Howard Morgan of B Capital Group joins Nick to discuss Founding Renaissance Technologies & First Round Capital, The Origins of Idealab, Reasons Not to Grow your Fund, and How Sourcing from YC Has Changed. In this episode we cover:
Starting the First Quantitative Hedge Fund Idealab’s First Check The Story Behind the Founding of First Round Capital What’s Ahead in The VC Funding Markets
The host of The Full Ratchet is Nick Moran, General Partner of New Stack Ventures, a venture capital firm committed to investing in founders outside of the Bay Area. To learn more about New Stack Ventures by visiting our Website and LinkedIn and be sure to follow us on Twitter.
Want to keep up to date with The Full Ratchet? Subscribe to our podcast and follow us on LinkedIn and Twitter.
Are you a founder looking for your next investor? Visit our free tool VC-Rank and we’ll send a list of potential investors right to your inbox!
Transcribed with AI: 0:00 Howard Morgan joins us today from New York City. Howard is a true OG of venture, often featured as one of the best investors of all time. He has co-founded and led prominent investment firms including Renaissance Technologies, First Round Capital, and most recently, B Capital Group. He has invested in companies including Idealab, Climate Corp, Pandora, Augury Systems, Franklin Electronic Publishers, and Energy Vault just to name a few. Howard, it’s a true honor. Welcome to the show. 0:28 Thanks, Nick. Good to be talking with you. 0:30 Yeah, likewise. So I’m going to ask you just to take us back a bit in time here, Howard. You co-founded Renaissance Technologies with James Simons in the early 80s. What insights led to the first quantitative hedge fund? 0:42 Well, the main insight was that Jim had been spending time with his mathematician friends. He was a mathematician and Head of Mathematics at Stony Brook, and had always wanted to see whether or not you could predict the markets, and the answer was at certain timescales, you can predict the markets. And if you could predict them better and faster than other people could, you can make money. And what he decided in 1982, after three or four years of doing this with people out in the Stony Brook area, was that it was time to set up a real firm and create a true fund, if you will, that would specialize in collecting as much data as possible. And then using statistical and computational methods to try to do the predictions with people who did not know how the markets work. We didn’t know the underlying instruments. So even today, Renaissance and their research department won’t hire people who know the stock market. They don’t want people to think that they are predicting anything that has to do with actual companies or actual commodities. They just want them to be great at looking at a series of numbers and predicting the next number. 1:45 Interesting. And how did the two of you overcome objections from the early investors that come with being first? 1:52 Well, it was very volatile at first, but very lucrative. So Jim had already between 1978 and ’82, turned about $2 million into 70 million. The problem was that the mathematical methods that we were using, could only handle about half of that much money, the capacity of the strategy. One of the key things about any quantitative strategy in the markets, is it has a capacity, there’s just so much trades that you can get with that strategy. And in 1982, he said, you know, we’ve got strategies that will handle 30-35 million, we’ve got 70 million, let’s put some of it into venture. Let’s put some of it into quant. Can you give me your expertise – my expertise at that time was database management systems, so that we can help store giant amounts of data. And today Renaissance has literally 50 years of data, tick by tick data, that no one was collecting back in the 70s. It was very expensive to collect, because disk drives were not cheap. You couldn’t buy three terabytes for 100 bucks or five terabytes for 100 bucks on a USB key. So, and that’s a huge advantage, because they can then test and back test on monster amounts of data and monster amounts of. But by 1989, at Renaissance, the algorithms could handle a billion dollars. And when we got to that point, we decided that we would take the venture part out of Renaissance and leave it as a pure quant fund, which it is today and with, I don’t know how many, $60 billion or something under management, very large amounts of money under management. 3:18 More specifically, was there a key strategy of the signature Medallion Funds? And was there a point at which you knew it was working? I imagine before you launch you probably back tested some of the data but you know, was there.. 