretty well.
Jason: Now, those companies, what was your role of them, founder, developer?
Naval: Yeah, founder and CEO.
Jason: Founder and CEO.
Jason: When you look at that first wave and that crazy bubble with live through, compare it to what some people are calling Bubble 2.0 or at least frothy and it tells we use it.
Naval: Yeah, it’s frothy for sure in that it’s completely different. First Bubble, our first wave you could say there were tons and tons of money being poured into silly projects. You get literally a hundred million dollars behind something that was just made no sense and had no chance of ever making that money back.
Things are very, very different now. The markets are much larger. Everyone is on the internet. Everyone’s got on their pocket. We had broadband. We have 4G. People are much more comfortable using things, paying for things. There’s API stock structure. Scale the opportunities play a hundred experts. It was back then.
It’s much, much cheaper to build a company. Epinions, we spent five million dollars just building the first original Web site because there was no open source softwares. You use [Solaris 00:03:41] servers and Sun and pay some licenses in oracle databases and host our own data center because data centers weren’t established and key one lines and there was no CSS. We had to write everything in tables from scratch. No deploy scripts.
You just have to build everything from scratch. What used to cost five million bucks to do, eventually cost five hundred thousand dollars to do and that cost fifty thousand dollars to do. Now, we’re in a completely different phase. We’re in [inaudible 00:04:06] explosion sort of scenario where there’s tons and tons of small startups and more and more being started every day. They’re very cheap and you need to start.
That said these are still winner take all markets with huge network effect. You get this incredibly outsized one of hundred companies winning and they win huge. The nature of the market has fundamentally changed. The distribution curve is very different. I do think it’s actually a much [inaudible 00:04:31] time to be investing as long you’re evaluation is sensitive and you stand the portfolio dynamics of how few companies actually succeed. All your returns are going to come from being in one or two at those winners.
Jason: A lot of talk of a Series A crunch and being very hard for people to get that series A. Is it true or not?
Naval: It is true. The numbers of companies is split, gone up by a factor of a hundred at this point or rounds to that. The number of venture capital is not increased. What venture capital now does is basically growth capital. Series A has really done by the accelerators and the Hackathons. Series B is done by the Angels, super Angel seed funds. Series C and up is where the VC start.
I think I believe there is a fundamental misunderstanding still working its way through the system about how to build a company, which is people think the right way to do it is to kind of raise a million, a million and a half from some Angels then prove something out then go to VCs and try and raise series A which is a classic model and that fails.
Really the right answer is fifty thousand dollars build your product, launch it, accelerator style. Five hundred thousand dollars prove something big. Then maybe your next round is a million, million and a half and get to profitability because it doesn’t take that much to build a company anymore.
Those of you who are trying to hire like 30 or 40 people into your company, you’re setting yourself up for a very, very tight funnel that you you just squeeze through on the growth capital side with the VCs.
Jason: Why is this go … I will go small or go lean strategy in your opinion being probably the most knowledgeable person of the angel space? Why in your opinion is this the best reason?
Naval: Small teams are innovative. They move faster. They get more stuff done. Most of AngelList was built by six people. Even today after ramping up quite extensive last year, we’re only still 15 people. It doesn’t take a whole lot of people. Look at Instagram four people, Snapchat, 20 people. To the extent that you’re hiring a lot of people especially in software, that’s most your ego talking. Unless you’ve got a billion dollar opportunity tiger by the tail and you’re clearly winning and people are throwing money at you, do not scale up. Pretty much you’re scaling of a debt at startups.
Naval: Keep in mind it’s in everyone who’s writing checks incentive to tell you to scale prematurely.
Jason: Why is that?
Naval: Because they’re going to bill you out. They’re the ones who are going to show up and then negotiate the terms to bill you out. It keeps them relevant and powerful. The part of the reason why AngelList exists and Angels exist is because the cost of building companies come down so much that the number of people who qualifies investors are some exploded. They couldn’t make a difference before. A twenty-five thousand dollar check would make no difference to you in 1999.
Jason: Yeah, but a twenty-five thousand dollar check can build a proof a consent.
Jason: Today, right.
Naval: But not through early.
Jason: The VCs or some of the bigger investors, they suddenly realized they might miss the chance to own 30% of your company?
Naval: They do it now but they do it through growth capital. They wait until you’re big and growing then they come in and they bid. The venture capital game at this point has become about branding yourself and to get good deals that are obviously hot deals.
Jason: Let’s talk a little bit about AngelList in the early days. Maybe we pull up slide number two for a quick second here. You started doing Venture Hacks and you even started writing this great blog post unpacking the venture capital industry back in 2007. People were not happy about you doing this, were they?
Naval: Yeah, it was mixed results but at the end of the day, information knowledge wants to be free. [Nivy 00:07:59] and I used to sit around and talk a lot about how opaque term sheets were and how as a founder. This is basically pre-accelerator, pre-blogging. You go in and you get a term sheet from a venture capitalist and we have no idea what it means. You basically sign … The series A was the most important. You basically sign your life away then you find out five years later what it meant.
There was a game theory to this whole thing, to all the different terms in how to play out [V2 rights 00:08:22] and all that stuff. Even your lawyer, although your lawyer you could trust as your lawyer, nevertheless they spent more time serving to VCs and working with an ecosystem. You were the single move player in the game. The VC is the multi move player in the game theory sense. They’re always going to get the benefit of the doubt.
They created all these things that were standard and then your lawyer would say, “Oh, that’s standard. That’s standard. That’s standard. That’s standard.” Pretty soon you find out, that evaluation is temporary but the control is forever. You’ve given a control every company. It’s no longer your company. You’re the entrepreneur but you’re working for somebody else. I think it’s a horrific situation that still many, many entrepreneurs I know it.
I just had another coffee meeting yesterday … I’m sorry, Friday with an entrepreneur who was doing great or his company is flying high. VC took over, merging another company, kicking him out. It still happens. You got to be very careful.
We wrote it all down in Venture Hacks, but people said stop telling me how to negotiate the term sheet. Get me a term sheet. That’s where Angel started.
Jason: Next slide. As you can see here, the most recent version I think of Angel is or one of the recent versions, it started just you e-mailing people opportunities that you met when you were out having coffee with people, is that correct?
Naval: That’s right. Yeah, I was having all the … I had a small investment fund that’s investing startups. I remember I had coffee with one investor and he said, “Oh, I got this great high deal and we should move on it before anyone else finds out about it.” I was like, “Really, 2007 still works like that? You can use proprietary information to get some kind of advantage, it doesn’t make sense.” This is 2010.
I’ve talked to [Nivy 00:09:51] and I said, “Look, I know a bunch of investors. I do deals. Let me just take my deal. I start sharing it.” We took 25 investors. We started sending out just e-mails, and then we searched the Mailchimp, and then we had Wufoo and then nine months later we started coding.
Jason: The site has gone quite big now. How much has been raised through the site as a result of the site?
Naval: Yeah, it’s about a couple a hundred million directly through the site. It’s a result of introductions from the site. It depends and I count them like for example we introduced Sherwin to Uber. He put in 100K into C and then he did the Series C later. Do we count that or not? We don’t.
If we just count direct investment through introductions, it’s about two hundred fifty million bucks by ten million a month. Then on top of it, we also move it by four million a month, five million a month, just purely online through what we call it through syndicates essentially where people commit to small dollar checks online.
Jason: You yourself have invested in a lot of companies. Maybe we can go just to slide number five.
Naval: Yeah, it’s far too many.
Jason: Yeah. It’s impossible to read but how many companies have you invested in and what you’re current pace of investing?
Naval: I probably do one a month small, one a quarter big, very opportunistic. I mean this business that I love technology. I love entrepreneurship. Even I keep telling myself, “No. I’ve done investing.” I’ve got enough investments. I always see something new and shiny come along like “Oh, that’s a great idea.” Those people didn’t know what they’re doing and I just get caught up in it. It’s hard not to.
Naval: Let’s talk about something big which is the [jam-zat 00:11:27]. Explain to folks who don’t know what that is? What exactly happened that is relevant for founders?
Naval: Yeah. It’s sort of the only major by its partisan built to pass congress and get signed into law in the last four years which both sides sort of supported. It is a whole bunch of nice things for startups. There’s stuff that allows companies go public more easily. The shareholder rule, the 500 shareholder max is raised getting up 2000 shareholders.
But the two most interesting things from most recent in the audience is one is the general solicitation band is lifted, which means you can fundraise in public. It used to be illegal to go out there for example even at this conference stand up here and say, “I’m fundraising”. That used to be against the law. Supposedly you’re only allowed to market to accredited investors basically proven rich people that you had a pre-existing relationship with.