3:30 Yeah yeah, Medallion Fund went through a number of iterations before it was really quote working, which is to say; so there were other funds between ’82 and ’80-sort of-seven, when we really started Medallion, and they all made money, but it was much more volatile, there were down quarters and so on. By the time we got to Medallion, and the key insight in Medallion, I think, was that it was a collection of signals where you added signals every month, and you removed signals that weren’t working anymore, and remains to this day, a constantly evolving model. There are some signals that are in Medallion today that were there at the very beginning that have still not been carved out, as they say in the trade. But most of the signals are signals that have evolved in over the last 30 years. And Medallion remains an insanely profitable fund, but it’s only for the employees of Renaissance now. They have other funds which do different things. And those funds are doing what they said they would do, which is beat the market by 5- or 6- hundred basis points. The problem is that when the market goes down 800 basis points and you beat it by 500, you’re still down 300, so people. Whereas Medallion because of what it’s trading, which is mostly currencies, commodities, things like that. Can’t do that. Now Medallion has grown in capacity, but it’s still limited. It’s still, I don’t know, currently 8 billion or something like that. If you trade more you reduce your returns. 4:47 Was there a reason why you remained open to external investors and you didn’t become a prop shop with the other funds outside of Medallion? 4:56 Well, Medallion was open to other investors until the early 2000’s and then, I had already left the firm formally, although I was still physically there and working with Jim and other stuff, and that they threw out all the other investors. But Jim had always wanted to create a fund that could handle $100 billion. And in the 90’s, that was kind of unheard of. Nobody thought you could move that much money quantitatively in the markets. And the only way you could do it is by having a longer time horizon strategy, a strategy in years or decades. And so he worked very hard to create what’s called RIEF, the Renaissance Institutional Equity Fund, that was meant for insurance companies and people who had a 30 year horizon. And the goal was to be 400 to 600 basis points return above the SEP at half the beta, over 30 years. And that’s what it’s more or less been doing. But in a very highly volatile or uncorrelated way. For example, in the first three years, it raised $26 billion. And people loved the idea, people knew about Medallion. Everyone was told this is not Medallion, it’s different strategies, it’s different time horizons. In year three, the market was up 13, or something and Medallion was down 1, and so the $26 billion poured out and dropped back to sort of 6 billion. By year eight, if you stayed in it was 500 basis points above the SEP for the period at half the beta, just as promised. The difficulty is that the managers of the money in the institutional shops get evaluated either every quarter or every year, or in a rare instances, over a rolling three years. 6:30 Right. 6:30 And we had it, as a focal point, 30 years. And so the people who were making the investment decision to go in or out on the fund, were doing it on a different time horizon for the fund. And that’s kind of been fixed a little bit. So people are understanding a little better, but that’s the fundamental disconnect, particularly with the open fund. Now, institutional equities, they had a couple of other funds, some of them open and close, one in futures, Renaissance Institutional Futures fund and some others, but their performance has not been as slow and steady as Medallion, and Medallion continues to earn 35 to 40%, a year net after some of the most outrageous fees in the industry. 7:04 Yeah. Right. Howard, if we were to ask Jim, what the key to success was, what do you think he’d say? 7:11 Hiring absolutely the best people, and making sure that they were focused on mathematics and statistics, and not the fact that this was being used in real markets. So it’s pretty well structured, right, there’s a research team, they come up with the signals, they back tests, they do all that. And then there’s a trading team that does understand the real markets. You know, the research team basically says, okay, this should earn so many cents on this particular trade, or share or per contract. And then the trading team’s job is to try to beat that by executing better. And they’re evaluated on how well they do versus what the model says. So works great. 7:46 It’s true first principles and practice with both people and the data. Howard, you mentioned technology, I know that you were an early user of ARPANET. How were you able to leverage technology at Renaissance in the early days when computers were incapable of handling significant processing? 8:00 Well, as you say, I had in my research, when I was a professor for 15 years, I was a, I had machine 50 on the ARPANET. So I was in that very early, and I was the editor of the Database Journal, and the head of the first very large database conference. And a very large database in the 1970’s is far less than is on your phone today, way less than you can fit on your phone today. But we had a 5% fee structure to cover costs of much more processing. I think up until sort of 1999 or 2000, Renaissance had more processors per employee than Google, a huge amount of processing power, it had more storage than almost anyone else in the world for market data at a much more detailed level in many more instruments than anyone else. And we just poured money into that. And the fee structure, it started out as 5 and 25, 5% fixed fee and 25% incentive fee. And then one day, we looked at it all and somebody said, we’ll raise it and they raise the fees to 5 and 36. And the phone would ring off the hook the office of the CIO next to me Mark Silber, and I’d hear the conversation, which is “Yes, 5 and 36, it’s pretty outrageous. We have your wire instructions. Would you like your money back today? Or tomorrow? I know we can get it tomorrow, I can try to get it out today”. And they would say “no, that’s okay”. All right. And then they raised it to 5 and 44, which is what it is today. And and the same phone calls happened with, “Oh yes, 44 is pretty high. We’ll get you your money back when do you want it back today, tomorrow?”, because they were getting a net of 38. And so if somebody’s giving you a net of 38, why do you care how much they’re making, right? If I say to you, “I’ll guarantee you 38%”, do you care the fact that I could guarantee you 42%, or you know, 25%. It was so far above whatever anyone else was doing. And that’s where it stayed. And part of the reason is, it is a very expensive thing, probably much more computing power and much more storage. And you know, what we’ve learned in the last couple of decades about everything including AI and so on is that when you have much more computing power, you can get better results. Look at speech to understanding. 10:03 Yup. 10:04 We were doing speech recognition stuff, and the team that actually ran Renaissance’s research for many years, in some ways still does, came out of IBM Speech Recognition group. And a lot of the modeling Hidden Markov Model technologies that were used were very similar to what was being used in speech. However, starting in the late 20-teens, deep learning came in. And I heard Yann LeCun, who was a very famous AI researcher say, “you know, we thought we could make speech better if we had, sort of, 10 or 100 times the computing power that we had in 2008. It didn’t work, we actually needed about a million times the computing power”. But we got that by 2014. And then all of a sudden, you got Siri and Alexa, and Hey Google, and all those things that have dramatically better speech recognition, done in a different way, and in hundreds of languages. 10:53 Amazing, amazing to see how far we’ve come. And I think not all products are equal, and not all investment funds are equal. So they probably shouldn’t be priced the same. Howard, before we talk about the launch of First Round in 2004, it sounds like you started venture investing at Renaissance, and I know you’re doing some early angel investing. Can you tell us about those early days of angel investing and maybe venture investing at Renaissance and how that led to First Round? 11:15 Sure. Well, so I was a professor, as I said, mostly at the University of Pennsylvania at the Wharton School and the Moore School, both engineering. I was dual tenure in the Computer Science Engineering School and also at the Wharton School. And the ARPANET came in, I was doing research for DOD and at DARPA, and the Office of Naval Research. And one of the guys at the Computer Center said, “Gee, this stuff you’re doing”, which was windowing systems, so we had systems running Windows back in 1974 and ’75, long before you saw the graphical kind of Windows, things that you saw later on. He said, “I can make that much faster if I had a little microprocessor in the terminal, and the terminal could do all this”. And I said, “That’s great”. He said, “I’m gonna set up a company”. And so I said, “Okay, I can, I have a little money, let me give you a little money”. And I invested in that, and that was my first angel investment. It eventually went public was a company called Neoware, because something called the X Terminal became big and that was one of the first X Terminals. And that got me interested in venture investing. And then in ’77, a friend of mine at the RAND Corporation, Peter Weiner, got the first commercial license for Unix. Until ’77, Bell Labs only licensed Unix to nonprofits to universities and research labs. And Peter convinced them that if he could have a commercial license, he could build products and sell on this, and created a company called Interactive Systems Corp. And in fact, that’s how I met Jim, because I went out for the summer to work with Peter in building this company, and he said, and the funds are coming from another professor friend of mine, Jim Simons. And so I met Jim and we hit it off. And Interactive Systems Corp eventually got bought by Kodak and Son, and it formed the basis for AIX and for a lot of the Unix systems that were created in the ’80s and ’90s, and ended up really as a part of some responsive operating systems. So that sort of got me into the bug of doing that and doing it with Jim. And then Jim and I – in 1981, I came to Jim and I said, my friends are starting this Apple clone called Franklin Computer, it was called originally, and we had the first clone, it was an Apple Two clone. And we got sued by Apple and all the lots of interesting stories about that one, but that worked out well, and Jim said, “You know what, why don’t we just formalize this, I have this 70 million, I can only use 35 for the quant. But let’s do some quant types of venture together”. And we started Renaissance. 