Now that’s changed. You can market to anyone. The downside is you can do a lot more work after the fact to prove that they’re rich and sophisticated and can take the risk. That can be pretty the owners. You got to check their income tax forms, talk to their lawyers, credit reports, those kinds of things. We’re automating that and building that in AngelList and we got to provide that as a free service to everybody. Be accredit ones and then we use it wherever you go.
The other piece that’s relevant is through crowdfunding where non-rich people can invest small limited amount using Startup. That’s coming down the pipe. It’s still not legal yet. In all these cases, the SCC drives additional regulations to kind of mandate how it works. SCC being a C, they’re going to automatically lead a little bit more towards what Wall Street once or how hedge funds work. We always have to write to them and get the community together and say, “Hey, look this is how Startups is doing. This is how it needs to be a startup friendly.”
Jason: You decided when … I guess it was September to launch public facing syndicates. I was one of the first to sign up for it.
Naval: Yes, if there’s a hundred times many startups, we need a hundred times many investors. That just seems logical. You’re not going to have billion dollar funds allocating capital into all these little startups here fifty thousand dollars at a time. This micro VCs must exist and they need a platform. We created a platform for micro venture.
What syndicate is it allows any Angel who is putting their own money, [scheming 00:13:44] the game which keeps you ethical and moral. To also raise money on the fly in the Cloud from people who want to back them like kit starter before equity. You can build up and put them in the backer base. Get first look at your deals and exchange for having first steps on them. We try to make it very simple and move it all online. What we have is we now have close to a hundred Angels who across them have close to a hundred million dollars in annualized backing.
Every time they write a check, their money gets multiplied 5x, 10x, 20x, comes into the same terms, gives them a piece to the profits. We formed a single purpose LSC in the sky that aggregates the money and puts one signature under the cap table which that Angel controls.
The backers are basically backing people who are putting their own money at risk, only paying a piece of the profits. There are no fees. They can start or stop backing at anytime with one click. They owned interest in the LSC which good itself is a secondarily traded and they got a beautiful in the sky dashboard brokers to view of all their holdings.
For the lead, they basically have a liquid back end to LPs now. They don’t have to be a full time venture capitalist. They can continue to do their job as an entrepreneur or whatever they want. Once in a while when they see a good investment from our entrepreneur, they are now in trust. They can scale up, do the series, this whole C drown with the Series A and then move on.
The folks that snapterms really value this weekend’s Startups audience, they tell me that every time I talk to them and we use it here ourselves, go to the weekend’s startup.com/legal to see our funny terms of service. Thanks again to snapterms for providing such a great services.
Everybody knows we have our choice of partners here at this weekend’s Startups. We’re that lucky. We use Snapterms. We’re like, “Hey, if we love Snapterms and we use it ourselves, why not …” We’re telling everybody about it anyway because we’re saving thousands of dollars. We might as well ask them if they wanted to be a partner in the program. They’ve been a partner in the program for over two years.
They’re second only to probably Mailchimp in terms of their support of the program. I really do thank them. You should too on your Twitter account. Go ahead and say thank you at Snapterms and tell all your friends, don’t waste the ton of money on your terms of service. Go to the pros, and the pros are the folks at snapterms.com. Thanks again, Snapterms.
Where do you think this is going in terms of your doing it right, doing it lean strategy which you outlined earlier which is to say, if somebody was going to do the fifty thousand, five hundred thousand, one point five million dollar deal, it seems to me that alliance would exactly what AngelList syndicates and AngelList sub classic is capable of doing.
Naval: Yeah. Essentially doing incubators AngelList and syndicates, Super Angel, seed funds, you should be able to get to escape velocity, and that’s the point at which you go for venture essentially especially in the consumer side. It’s a little different in the enterprise or hardware but especially in the consumer side. No, series A “VC” wants to talk to you unless you have escape velocity or unless you’re an extremely branded entrepreneur and even then they want you to have a product in some indication that you’re going to hit escape velocity.
Jason: I kind of hit that road already now.
Naval: Yeah. I think even the Super Angels and the seed funds are starting to operate in that basis. The bar has gotten really, really high because there’re so many startups.
Jason: You have to show traction.
Naval: You have to show some traction, or you got to show something, or you have to show, you do something really unique and difficult which is very hard to calibrate because everyone thinks we do something unique and difficult but you need a third party validation on that.
Jason: Something like Pinterest or Airbnb or even Dropbox might own its surface not look very difficult. Those things if they existed today, what would they need when each of those examples in your perception to get an A round from a top to your VC?
Naval: Today, the need like a hundred thousand monthly users climbing fast kind of thing, at least.
Jason: Dropbox would have to have a hundred thousand monthly users climbing at 20%?
Naval: At least 20. Yeah, 20 is probably the minimum.
Jason: Wow. They weren’t doing that when they raise previously?
Naval: Yeah, but the bar just moves up in the market constantly.
Naval: Yeah, that’s just the nature of the [beast 00:18:47]. I think Dropbox its first round was pretty money of three million or four million, something like that. The evaluations were lower back then. In 2007, ’08, ’09, the evaluations are probably like two, three, four million. In 2009, ’10, ’11, ’12, they kind of went little crazy. Got up as high as 6, 8, 10, now they’re coming back at that 3, 4, 5 range.
Jason: What do you think of the accelerators and their future and all of this? I mean we have obviously the [wide commenter 00:19:17] and texters are for sure, the gold standard went there. Then a lot of people creating derivative, knock off, or with good intention and sometimes well-executed, sometimes not well-executed. What is your advice to folks? Should you give 6% of your company away for 15K and join with our company or was it worth it?
Naval: Yeah, accelerator is a very complicated topic. We’re basically witnessing the unbundling advice controlling money, knowing what’s to give controls. That’s where the Angels come in. They come in with money. Syndicates bring backers and more money. The accelerators have sort of branded advice and institutionalize it. I think [inaudible 00:19:58] the new graduate schools except and instead of paying to go, you get paid to go. Instead of going out and getting a job, you got and create jobs. Instead of doing derivative work, you do original works. They’re better at graduate school and they’re shorter and faster.
If you’re just starting out, if you don’t know the context, if you need some help in which you’re building, if you need a network and if that first 25, 50K really matters to you, then by all means do an accelerator. If you have already built your product and start to get a little bit a traction, you can get their on your own. If you’re pretty savvy about how to build a company, and if you already had your network in place, then it’s a pretty expensive deal. But obviously it depends on the accelerator.
Yes, there are too many accelerators but there are also too many VCs, there are too many companies, there are too many Angels. There’s always too many of everything in this business. It’s like the Kentucky derby. Yeah, there are 16 horses running. Their 15 too many but you don’t know in advance which 15 are too many. That’s the whole point. You have to let them all run.
On the accelerator side, you could have argued that [texters 00:20:54] were redundant but they turned out to have some great classes. Angels Pad is redundant. It turned out that to be a great set of ex Googlers. Lemnos is redundant.
Jason: Which one is the most impressive to you in the last year? Pick the top … pick your top three most impressive accelerators that aren’t like [inaudible 00:21:07] than last year or texters?
Naval: Yeah, I don’t want to so much pick favorites. I would say I was …
Jason: What’s the most impressive class?
Naval: Yeah. I was surprised by a couple of outliers that came in from that field. Startmate seems to [inaudible 00:21:20] scheme Australian and bringing the best Australian company.
Jason: Which one is that?
Naval: Yeah. They bought some really good companies from Australia. They’re little off the radar. MuckerLab has had some great companies out of LA. There’s been Lemnos Labs has had some great hardware.
Jason: Where are they?
Naval: Lemnos are right here. They’re in the city.
Jason: Specializing in hardware?
Naval: Yeah, they had momentum machines which just get to find out by coastline and airway which a guy done by Andrei Sin and a few others. Angel Pad obviously crashed it MoPub exit but there are lots. To find your Startups is killing it internationally. Maybe they have a great space in mountain. Even the US they can beat the [inaudible 00:21:57] and texters and all that. But you go international, do the only game in town and they have a great brand and great reputation.
Jason: If you had an idea today which you can get your basic funding for … but you could go either enterprise or consumer. Let’s say you had the idea for Twitter and Yammer and you’re sitting there with your team of four. You’ve got your funding equal for either ideas, so you could get your 50 or 100K to start building, which one should you build and why in today’s market?