13:26 Howard, you were the first investor in Idealab, sort of the original venture studio or venture builder back in 1996. Do you recall your thesis and your viewpoint at the time of investment? 13:36 So I had known Bill Gross. So let me go back. In 1977-78, I taught at the Harvard Business School for a year on sabbatical to learn the competition to the Wharton School. And one of my students was Dan Bricklin, who created the first spreadsheet, VisiCalc. And then that got morphed and became part of Lotus 123. And I knew the folks at Lotus, and they bought a company from Bill Gross, which was his first company called HAL, Hierarchal Languages or something, but basically, it was Macros for Lotus 123 back in those days. And so I knew Bill, I met him, I beta-tested some of his late products. And we met at the TED conference in 1996. And Bill said he was leaving Lotus, and he was leaving Knowledge, he actually left Lotus to build Knowledge Venture. And what he really wanted to do was to start a company to make companies, and I said, “What do you mean?”. He said, “Well, you know, instead of building a 10,000 person company, he’d really rather build 100, 100-person companies”. And he said some of those would break out and become much more successful. And I said, “that sounds like a great idea”. And so I became the first investor actually, to be fair, Marsha and Steve, his wife’s parents, were the first investor, they put in the first check, but I put in the first outside check to Idealab and I’ve been involved ever since. 13:42 Howard many other venture builders or studios, those models have failed over time. Why do you think Idealab and some of the few names that we know haven’t? 14:57 I think the reason is there’s a single purpose driving them. So Y Combinator obviously succeeded for a long time,but the first guys there understood what they were doing. I think Bill was both the source of many ideas and his constant source of ideas, and a great salesman to get great talent in. I mean, if you look at the great companies that Elon Musk has built, you hear about Elon Musk, right? If you look deeper, you hear about Gwynne Shotwell. You know, the people who really make SpaceX work, the people who really make Tesla, work the people who really make solar work, but they work with the guidance, with the drive, with the inspiration. And that’s what’s happens at Idealab. So Bill Gross will say, here’s an interesting idea, good idea. And then we’ll go out and find a CEO for that. And the idea typically will have that CEO move the idea from true North where Bill was pointing, to maybe 20-30 degrees East or West, and make it their own, and then grow it. And by living inside the Idealab building, and working with the Idealab Team in a pretty constant manner, you get so much benefit. The goal, originally at Idealab was to say, we want to take away from the entrepreneur, all of the distractions, “Alright, I gotta find some space, I gotta buy some computers, I gotta buy desks, I gotta outfit, I need a lawyer, I need an accounting firm I need to do”, we took all that away. So you focus on product market fit and getting the best thing out, we’ll give you the help and everything else. We’ll give you the space, we’ll give you the accounting, we’ll give you the legal services, we’ll give you a team of developers and coders if you need them. We have a team of designers and all of that, which was very streamlined. So we created I mean, Idealab says they’ve created 150 companies, but it’s really been more like 500. And it had an advantage over the venture fund. The advantage is this, we could start a project at Idealab for $50 grand. And if it didn’t work, we could shut it down. And if it didn’t work, we would go get additional funding from Idealab’s own money, but also from other venture capitalists. But if you start out by going to the venture firms, and you raise a million bucks, and that says in those days, today, it would be much more money, you’re sort of committed, you can’t really, you know, fail as quickly as you’d like to. The mantra was fail fast and fail cheap. And you could fail fast and cheap. Because it was all one bucket of money. It was Idealab’s money. It wasn’t this company was failing. It was “that’s another idea, another project that we shut down”, and we shut down a lot of them. But we had a number of pretty successful ones. 17:17 Did you maintain that fail fast thesis when you launched First Round in 2004? 