Naval: Well, it depends. First, build the one that you’re more passionate about because you have to be the best in the world to what you do. It’s the old Glen Ross. Number one gets a Cadillac. Number two, gets a set of steak knives. Then number gets fired. That’s the salesman hierarchy and it’s kind of the entrepreneurial hierarchy too. You have to be number one. In that sense you need to be passionate because you’re competing against the best in the world doing what they love to do.
That said most people are much more passionate about consumer stuff. If you can get excited and be passionate about something on the enterprise side, if you have some experience there, it would probably be foolish to wander away from that.
Jason: What are the chances of success today for Startups pursuing an enterprise strategy versus a consumer’s strategy? If you just say out of 10, this many will have a pretty good outcome versus a not good outcome?
Naval: Yeah, a consumer it’s some fraction of one out of 10.
Jason: Less than one.
Naval: It’s less than one. Yeah, and an enterprise is probably … it was like one out of three, one out of five. It’s probably one out of 10 now because it’s getting crowded. It’s just good to just teach your class. But that said entrepreneurial ventures fail all the time. Most of my companies fail. I actually started seven companies and I’ve launched for about 40 or 50 projects over my career. Angel is the first one that I would truly say its product market entrepreneur fit for me and might succeed down the road.
It’s a low heat rate over your career but you only have to be right once. It’s like your relationships. You only get married to one person. It’s like what you study in school. There’s only one set of things you study that stuck with you that you made a career out of. You only have to be right once or twice. Just keep trying. Just keep reiterating. Flow with it. Stay open. It helps if you have an arc or a theme through your life like people Evan Williams I think are always hammering on the same problem.
Naval: Yeah, micro publishing or easy publishing on the web. He just keeps hammering the same problems, so he just gets better and better and better at it. It helps if you have sort of an arc or a bent to your life and you’re not just hopping from sphere to sphere project. But that said, just a lot of this market timing and you learned a lot of lessons.
At first you learned how to … you do everything wrong. You make a whole set of mistakes by your first startup. Then the second one, maybe you do things right but you do it with the wrong people. The third one, you might do with the right people in the right way but at the wrong time. You learned all these painful lessons and so it’s just don’t lock. Just stick with it.
Jason: How many years did it take you to get to Angel Hacker? How long have you been into this profession?
Jason: AngelList rather.
Naval: I started early for my generation but back then everyone need to start Startups later. My first Startup started when I was 23. I started AngelList when I was 36 and now I’m 39.
Jason: Thirteen years.
Naval: Thirteen years pounding at Startups non-stop. I always like I would do my startup during the day out of my day job. I ran home at night. I brainstorm more startups. I just love it.
Jason: We’re going to get to audience questions in a minute. When you started AngelList, there was significant pushback from powerful people.
Naval: I don’t think anybody who is doing well under the current structure likes change but AngelList is a byproduct of the fact that Startups have gotten so cheap to build. They’re so many of them. Whether it was called AngelList or it was called something else, it would exist in some form or another. It’s one of those things … There has to be a place in the ecosystem to organize all the companies and all the investors.
Jason: What was the thing you found out about later that you thought was like somebody try to stop you the most because people have tried to stop this training. What some examples of people who tried to see the names of people. But have you tried to [scuttle 00:26:17] this project?
Naval: Sure. I mean the number one way in which people try to scuttle is people have a lot of money and they’re used to the old way of investing. We’ll tell the entrepreneurs “Don’t go in AngelList. Only the desperate companies go on there.” I think we’re sort of …
Jason: The desperate people go on there?
Naval: Yeah. It was the same thing they got said about [White Com 00:26:32] in the early days. I think at AngelList we work very hard to make sure there’s no adverse selection. We don’t take a cut from you from listing or forgetting funded. We don’t … All companies are on there. Now you have Yelp and Pictorn and others is recruiting on there. There’s no social stigma to being on there. That was something we fought with over the years.
Jason: They tried to make the companies look like they were the desperate ones. I agree with that.
Naval: Yeah. I don’t think it’s any like grand conspiracy. I think it’s more just human nature. It’s just like when you’re telling your investors that “Hey, I have unique access to deal flow. I’m this person that sees things that other people don’t see.” You have to be consistent with those statements that you’re making all the time. You go to entrepreneurs and you basically keep under this illusion of proprietary deal flow. I think people have given that up to now.
Deal flow is public, its access as proprietary. Access as proprietary still based on fuzzy brand but I think what we are arguing is that first we’re going to make that transparent. What you’re value add is, let’s see, put your cards on the table and let the entrepreneurs know and give the choice.
The second is there are a whole bunch of people today who are highly value added who have good access but don’t necessarily have the capital and that’s what syndicate sells. We’re sort of unbundling all the pieces to make them transparent. This idea that there are only certain people can invest and see and other people who can do growth is a false dichotomy.
If everyone can see everything, than you can cross boundaries all day long. You should have …
Jason: You’re back.
Naval: We’re back. You have Angels who should be able to do a series C or series D. There’s no reason with someone like you or David Sax at some point shouldn’t be able to put a million dollars into a company, have a ten million dollars syndicate behind you and do series B. At the same time a venture capitalist should be able to say, “Hey, the accelerator deal, I want to set up an accelerator in the cloud but I want to offer a hundred thousand dollars or hundred fifty thousand dollars for the 6% instead of the twenty five thousand dollars and here’s my structured network.”
The moment you take all the companies and all the investors and you fit all them in one place together, people can start cross pollinating.
Jason: I got it. When you started getting that resistance to AngelList, did that act as motivation and a signal that you were doing something right? Or did it make you feel like maybe I’m becoming an outcast? I’m too much of a rob-aroused?
Naval: Well, I’ve always been an outcast, so that’s easy.
Jason: It’s kind of easy to accept.
Naval: Yeah. [Red Herring 00:28:57] hit a job on me back on the day which says something like this guy will never work in this town again.
Jason: That’s what people say when you went to AngelList.
Naval: Yeah, and Red Herring is gone and I’m still here.
Jason: Yeah, you [inaudible 00:29:09] Red Herring.
Naval: Yeah, but I don’t care about that. The reality is just made sense. I mean there’re some things you just know with the core of your being that you’re meant to do in that array. You don’t care what other people think. It’s like when I married my wife, all kinds of people start coming up and giving me opinions, “Oh, she’s really good for you or I think this man …” I don’t care. I don’t need the opinion. I got it.
Jason: You should hear what they think with your wife?
Naval: Yeah, exactly. They try to talk a lot of it. Yeah, I would say that if it’s a right company then that means you know your domain. You know you’re meant to do it. Yes, other people’s data matters. If they really mean well for you and they have data, keep an open mind. But if they’re just giving you opinion on whether it worked or not, that’s useless. Just go to write a comment in TechRanch.
Jason: Right. How bad is journalism today? Specifically … I’m not going to mention any publications, but just in general, being inside who sees the actual real data, the real information and you have journalist who are writing who have never run a company, never failed, never succeeded, never took the company public, never sold the company, never did lease, never done investments basically opining on the industry, what percentage do they get right overall this sort of tech press? On a percentage basis, a hundred facts, how many they are right?
Jason: Facts, opinion they have.
Naval: They probably have the most to the facts right but they missed all the new ones and the new ones matters a lot. The problem is that journalists are not domain experts. Not really. Then they do not have much time. They’re paid on page user whatever. They don’t dig in. They don’t have the time to dig in. A lot of reporting is very superficial and incorrect.
The interesting part is the people who know the most and who can do the best writing are bloggers like Samuel Shaw for example. He is entrepreneur and now a VC. Understands that space, builds a brand by blogging, and you see there’s a lot. A lot of the great bencher brands now get though by blogging because those people can really, really dig in and understand the new ones.
The problem is that they’re incentivized. The moment you’re a venture capitalist, then you can start talking about how it’s really important that you take growth capital and you take money from the right people a.k.a. me and that you scale it fast. You have to filter out all the bias in the system and there’s a lot of bias in the system. Everybody has something that they’re going to try to sell you. I’m frankly when in business, right, because this is all business, at the end of the day, you’re here to build the product to make money. You’re not here for just to …
Jason: It’s not charity.
Naval: It’s not charity or you’ll be in the art gallery or you’ll be in the studio or something? Because the business everyone has bias, everyone has an incentive bias, and you guys attract their bias out when you listen to them, and what calls me is just a huge amount of blogging that goes on with no disclosure of the bias or maybe it’s just little line at the bottom. But the whole thing just drips of bias.
Naval: I think fundamentally people get passed that. People don’t like … They don’t necessarily it was like the sound of truth because it is harsh but I think people like the feel of truth. They know what truth feels like. I think it’s very important whenever you’re communicating with other people in this industry to try and drop your bias. That’s very hard to do. Very few people do.
I think that’s one of the reasons why Fred Wilsons ABC blog is so popular. Whenever I read it, there’s just like a certain honesty that comes off the pace. James Altucher, a guy bleeds on every page. He’s so painfully honest, I cannot believe him.
Jason: It’s train rack and awesome.
Naval: But yeah, people if like such honesty, such truth, I wish there are more journalists and bloggers like that who spent the time, did the homework and try to remove the bias from the reporting.
Jason: What makes a company … because I’m going to start putting companies obvious at the back? I’m hoping to put the top two or three companies here and put them out to syndicate if we make them real companies. What makes a company … this is final question before the audience goes … what makes a company in your mind worthy of me or Kevin or Dave putting it out to our syndicate? What do you want us to feel [inaudible 00:32:58]?
Naval: Yeah. I mean what makes a great company is actually different than what makes it a great syndicate and obviously I’m trying converse it true but one of them is reality. The other one is perception of reality. Reality, what makes a great company these are great founders who can build the great product that are going after an open market. That is your judgment call. That’s why syndicates exist. That’s why we don’t believe in this concept that you can just have a giant list of companies in the Cloud and you can somehow pick the winners and the losers without ever meeting them.
Someone has to go in there with judgment and access and do the homework and spend time with the company and figure out if the founders know what they’re doing.
Jason: Great founders, great execution, great market.
Naval: Right, that what’s makes a great company. What makes a great syndicate is you have to convey that to people somehow. The way you convey those will be just use a catch all term signal. It has to be a high signal deal. Are the founders accomplished? Have they done something in the past that would indicate their capable of things in the future? Is the syndicate lead a good judge? Do they have a good track record? Do they have early traction? Do they have any indication that the dogs, [or eating the 00:33:53] the dog food so to speak?
Is the product visually appealing? Or is the space not too crowded? Is it novel? Is it new? That matters too because on AngelList people see everything. By the time you see the third, or fourth or something, it almost is a matter how well-executed. These people are sort of already made up their mind with the space and probably invested in something. That’s what we call signal.
The backers and the syndicate case, right now, today most of them are actually pretty high value angels. For example, just this morning I’m looking at my dashboard, and one of the syndicate deals that’s out there right now, Jeremy Stoppelman and Charlie Cheever just committed CT, one of the founders of Quora and the CEO of VL.
Those people really shouldn’t be backers. I mean they are backing because they don’t necessarily want to do coffee meetings but they’re interested in this company. They believe in the lead but they don’t necessarily spent a lot of time in the company, this money economic state.
But the real backers are going to be limited partners, funds, people who are out of market, wealthy individuals who are not professional investors. Those people, you need to also send them a high signal but also show them there’s not a lot of access in this deal because otherwise someone is going to say, “Why am I paying you the carriages? Why am I following behind you? Why not I just go direct in the deal?” We call it high signal low access. That’s kind of our balance.
Jason: I was thinking if I did a video with the founders talking about the plan, and maybe showing it in me preparing with questions, would you think it would add value to the people where they want to … the backers want to see me interviewing? Could that be an advantage as a backer?
Naval: It might. My experience so far is that most investors don’t watch videos just statistically speaking. That said we have Q&A function on the profiles now, so you could ask questions and they give that answers. I think you get the same effect but in the more easily digestible format.
Jason: I got it.
Naval: You could add a video too. I don’t think it hurts, so just thinking on a ROI basis you [crosstalk 00:35:41].
Jason: What if I wrote a blog post about why I’m jazzed up about the company in the team? Would that be …?
Naval: That would be helpful.
Jason: That would be helpful. Okay.
Naval: But I think one of the very important things here is that the backers really need to understand. You need to convey to them things like what deals am I showing you so that if they know that there’s no adverse selection, you’re not just keeping the best deals for yourself and then send at the [inaudible 00:36:00].
They have to also understand that this is extremely high risk investing and that they should spread their bets over a long period of time. Be very careful and go slowly. It’s not gambling because gambling is a negative expected value in the house. It’s working against you actively. Hopefully it’s a positive expected value and it’s good for society. But the distribution curve of outcome looks a lot like gambling.
Jason: I got you.
Naval: People have to understand that and be comfortable with the risk and the fact as no liquidity.
Jason: What should an investor do in terms of number of deals? Ten deals? Thirty deals? Fifty deals? If they were going to optimize for a number, if it was your cousin and they said, “I have a hundred thousand dollars, should I do a thousand dollars into a hundred if I’m able to do that or should I do to two thousand into fifty?”
Naval: Yeah, I don’t think it’s a single right strategy there. It depends obviously in the conviction that you have around a given company. But just realize you can have enormous conviction on a given company and still you’re usually going to be wrong. That’s the nature of the business.
I would think that to get a decent portfolio in the Angels space, you probably need to be at least 25, 30 companies if you might guess. You could certainly take the point of view that you should do a lot more. There are practitioners who do less. They’ll do 10. But for those people that are diligence using to hack on those companies or putting a lot of wood behind that, already spending a lot of time with the companies. These are classic early stage venture capital type approaches. Unless you want to put in that time in diligence, you cannot narrow your portfolio down that much.
One thing you could say is maybe a little unfair but diversification is a hedge against a lack of knowledge. But here no matter how much work you do, you’re probably still not going to have more than 30% knowledge. Seventy percent is still up to the market and out in the air and so many things can go wrong.
Jason: Sure, pivots yet to come. Let’s take three questions or four questions from the audience and then we’ll get back to work. If you have questions, raise your hand. I want to make sure people in the back of the room also. Scott you’re a runner. After you get this first person, I want you to run all the way at the back to the next person. I’m joking. You can do two in the front, two in the back.
Speaker 4: Hi! What’s your position on [inaudible 00:38:03]? Is AngelList plans on using Bitcoin or something?
Naval: I’m personally a fan of Bitcoin protocol. I think it’s very innovative and world changing. Anyone who is interested in http for example should understand Bitcoin because I think it’s a next generation kind of protocol for distributed contracts. I have a little blog post that I think that it’s put up recently, as far as Angel itself … except in Bitcoin that’s kind of just a gimmick at this point. Like no one really uses it is a currency at scale. Those days have not yet arrived and may never.
Angels already operates in a highly regulated space with all kinds of mine fields. I don’t need to introduce Bitcoin in that I make my life a headache. That said, the moment it makes sense your bet, we’ll support it. I mean it’s a great thing.
Jason: Do you think there’s a chance that Bitcoin could become a total … what are the chances big coin becomes total wipeout like somebody hack the system or something bad happens where people just loses complete trust or government intervention or something?
Naval: Extremely high. I think right now the value of Bitcoin is based on like a 1% chance of a huge positive outcome or a single digit percentage chance of a huge positive outcome. That’s why it fluctuates so much because when those odds change like 2.5%, 3% the evaluation goes up drastically or down. That’s why it’s so [bottom 00:39:22]. If there were a hundred lottery tickets to buy like Bitcoin, I buy more and diversify. Unfortunately there’s only one.
Most people say Bitcoin is going to fail are probably right in the absolute sense like you may lose an investment but are they right in the sense that crept occurrence is going to go away? Or the proof of work function or the distributed ledgers is not going to make a huge impact in computing? They’re wrong about that. It is a fundamental technology melt.
Jason: The distributed ledger … the distributed ledger, explain that in plain English to somebody why that’s important?
Naval: Basically when you exchange money with other people electronically, you always go through a central authority and you need the central authority to keep track of who sent who money and so who’s got it and to prevent double spending. In other words, I get write you a check, I write somebody else a check, the way we know the check bounces, go to the central authority. The banks reconcile it and know that, “No, you can’t send that same money to two people at the same time.”
Bitcoin has the thought that probably decentralized way. It does that through a proof of work function which is way beyond the scope of this discussion but just go to my blog, look it up, and then through a distributed ledger to keep track of the transaction. Distributed ledger means that the record of all the transaction is stored in every wallet. Every user at all times has a record of every transaction, ever done in the history of Bitcoin.
Jason: You get pretty big at some point.
Naval: It is. There are ways to prove it and to put it part of it in the Cloud and its security look up and make sure that you’re only growing on the right part that you’ve authenticated in the Cloud. It’s solvable, but the basic idea is that you should not have to rely in anybody else to authenticate that a transaction happen. You can look inside your own wallet and say, “Oh yeah, that transaction happen in an absolutely, verifiable, irreversible way.