17:21 To some extent, but it was, it’s harder, because once you’re really committed to going into a company, and you have a company’s team as the mainstays, they’re not as willing to fail fast. When Bill had 12 companies running at a time, or 15, and you wanted to shut one down, you went to the guys running it and the women and said, you know, this isn’t working; why don’t you join these other three projects that are in the building here today, or here’s six more ideas, which one of them interests you to do? So we’ll take you off of this one, because it’s not working, as opposed them saying, oh I have to make this work, it’s my only chance of success. And that’s a big difference. Now at First Round, we were more traditional, right? We were investing into companies, but we were doing it very early. I mean, the insight that Josh Kopelman had in 2004. So let me move back – 1992, Jim and I invested in a company called Infonautics, started by Marvin Weinberger and Josh Kopelman. And it was, the product was called Homework Helper. This was more or less pre-internet. It was going to be used on Prodigy and AOL. Yes, I know I’m going back in time. As you say, I’m one of the OG’s. But as we got closer to reality at ’95, I had been involved with the internet, obviously. So we built it on TCP IP rails and I suppose Prodigy rails and AOL rails, were able to move it to the internet, and it eventually got sold to ProQuest. It became ProQuest. But what Josh said was when we were really mad in ’92, it took us $5 million to get to first product ship. We had to buy computers, we had to buy Oracle databases, we had to buy giant Sun Servers, we had to have people to manage those servers 24/7, etc. Josh and Infonautics went public in ’96. In ’98, Josh, while still at Infonautics, said I want to start something else. And we said fine and Infonautics took 10% of it, and I invested in it, and some other people, it was called Half.com. And Josh’s idea was, Amazon was starting to get strong, and I would like to be able to sell things for half their price. So he bought a whole bunch of domains, HalfOffAmazon, HalfOffeBay, HalfOff, but eventually just got Half.com. And you would sell things on Half.com at a fixed price, whereas eBay was bidding all the time, and lots of amazing, clever marketing stories in that one. One of them was they were brainstorming one day about how to get Half.com on the map, right? And somebody said, “Well, why don’t we get Half.com on the map”. So they made a deal with a town in Oregon called Halfway, Oregon to change its name to Half.com, Oregon for a year and gave them, I don’t know, $25,000 of computers for their schools and stuff like that. And all of a sudden we’re on the Today Show, we’re in every magazine, you know, this town is renaming itself to the internet Half.com, Oregon and Josh still has the “Entering Half.com” sign as a souvenir somewhere. But building Half.com only cost $2 million, because we had open source software. So we had the LAMP Stack in ’98: we had Linux, Apache, MySQL, and Python. And so you can build things much cheaper. And you started having even in ’98-’99, Rackspace-type of places, where you didn’t have to have the computer staffed; you had to buy them, they were still yours, but you didn’t need them staffed 24/7, they were staffed 24/7. So the cost went way down. We sold Half.com to eBay for a couple hundred million bucks. And it’s the basis of Buy It Now on eBay. But Josh didn’t want to move to California. So after a couple of years with eBay, and very successful years, where he became sort of a part of the old PayPal Mafia, which had been acquired in there. He said, “You know, let’s start something else”. He started an anti-spam router TurnTide was the name, so we started TurnTide. And the idea TurnTide was it would look at IP packets coming in and look up in a table. And if that IP address that was coming from was a spammer, well we thought it was a spammer or a spam bot, it would not acknowledge the packet for 30 seconds. So normally a packet comes in and you send an ACK to say that was received. And if you’re a normal person that that that comes 10 seconds or 30 seconds later, that’s okay, it’s unusual, but it’s okay. But if you’re a spam bot, and you don’t get a response in a quarter of a second, you just stop sending stuff. So it cut out 85% of this incoming spam right at the edge of the network, it never even got into the network. And so that was sold to Symantec in pretty much six or seven months. And the interesting thing about is only cost us $750,000 to build because the machines were much cheaper and the way it was working. So Josh said to me in 2003, 2004, I think we could start a fund to invest $250,000 and get companies started nowadays, things are much cheaper, and that I’d like to spend half my time doing that. And I said, “Okay, let’s try that for