Jason: Next question.
Speaker 5: First, I just wanted to say, thank you guys for creating up a platform and a space for us to be creative and just do some great things. Secondly, I just wanted to ask especially with the AngelList you and the syndication model that you’re using, you’ve really kind of gone out of the box about how to come up with different ways in solving problems. I just wanted to know … because you were talking earlier about the issue with series A funding, is there anything that you were aware of with regard to flexible purpose corporations in California and the models that those might make for different types of investing in the future as well as your social startups as well?
Naval: Yeah, there’re definitely investors who focus on social startups and dual purpose startups and started submitting but it’s not a lot of investors. They’re solely not on the venture side. They tend to be much more in the Angel side. I know for example Tony Hsieh at Zappos and Tabreez Vergee and Mitch Kapor, and Johnny when people like maybe want to look at Startups that have a social mission with them but they’re not getting the huge amounts of LP dollars. These tend to be people investing their own money or raise small amounts of purpose driven money.
There is more of becoming but it’s still difficult. I would say you absolutely have to make it viable in every respect as a for profit business. Then if you happen to get one of these social interest investors consider that gravy but don’t count on it. Don’t think it’s going to give you any advantage.
Jason: It might actually give you disadvantage, would it not?
Naval: I think it does give you some slight perceived disadvantages with most investors. Yeah.
Jason: When a pitch starts out with, we’re going to give this amount of our profits to charity, the first thing I think is what is the business that’s going to enable you to do that or one just make a really great business and go be Bill Gates, [inaudible 00:43:06] away after you build the great business.
Naval: That’s one good point of you. Some of it is you have to believe that every company that we’re building here, these are not hurting people. Hopefully we’re not launching drawing the tax or whatever torching people. We are helping the system. We’re creating advancing forward at creating jobs, creating very great products. Look at how the iPhone has just improved the lives of people all around the planet. I would argue Steve Jobs is the greatest humanitarian in the last century in that he empowered more human beings with the computer in their pocket than probably anybody else.
Is AngelList socially conscious? I like to think what we’re doing is ethical and helpful and stands up for the underdog in a little person. But we can’t help everyone. We fail all day long and I don’t pitch that mission. I just pitch it as we’re building a great company in the business that we’re passionate about that allows us to recruit people and raise money and stayed focus and motivated. But of course we’re going to do it in ethical, socially, helpful kind of way.
Jason: Okay. Next question?
Speaker 6: Hi, I’m James Young for Telex Networks. Just curious on how long does it normally take traction, take to get some good traction on AngelList to get recognized each because [inaudible 00:44:19] on there for a while now but it just seems like many were not messaging it right or whatever.
Naval: Look, very few companies get funded as a percentage of the ones that are in the entire ecosystem, whether it’s here, whether it’s at AngelList, whether it’s wherever. It’s not a question of messaging necessarily. Basically right now companies get funded in one of two ways. One is … actually three ways … one is they have an investor offline who has a presence online and they bring the company in, other syndicate lead or send it to their followers and gets fund on that basis.
Second is you can just get discovered which is there are investors to say, “Well, I want to invest in that parking metered company, smart parking metered company.” They go look at everything in the space and Angels may reach out to all of them and then they find one.
Third is we feature it and we feature … we review for featuring every week. We review hundreds of companies and we look for which ones have tractions or novel interesting. In anytime of profile updates we go back to reviewing it. Unfortunately there really is no self-service reach out to investors for everybody because that turned into a giant spam fest.
You just have to basically keep updating your profile, using it as a calling card with investors. Perhaps it will get picked up and they do all they long. I mean five, 10 companies a week get funded but hundreds apply every week or not even apply but list. I think you have to treat it a little bit more like a LinkedIn. It’s not necessarily going to be the only way you find a job but it helps to have a resume on LinkedIn if you want to get a job.
Jason: What are some of the hacks if you will things that work really well? I have a lot of people say like, “Oh, I’ll just give you free shares at my company to … let me put you down as a [inaudible 00:46:03] or not as I take it.”
Naval: That doesn’t do that much. Those hacks don’t work. We close those hacks whenever possible. Really, the hack the works the best is having a great company, having a really kick ass up to date profile and keeping it up to date. Every time you get an investor for example at each year round, you add them and we notify all that investor’s followers that Jason is now investing in the round. People pile in.
People get the signals and notice and we’ll notice, “Oh, they put up three months of 30% growth. We should feature this company.” The data flows through the signals. Data are all structured and so it’s easy to track and there are lots of investors now using our API. We do probably five times of volume through our API that we do through the side browsing itself.
Jason: So many can hit the API and say show me the company. Is that a report that this amount of growth?
Jason: You’re a big fan of this rolling close on a convertible notes because it allows you to put, “Oh, Dave more invested this week and Jason Calacanis this week and whoever, Jeremy next week.”
Naval: I wouldn’t say this is to be a fun. I just think that it has certain advantages and one of those that you can keep fundraising all the time. The idea that you fundraise once then it’s closed up for 18 months, then suddenly somehow magically reopens and closes again, that is an artificial old model driven by the realities on how venture capital and fundraising used to work. The reality says now more of a blend. You can probably do your fundraising over six months where you can keep convertible note open.
After that to keep it open, you probably have to show some real traction because the signal from the original investing will now sort of be dying down.
Jason: I got it. One more question and final question. Two up here? Here comes Scott. He is running, get that real quick. Thanks God.
Speaker 7: Hi! Given what you know about this space, do you think it’s possible for Startup that does crowd source knowledge similar to Wikipedia to succeed? What would it take?
Naval: That’s a specific question. Everything that’s great on the internet is crowd sources some level or another. It’s very rare to find a business that isn’t some variation and crowd sourcing. Uber is crowd sourcing the backend. [Limos 00:48:10] and resembling them together. Google is crowd sourcing the intelligence says in the backend. Even the companies that don’t look like their crowd source are fundamentally crowd source.
With the internet … internet companies generally do that works really well is they aggregate huge amounts of desperate knowledge of individuals and pull them together. In that it says theoretically yes but the devil is so much in the details.
Jason: One thing I would add to that is just people often look at the Wikipedia as like a roadmap and I think it’s kind of a black swan. It’s like it needed to exist in the world. It exists and now it no longer needs to be done. This idea that you could make a better Wikipedia or that that is some sort of a roadmap, I think [crosstalk 00:48:53] because there’s a lot of pitches.
Naval: Yeah. You have to do some fundamental orthogonal way, either something will have to change. Like for example, mobile came along and Wikipedia didn’t keep up which is not true but something would have [inaudible 00:49:02] a lot of change. Or you have to have a fundamentally different take on the problem.
For example Quora you could argue is trying to augment or not replace but stand alongside. Wikipedia isn’t equally powerful and important entity but they’re basically saying is, “Oh, we’re going to crowd source knowledge for things where its specific knowledge and you need the answers from multiple individuals and there’s no single canonical factual answer.” Quora works much, much better for those kinds of things.
I think you may have to look for an adjacent domain of knowledge that Wikipedia cannot address or doesn’t choose to address.
Jason: There’s also a little bit of just one [inaudible 00:49:34], a little bit of fatigue in the crowd, so a lot of people think like you just started and then just a lot of people going to show up but when I started over the last decade of people doing crowd funding stuff is you build and nobody shows up or few people show up. You’d have to really have a long term slow build a fire from tinder and twigs and then build it up to a fire.
Hey listen. This has been a fascinating discussion. Naval, as always you’ve been very insightful and honest which always makes you a great fire as I chat. Let’s give it up for Naval.
Naval: Thank you everyone.
Next up on the list: Book Recommendations from Dustin Moskovitz.
Click "Subscribe to updates by email »" to get notified.
Scott K: Hi. This is Scott Kupor from Andreessen Horowitz and I'm here with Preethi Kasireddy and Jamie McGurk from our corporate development teams. We want to talk today a little bit about SaaS evaluation, so I'm going to start it off. For me, if you watch any of these news programs in the morning on all of our favorite cable news channels, it's pretty easy to hear the constant refrain which is holy cow this SaaS Companies are overvalued, what is going on here? Are we back in '99, 2000 just like we were kind of 14 years ago.
I think what people need to understand is a framework for how do even evaluate if that's actually a true statement. That's what I'm hoping we can cover today and I thought we'd maybe kick it off with Jamie. Give us a kind of a perspective on, what do you think is going on? What is causing all this bubble talk? What would do people need to understand and actually figure out, look is it in fact the case that we're in the bubble or is this all just normal behavior?
Jamie M: I think there's two things that people look at and point to the bubble and one is revenue multiples that a lot of these SaaS Companies are trading at, and I think it's both current and it's future expected loses, so lack of profits of a lot of the current public SaaS Companies that people look at those two things and they say, "Holy cow, Workday is trading it at 1.20 plus times revenue and expected loses for years are come," and they say, "This has to be a bubble." I think those are the two metrics that people tend to focus on that leaves them to conclude that are in a bubble.
Scott K: I think that's right, so we've got to train the world to say the way you evaluate companies is you look at the income statement, and you say, "Okay, how much revenue did you say generate? And how much only for share does it generate? If those are good then we all clap and applaud and go by the stock and if they're not, we all run for the hills. What's the problem then it maybe Preethi you're going take us off here, so my wise at wrong so what's the problem would actually just looking at revenue in EPS for this SaaS companies?
Preethi K: For SaaS it's a little bit different from perpetual models. In SaaS, you're getting the revenue from the customer over an extended period of time whereas perpetual models, you can look at the income statement because you get the revenue for the license in the period that they pay for and then you get maintenance fees overtime. You're covering your cost right away from the perpetual license, but in the SaaS you get it overtime and so that profitability doesn't happen until a certain period in the customers lifetime. Once it become profitable you have this huge existing base of customers that are paying your recurring revenue over time and your margins are actually higher than a perpetual model.
Scott K: Yeah, maybe we should unpack it all because for everybody. Maybe there's a simple example the way I think about it right. If you're a Oracle in the traditional model, I go sell your license for $10 million, I get that $10 million from you and then right in the [corn 00:03:03] which I sold you that business. I got $10 million that [throws 00:03:06], flows right through to my income statement. So all is good and I can pay my sales guy, I can pay all my other folks. I think most of you guys disagree that for a company like that the income statement is actually pretty good, that's a pretty relevant indication on what's happening to business, right?
Jamie M: It is an importantly the cost to acquire that company was incurred in that period before or during and as they paid for it. That cost is realized up front.
Scott K: That's right. We can actually see the profitability of that customer right away. We know that customer paid us ten million bucks. We know we paid our sales guy a commission. We paid some engineering guys to develop that software and we know right in that quarter whether that was a good purchase or not for us as a company. I think if we use that same example on the SaaS side if you work day for example and you sell a $10 million deal to a customer, that deal could over 12 months, 24 months, 36 months depending upon what the contracts are and Preethi to your point you've got this weird imbalance, which is, I've signed a contract that says I'm getting $10 million for the customer. There's no way that customer is actually paying me that $10 million today. They are going to pay me month by month as they use the service and so I've got this weird thing, which is I've got to go still pay the sale guy. I've got to pay all the engineers.
I've got to pay all the support folks immediately for that thing but that $10 million now is not coming in, in one quarter. It's coming in four quarters, eight quarters, 12 quarters depending upon how it works. Is that the right way to frame it? Is that the fundamental thing that you think is missing when people look at simple revenue and EPS.
Preethi K: Yes, exactly and that's why we tend to look at billings as fall, which is the sign of contracted revenue that you have coming in from customers and that really is like a leading indicator of revenue growth, and that's why you see a lot of investor focusing on billings more so than revenue itself. Because if Workday for example, signed a three-year contract with a customer and they have that unearned revenue on the balance sheet, you know that revenue is going to be coming in on the next three years. There's predictability in the cash flow, there's visibility into the future and what's different about SaaS is that a lot of financial analysts like it because you can really predict the cash flows over time.
Whereas in a perpetual license like you said, it's really every quarter, or every year you have to go out and acquire new customers. With SaaS you have an existing customer base and so you know that you are going to get revenue from that.
Scott K: Let's come back to that and I actually want to pick up on the billings com you made because my like spidey senses start going off every time we say, "Hey, forget about metrics that are actually in the financial statements. Let's go make up some new metrics to actually measure the business," and again, for those of us who are old enough to remember 99/2000 we used to talk about things like eye balls and stuff like that. You couldn't find those things anywhere in the financial statements and so I think people always get nervous right and this maybe some of the skittishness you have in the market where people say, "Once you actually have to invent metrics to explain why a business is valued in a certain way, that's scary."
Maybe you could just unpack billings right and maybe Jamie you can jump in here. Give people a sense of what that means and also how they actually see that in the financial statements as opposed to three of us just saying, "Hey, just trust us like the billings are really big."
Jamie M: Sure, I think a great example of that is Castlight's recent IPO where if you look at what the revenue was on the income statement on a trailing basis or even on a forward projected basis it was fairly modest, and certainly didn't justify the valuation that Castlight achieved post IPO. What investors are really looking at is the billings and I think that all comes down to predictability with a big bookings billings backlog that helps investors predict what next year's revenue is going to be and even the following year's revenue is going to be.
Think of billings as a backlog of future revenue. That really gives comfort that the growth rate is there and that they will achieved those revenue milestones.
Scott K: I know it's not always the case but often times you can get a good proxy for billings by looking at the balance sheet. You can look at things like deferred revenue, and there's all kinds of funky accounting rules that sometimes some of it may be in there and some of it maybe not but my understanding at least is when the Wall Street analysts look at these things they are looking at kind of change and deferred revenue, quarter over quarter as a proxy for this concept of billings. Is that pretty right?
Preethi K: Yeah, exactly. Deferred revenue is a liability on your balance sheet and it definitely represents ... usually represents new billings that you acquired in a quarter that you haven't serviced or renewal billings that you are getting from existing customers that you haven't serviced yet. Any change in that will indicate whether your billings are growing or shrinking, and that's really a good leading indicator into revenue growth.
Scott K: Right, well that's good to know. We are going to talk about billings. We are not literally inventing new financial voo doo metrics. We are actually talking about things that people can look to the GAAP financial statements and get good sense of how they are. I want to come back to something Preethi said earlier either, which is this whole concept of acquiring a customer with the idea that that customer is profitable but potentially over a longer period of time than we are used to thinking about.
Again, if we go back to maybe the simple example we started with, that Oracle customer. That customer is going to give us a bunch of license revenue upfront so they are immediately profitable, which is a great and wonderful thing and then they'll typically pay us a maintenance, an annual maintenance contract maybe 15 or 20 percent of the software license revenue. That's all good and that's actually quite profitable revenue but Preethi you kind of mentioned which people don’t quite have a firm grasp on it is in a SaaS company, what we are doing is. We are saying, "Hey, we are going to spend a bunch of money upfront to acquire this customer. On our income statement it's going to look bad."
"It's going to look like we are crazy and we are burning cash in parking lot because of that," but the whole theory is that customer probably is going to be with us for a long time. They are going to be sticky. We may be able to sell more stuff to that customer over time. Maybe just give a perspective on what is it about SaaS, why should people believe that that SaaS customer is going to stick around for 3, 5, 10 years any more so than that Oracle perpetual license customer and how does that play in to how people think about long-term profitability?
Preethi K: Yeah, sure. In SaaS businesses it's two different sales. First you're acquiring the customer and then you are retaining the customer and so we look at a lot of metrics around that around churn and so forth, which give us that conviction a customer is going to stay with you for a long time. If you are trying to [inaudible 00:09:26] then you know that you are going to keep that customer for let's say three years or so. You have an idea of how much that customer is going to provide value for you over their lifetime. If you can compare that to how much it costs you to acquire them and see whether you are becoming profitable on a company that's a good indicator of whether you can sustain and continue to spend money to acquire customers.
What happens is that the more you spend money to acquire customers in the beginning the more negative you go in cash flow, and a lot of investors have trouble understanding this. For example, a company will spend a ton of money, and acquire customers, and they are just starting to reach profitability but when they actually accelerate that growth they'll go into an deeper negative cash flow. Then when they come back up to reach profitability they'll start growing faster and that's that ... like that balance a lot of investors and board partners don’t understand, and right when you are about to press the accelerator a lot of people will tell you to stop, no slow down growth but that's in fact not the case for SaaS.
SaaS it's more of a land grab market where you have to acquire a lot of customers early on, and become the market leader, and that's what investors are paying up for. That idea of negative cash flow in the beginning versus a perpetual license where your cash flows are generally pretty not as negative is something that investors struggle with.