a year, I’ll give you half”. So the first year at First Round was, as we like to say, two full-time equivalents, that is to say half of Josh and half of me. And we made 48 trips to California that year from the East Coast, we were based in Philadelphia, I have places here in New York City, as well as in the Philadelphia suburbs. And then we decided this was actually going to work and we did 14 investments, and we had set it up as a one year fund. We then added another one year fund and hired Chris Fralic and Rob Hayes, as partners, Rob to cover San Francisco, and to grow the thing, but the idea was you to start companies and invest really early at seed stage, and that was kind of the first seed stage firm. 22:32 Howard, can you talk for a minute about First Round’s 6 P’s, and how that relates to strong founder execution? 22:38 Sure, at First Round, we voted on each deal, and we voted in a number of areas. We voted on the product itself, on the people on the team, essentially, and so on. And my own framework, the 6 P’s you’re referring to is what I use, Josh use’s something a little different, but similar, and the 6 P’s are people first. And of course, the myth is that people are the most important thing. And there’s a wonderful Dilbert cartoon where the pointy haired manager says, we just got our survey on what’s most critical for the company, and Dilbert said, I thought it was people, he said, No they’re ranked number seven, he said what’s first? Uh, paper clips, paper clips are first and then garbage. But people are obviously the most critical thing, you know, if I’m going to invest in you, I want to know that you’re honest, that you’re intelligent, that the chemistry is okay, because I’m gonna have to live with you for a decade very likely, and that you’re coachable. That’s the people part. The second part, product, is really more market. And one of the things that I learned over the Idealab years and talk to a lot of other venture firms was that you want a company created in a giant market, if the market is big enough, the product which is shown on day one, which is never the product that’s actually going to ship at a year, that the market is big enough to accommodate those changes, and that you’re able to morph and position and tweak and pivot to get to exactly product market fit. But if you’re in a giant market, that’s easy, because there’s enough people to do that. If you’re in a small market, that becomes harder and harder. And you might come up with a great $20 million a year product, but that’s not a venture scale return. So people and products are the first and most important and really people in market giant markets. The third is plans. Do the entrepreneurs understand planning and so many entrepreneurs are purely technical founders that they don’t understand business planning, business model planning, and understanding how to show that on a spreadsheet, which shows how much cash they’re going to need, over what period of time. And then profits, of course, we’re not in this for charity. I mean, I am in this for charity, in the sense that most of the money I make nowadays goes to charity because I’ve already got enough to live with my kids, and grandkids and so on. But the profitability is important. And they’re both Bill Gross, and Josh Koppelman and my current partner Eduardo Saverin at B Capital, are all fanatics about unit costs and understanding your unit costs. You really do have to make a little bit on each sale. You know, you can’t make it up in volume. And what we saw in the first internet bubble, was people trying to make it up in volume when they had negative gross margins on every sale. So you need good unit economics and that’s critical. So those are the four sort of traditional things, people, product plans and profits. And then the two more that are really important, are passion. And the passion is what lets the entrepreneur convince you as the venture capitalists, convince people that he or she is going to hire, convince suppliers to go with them, they have to exude passion, and in fact that they don’t exceed passion then we worry that they won’t be able to raise the next round. They have to really love what they’re doing. They have to really believe in what they’re doing. And that’s very critical. And then the final P is persistence, because they’re going to be setbacks along the way, there’s going to be their spouse saying, what do you do and working on Sunday night? And there’s gonna be that great that contract, it’s almost almost signed, and then it’s not getting done, they went with someone else. So how do you power through the bad times? Do you have the persistence to stick with it as opposed to cutting and running. And those are the six P’s that we use to guide things. I will say that the people and the market size are the two most critical ones. But the passion is probably a third in my mind, because you need to be able to show that passion to sell people on things. And when you get people like Steve Jobs or Elon