Scott K: That's definitely an interesting point and I do want to come back to that one. Jamie, lets dig a little bit deeper on this concept of customer acquisition cost and then Preethi did mention life time value, right. Just maybe help people to understand how to think of those because again, the way I think without people understanding the details there's always this fear that, "Hey, we are spending a bunch of money to acquire this customer. We have no idea if that's actually a good thing or not." How do investors should think about, "How much money should u be willing to pay for a customer relative to how much value I think I can get out of that customer over a period of time?"
Jamie M: In the previous professional license model you go out, and acquire a customer and you can base that against what that sale was worth at one time. That's fairly straightforward. In a SaaS model as Preethi mentioned the customer lifetime is uncertain. It's hopefully in perpetuity as well but you really don’t know. You have to ... the stickiness of your customer is really dictated in part by the product really because in a ... you have constant updates. If you are a sales force customer, if you are any SaaS model customer you are constantly updates to the product even on a daily, weekly, monthly basis versus you have a product that you buy and you buy maintenance for a three-year-period you are making another purchase decision at the end of that three-year-period.
You almost have to spend to reacquire that customer if you are the company selling them the product versus a SaaS model the purchase decision is made one time up front and of course, the relationship with that customer dictates how long they stay on board but you have the opportunity to keep that customer and make them sticky. I think those are the two things you have to ...
Scott K: I think that's interesting. To me I think there's a very good argument for why a SaaS is a lot more sticky, and I think a lot of it has to do with that. I think the other thing is and we see this a lot on in companies, which is as purchasing decisions have moved away from the CIO as the central purchasing control of all these things and now you've got department level purchases. You see a bunch of SaaS companies, Machado would be a great example of a company that's selling to the marketing department, right whereas in the old world you had to always sell to the CIOP.
To me part of the stickiness argument I think in favor of SaaS is you've got decentralized budgets now. You don’t have a throat to choke any more in the company. You actually get to go sell to the marketing department, which I think as a vendor is actually a good thing. Then at the same time because SaaS is so easy to use in many cases, and user interfaces are really big component of this stuff, there are probably many more users in each department of these SaaS products than we actually ever had before. Again, if you go back to old ERP products they were probably five to 10 to 20 power users in these organizations who ever touched these things, and knew about them.
Now you go to salesforce.com literally every inside sales person, and every customer account person has a license for this stuff and I think to me those are arguments why these software is even more sticky than traditional software because it's just permeated the enterprise in a different way than traditional products. You guys think that's true or no.
Jamie M: I do think that's true and I do think there's one other component to, which is there's a lot more cross-functional collaboration. Marketing to engineering, or marketing to the direct sales, or to the front office ... to corporate office I think a lot of that collaboration has allowed these systems to proliferate. As they use the same systems they are communicating across the same systems that has ... we've seen a lot of these systems proliferate throughout an organization now as well.
Scott K: Great. I want to come back to a point Preethi you made earlier and I would call it the growth hurts problem, which is these SaaS businesses are weird in the sense that the more customers I acquire and the faster I try to acquire them you have this major negative cash flow problem. You almost exacerbate your financial statements and make them look worse even though you are actually trying to do a good thing, which we think is grow the business faster. In the end, does that all work out, and is that a rational strategy for these companies to do to or are we better off just saying, "Hey, look. Cut the losses, don’t grow so fast. Let's just harvest what we have and be happy with the business.
Preethi K: Yeah, sure. To answer that question I want to use NetSuite as a example. If you look back to 2010 to 2012 they spent a lot more S&M and they went into negative cash flow territory but that helped them accelerate revenue growth and although their bottom line looked bad their top line looked great and that's good but then how do you know over time that this is going to look better and you gain conviction from that because by looking at by understanding the fact that an existing customer costs a lot less to service and keep than a new customer. There's stats out there that show that an existing customer costs one third of the cost to acquire a new customer.
If you look at the breakdown of your revenue as the portion that's recurring versus the portion that's new increases. That means the more renewal revenue you have the higher the margins are on that renewal revenue because you are not spending as much money on the renewal customers. Your profit margins go up over time but it's hard to tell that in the early years of a exacerbate company. All these exacerbate companies are six to seven year in their life time and this is going to happen 10, 15 years down the road and it's going to take a while. They have to establish that big customer base, and a large recurring customer base before you can really say they have really strong margins but based on what they've been tracking, and their margin improvement over time so far I think we can say that they're on the right track.
Scott K: We've seen at least in the public companies that have been around for a little while we've actually seen demonstrable improvement in their margins over time. We are not yet in nirvana in the sense they are not throwing off 30, 40 percent operating margins probably for most of these companies but it sounds like at least there's enough traction to show versus a couple of years ago now that growth is paying off in many respects in terms margin improvement. Is that fair?
Preethi K: Yeah, exactly and SaaS businesses generally have higher gross margins as fall. It gives you more room to pump money into sales and marketing, and R&D and so forth to continue that product development that you need to do to have a sticky customer.
Scott K: Yeah, Jamie maybe you can come in on this. I think the other thing that maybe gets lost a little bit here is a little bit nuance in the technology is what are their economies of scale are there for SaaS companies over time. We've talked a lot right and Preethi just brought this up, which is this concept of sales and marketing scale, which is, "Hey, it costs a lot to buy this guy up front but the cost of servicing that customer over time are low."
The other one that I think people don’t pay as much attention to is on the R&D side. This concept of you hear people talk about this concept of multi-tenancy and my personal view that also means to me that over time it's scale you ought to actually have much more research and development leveraged in these companies than you do in a traditional perpetual license company. I don't know if you agree with that or not but maybe just give some perspectives on what that might look like and what that means.
Jamie M: Sure and I think it's particularly evident in models that involved hosting and storage, and networking. The multi-tenancy that you point to is often used. That's a term that's used when talking about cloud-type deployments. If there's a shared infrastructure or a built in infrastructure that ties over a larger number of customers they are in essence sharing that cost. You get a lot of leverage to the extent that you are fully utilizing that infrastructure and spreading it over a number of customers, a number of deployments and therefore a larger revenue base. You do get a similar type of leverage ...
Scott K: You've almost ... we've talked about timing issues in SaaS companies, which is timing of revenue and timing expense. This is almost a second timing problem, which is in order for me to go sell to my first customer I've got to buy some storage equipment, I've got to go buy some network equipment, I've got to build the software. That's a big fixed cost that I have to undertake, and my hope of course, is that as I build more and more customers that that cost that ... the margin cost that gets assigned to each customer ought to be better and better. Again, you've got this funny timing issue which is to go get those customers. I have to actually incur this first big, fixed cost.
Again, when I look at my income statement I've got small revenue, big expenses but I've got to wait over time to see how that works. Is that the way to think about it?
Jamie M: Yeah, that's right. It's not completely ... you don’t incur the entire cost upfront amortize it over time but it's also not modular. You are not adding a fixed number or a amount of incremental cost per customer. Of course, your infrastructure can scale over time as you grow the business but to your point, you will have to start with a fixed amount of infrastructure to the extent into the that your growth can support that. It doesn't grow [inaudible 00:19:58]. It's you are getting ...
Scott K: Right, certainly at a minimum cash perspective I'm going have to go to be at a cash on this stuff. The accounting laws may help me amortize and depreciate that expense over a period of time. I think the other part about this to which goes back to the core of multi-tenancy and a lot of us here came from the enterprise software world. When you have enterprise software you are always maintaining multiple versions of the software all the time. You had customers who were on versions 1.x. You guys probably saw Microsoft just deprecated what was it XP or something the other day and there was this whole outcry. It turns out that every ATM on the world is being run on XP and so now we've got to go figure out if our money will actually come out the next time we go to the ATM.
My point is that in an enterprise software world, you always are maintaining different versions. You've always got on different versions and that has a real cost that I think people who haven't been in the business don’t appreciate because now I've got engineering teams dedicated to it, my support teams have to be bifurcated. The beauty I think at least of multi-tenancy SaaS generally, which I think also goes to our broader point about why these companies ought to be more profitable over time is every customer or at least in most cases. Most customers are running the same version of the software, right and I've got all kinds of economies of scale of my R&D organization not having to worry about that version that I released ten years, ago and what legacy customers I have.
Again, Preethi I know we are not quite yet to maturity for most of these companies but is it logical to think that R&D expense again as a percentage of revenue over time for these companies ought to be lower than they are in traditional companies and therefore again another reason why we think profitability is better for SaaS companies?
Jamie M: Typically, yes because you are not maintaining multiple versions for multiple customers. As you said, you have just one customer with SaaS and that's the beauty of SaaS really.