Musk, you can see that passion coming through and sort of dominating everything else. 26:24 Well, it certainly helps with the persistence piece when things get tough, right? If you’re passionate, you’re just kind of a little stubborn about making it work, right. You mentioned Bill Gross again, one more question on him. I know he’s studied the reasons why startups fail. In your estimation, what’s the primary reason that startups fail? 26:42 Well, Bill said it and he did a big study. And it’s timing, market timing. So in 1999, if you look at the Idealab portfolio between ’96 and 2000, 2001, we had it right, we had all these things that we knew were going to work, we had a company called Z.com, which did streaming live entertainment over the internet, except that the internet was 28.8K, dial up modems, and streaming over 28.8K isn’t great. We also only had more or less 100 million people on the internet by 2000, worldwide, on those slow dial up modems, and a small number who are doing things in the office where they actually had T1 lines, one and a half megabit per second lines. And so we didn’t have a big enough market to overcome the difficulties. And we didn’t have the right technology. We were just too early. And I’ve had a blog, which I sadly stopped doing for a while, but it was called WayTooEarly.com. Because I like to say I made my money investing in two stages of companies, too early and way too early. And, you know, we could foresee most of what was going to be successful. And we were successful in a number of things. But like eToys, we did eToys, we knew that toys was going to be sold on the internet. And they were and eToys went public during that internet bubble Toys R Us had sales of five or $6 billion. It was trading on the market for about a one times revenue, five and a half billion, eToys went public with sales of $25 million and was trading at $11 billion – twice what Toys R Us was trading. So the bubbles we’ve seen the last year or so which is now cracking. We saw those back then. And the problem with eToys is we couldn’t convince the CEO that he needed to get positive gross margin on each sale. And that was one of those problems. So the timing, having the right idea too early. Andrew Weinreich with SixDegrees, which was the first social network, and then even MySpace, which was a huge player, okay, still a little bit too early in terms of number of users and so on, and the right thing. And finally, Facebook came along at just the right time when broadband really became widely available. 28:40 You know, Howard, something that I talked to my team about that nobody’s really ever brought up on the program. They talk a lot about timing the market, but they don’t really talk about timing the founder. And sometimes I’ll have this discussion with the team, you know, in our estimation, maybe this isn’t the right time or the right founder-market fit for the founder to take the professional ride of their life. And sometimes we get that wrong, right? You bet on a wonderful team, and it just doesn’t work out for a variety of reasons. But then that founder goes on to great success with the next business. I know that you’ve talked about this a bit. Can you talk about timing the founder, as well? 29:15 Sure. I mean, one of the great things about the United States, the reason the U.S. is so successful with venture capital, is that we allow for failure. We allow people to have a failure in their background and start something else. One of the nice things about Idealab was if we thought you would had given your all to something and it wasn’t working, and maybe it was too early, maybe you aren’t hiring exactly right, we found another idea for you to work on, which might have been very successful. So we’ve had founders do three or four companies at Idealab and it was the third one or the second one or whatever that succeeded. Jeffery Brewer did a couple of things. And his second one was a company called GoTo.com, which became the search engine Overture. And, you know, Bill had this notion that you could get people to pay for where they showed up in search rankings. And the world was aghast and but of course it were really well, to the point where Google ended up paying $600 million for the license, the patents that after Yahoo bought it for a couple of billion. And Overture is still part of the search engine at Bing, actually at this point. But that was a founder who didn’t do so well with the very first one and then did great with the second one. And he had learned more about how to manage what kind of things when needed to do to manage to both motivate employees underneath and also to do the the plans part of the people, products, plans, and profits part and became a great founder later on. 30:31 Howard, First Round famously passed when Jack Dorsey first shared Twitter. In your estimation, what’s the greater sin: failing to invest in a great company or failing to get the opportunity to invest? 