Scott K: Yeah, it's interesting. The other thing I was going to take us to a completely different industry but I think has a lot of analogies to SaaS, which is the cable industry, the telephone industry. What's interesting to me is none of this stuff is actually that new this concept that we are talking about here, right. If you think about a company like Comcast or you think about AT&T they've got the same problem, which is it costs a lot of money to acquire customers. It's one reason why the marketing right for these companies is so big but the reason they are willing to do that and spend hundreds of millions of dollar on marketing is once they get you as a customer the likelihood of you staying is pretty good.
Then now they've got this again, same thing like SaaS. It's a subscription business, so every month they are charging your credit card for a certain amount of money. I think it also explains why every time you threaten to switch from AT&T to Verizon they are very happy to accommodate the things and try to keep you as a customer. I don't know. Jamie, if you look at it maybe there's other analogies in the market but what is it that's giving people so much of an angst about SaaS when in reality subscription business have been around for a long time and there's really nothing new here. Is there something else you think that's making this harder for people to understand or is it just that because everyone's been trained on the perpetual license model that this is natural thing which people have to go through which is switching their mindset from perpetual license to recurring revenue businesses?
Jamie M: I truly believe that it's more of a switching the mindset because fundamentally the difference between whether it's an Oracle or one of the older perpetual license model, fundamentally they are delivering the same end result to a customer. They are just doing it in a different way. I do believe it's more of an understanding not just how the business model is shifted but how the impact on the financial statements is shifted and how the profitability of a new SaaS company or a relatively young SaaS company is going to look different than the profitability of that Oracle, SAP, Microsoft, et cetera.
Of course, those are in different stages of their life cycle but at the same time they are fundamentally realizing revenue and costs in a different timing period. I think it's a retraining. It's not a total brand new thing. It's just a new way of looking at a similar type of product.
Scott K: Just to net it out maybe I think where we came out is we've got thus fundamental problem, which is everyone loves to look at the income statement. We've trained the whole world to say, "Look at revenues, look at EPS and then lets figure out what revenue multiple, what PE multiples will be put on these businesses," and that's how we ultimately value them. It sounds like at least based on what we've talked about you've got a fundamental challenge with that model versus the SaaS model which is you've got all these timing problems that revenue doesn't all come in upfront like it did in perpetual license. Most of the expenses do.
You've also got this other concept of I've got incur lots of cost to incur a life time value that we think is going to be varied positive, certainly we'll see over time. It really is I think a function of helping people understand what are the proxies they could look for in the near time. Preethi you mentioned it. What is the customer acquisition cost relative to our projected life time value, and these are all things that actually we can estimate based upon the financial statements. We have to help people look towards those types of proxies as opposed to going back to the traditional valuation parameter we've talked about.
Thanks. I appreciate the time and I was with Preethi Kasireddy and with Jamie McGurk from the team and we are signing off.
Even small governments make for big customers. By revenue, many cities and counties would rank in the Fortune 500. These massive institutions, the number of them, and the need for productivity and technology gains means that extracting value is possible at scale.
Problem: State and local governments spend more than $30 billion on old enterprise software, built for the paper environment. Knowledge workers cannot derive insights, ask relevant questions, or manage information flow. Sometimes they cannot even tell how much money their entity spends. As a result, cities across America find themselves in financial crisis – in California alone, three cities filed for bankruptcy in 2012.
Problem: Energy consumption in developed and emerging markets continues to rise while production becomes more challenging. As easily accessible sources are depleted, producers must target more technically complex fields to extract natural resources. For example, more than 50% of original U.S. oil reserves remain down hole leaving 100 billion barrels ($10 trillion) which are not economically recoverable today.
Problem: Private wealth management firms spend more than $10 billion to separately create and maintain their financial aggregation, reporting, and analysis infrastructure. Innovation is slow because no single platform exists. Millions of people receive PDFs from funds and manually enter data into their systems from a variety of schemas, causing confusion, enabling fraud, and inhibiting sophisticated analysis.
Problem: The healthcare industry remains largely paper-based, and current systems are ill-equipped to handle transformational innovations, such as electronic medical records, cheaper testing solutions, and full genetic sequencing for individuals.
Problem: Although many functions, such as sales, recruiting, and business development, have transitioned into the cloud, data analysis still requires manual collection and input. Volume, accuracy, and timeliness of information are compromised in the process, which reduces the value of the information in these digital systems.
Sneak into the Old Chemistry Building (supposedly seriously scary)
The men's bathrooms next to the History Corner and Languages Corner
During the school year, these are almost always unlocked. Imagine the swankiest public bathroom you can, and now make it Chamber-of-Secrets-esque. They look like funny little huts from the outside, but inside they're pretty amazing. I don't want to give too much away, but there's a lot of funny/raunchy legends about these places. People even throw parties in them sometimes.
The Whisper Circle next to MemChu
One of my favorite Arillaga gifts. It's on the right side as you face the front of the church, and just looks like a sort of circular bench area. Go stand in the middle of the circle, and speak out loud to hear what happens.
Rooftop hot tub at the Four Seasons: Supposedly there is a waterfall! This sounds pretty epic.
2. Visiting the rooftop hot tub at the Four Seasons. It has really intense and pretty lights, and the pool is super warm. And there's a waterfall! A hot waterfall! It's totally romantic -- or, if you want to go with someone who's just a friend, that's totally fun, too. Because why are pools so beautiful?
You can go about this a few ways. Either just walk in and take the elevator to the 4th (I think?) floor. Or, if that would make you feel guilty, start out at Quattro, the bar/restaurant downstairs. After a $16 cocktail or two each, you'll probably feel a more justified -- plus, why not make a night out of it? Or, if you're a real goody two-shoes, you could actually just get a room for the night, depending on your budget.
Dave & Buster's: 33 minute drive. Expect to spend around $20 on food and $30 on games.
Movies: Mountain View
Movies: Redwood City
Go Karting: Gokart Racer in Burlingame is great. Other places have much worse reviews. Roughly $50pp.
Paintball: Santa Clara Paintball. Roughly $50pp.
Joseph D. Grant County Park. Head to Quimby Rd, and look for a spot to stop and look out of the San Jose area. The view is amazing!
Climb the observatory at Stanford. Technically trespassing. Amazing view.
Foothills Park. Go to the end of Alexis drive (Gmaps).
Just leave your car by that gate, hop over it, and walk up to the hill you can see in the attached image. Not only will you be able to see the stars, but also all of the Bay Area beneath you, and you'll also be the only people there, cause it's not a very popular spot. Source: Quora
Sky High Sports/Jump Sky High in San Jose/Santa Clara (25 minute drive) - $13/hr pp
Laser Quest in Mountain View
Shoreline Lake in Mountain View
You can drive if you have a car, but it's also right along the beautiful and legendary Baylands Bike Trail, which is an excellent way to start the date. Then rent a sailboat, paddleboard, windsurfer, paddleboat, or whatever.
Enjoy lunch or drinks at the little cafe by the boathouse before you head home.
Foot massage at the Happy Feet Foot Spa. $25/pp/hour. No reservations required.
Duck Pond in East Palo Alto
Note: All of the other items could also be date activities.
5. Contra dancing. Omg! YOU HAVE TO TRY IT! It's a New England folk dance that's sort of like line dancing, but different. No experience is required -- just show up on time for the free lesson and then dance the night away. TRUST ME. You will love it. You will laugh. And it will be awesome. You won't (and shouldn't) dance with each other the whole time... but after messing up a lot and wondering what the heck is going on, you'll be thrilled to be back in each other's arms. The Palo Alto Contra Dance people meet at the United Methodist Church (it's just the venue, though -- the event is not religious) from 7:30-11pm on the 2nd, 4th and 5th Saturdays of each month. People of all ages attend and are welcome. There's a caller to make sure you know what you're doing... and there is always live music.
Taqueria El Grullense in Palo Alto (well reputed!)
Jack in the Box in Palo Alto
Subway in Palo Alto
Happy Donuts in Palo Alto
Rose and Crown
Antonio's Nut House
I used to be a huge Verde fan (taro milk tea or taro freeze)... until I discovered Tea Era's roasted barley milk tea. Since then, it has definitely replaced Verde as my favorite boba in this area.
My personal favorite is Fantasia (Cupertino Village), followed by Verde (Mountain View). General consensus among my group of friends are that Verde has better pearl, but Fantasia has better tea... Also, don't forget to order half-sugar in Verde; they're super sweet.
I've tried Tea Era and Tapioca Express too, but I like them a lot less.
I recently discovered Ocha Tea Cafe on El Camino and Grant. They make the tea fresh using an expresso machine, just like the way the best places do it in Taiwan.