30:43 So we said this at First Round, I say this at B Capital; the great sin is not making the wrong decision on investment, it’s not seeing it so that you can make the decision. What happened with Twitter was more interesting, actually. We invested in Odeo, which was Ev and Jack and Biz’s first company doing podcasting software, here we are in a podcast in 2022. And I can’t remember the year exactly, maybe it’s 2006. And then Apple announced that, because of the iPod, that they were going to give away free podcasting software. And so Ev came to all the investors and he said, you know, this is gonna happen, we’re not going to fight Apple, here’s your money back, and I’ll come to you with my next deal. And at First Round, we had a rule Josh and I that we would invest the maximum pre-money would be $10 million. And when he came back to us with Twitter, it was $20 million pre-money. And we said, we love you, but it’s just too expensive for us. Fortunately, for me, personally, my friends at Union Square Ventures, Fred Wilson, and Brad Burnham, where I was an LP, did it anyway. That proved to be very, very successful, obviously, so successful that a couple of years later, when Jack came to us with Square and then an even higher pre-money valuation, we changed our rule to get it and ended up with a 25 or 30x, or whatever on Square as a result. That Twitter story was important because it did say you wanted to be able to see that you make wrong decisions. We saw Airbnb, and both Josh and I did the same thing – we asked our wives, would you let me rent out the couch to somebody that we don’t know for the night. And the answer was a resounding “Not on your life”. That is the craziest idea. So we didn’t do Airbnb. 32:20 Did you figure out a way to break your rules? Like did you have a rubric for here’s when we can break the rules and go off thesis and invest at higher prices or lower ownership, and here’s where we can’t. 32:32 I don’t think we had a rubric. I think the focus there was the people, right? If it was a serial entrepreneur, you could break the rule if it was a first time entrepreneur, you couldn’t, on valuation in particular. And we saw valuations drifting upwards from 2005 to 2010, we saw competition come in, we started we were literally the only formal seed stage fund and Union Square was doing a little bit more A than seed, they were doing some seed. And you had Ron Conway before SV Angels was really SV Angels doing a huge amount of seed. But Ron ended up as more of an index fund, very successful and good friends with him and have invested with him. But it was more of an index fund of working with everybody out there. By 2010, prices had started to come way up, there was so much money coming in post global financial crisis. I think one year we did a slide for our Limited Partners saying in the last year, there have been 12 seed stage funds started beginning with letter F alone. You know, you had Floodgate and Felicis and Forerunner and a bunch of others. So it started to get very crowded. And that did have to change some of the decision making rules. 33:33 Yeah, you know, let’s go there next. You’re at First Round, other seed investors are entering the stage in the asset class, how did that change First Round’s approach to sourcing selection or portfolio support? 33:43 Well, I think we started off at the beginning building what’s been called the First Round Platform. And that was a huge win for us, because it allowed us to support a large number of companies with a modest number of people by having the companies become their own support system for one another. 33:59 And that happened since inception, Howard, since the early days? 34:01 That happened in 2000, in the second year, pretty much. 34:04 Wow. 34:04 In the second year, we put up the CEO network, where CEOs could ask questions of one another. And CTOs could ask questions of one another. Because if you’re the CTO of a six person company, you have no peer group. And what we gave you was a 60 company peer group at that point, and now it’s hundreds of companies, where you could ask questions, and if you said, “Gee, should I be using node.js as my framework”, if you’re a CTO, you’d get back for answers in the next 20 minutes saying, “Well, I tried it. It didn’t work for us, but we’re doing this kind of thing”. And someone else would say, “Yeah, it’s great if you’re doing such and such”. And so you had a peer group. And that took off very quickly. And so we saw that getting the founders to help one another became a very, so people would ask questions and put out requests or needs that they had and the system itself would solve them. So we as partners didn’t have to do all of the heavy lifting. And there’s some great examples. I mean, the one we cite the most Aaron Patzer started Mint, and he was sort of set up for a 5000 person beta. And he was on the first TechCrunch Disrupt and he won, and 25,000 people tried to sign up and the system basically froze and grind to a halt. 35:07 I remember that that, I was one of the 25K. 35:09 Yeah. So Aaron put out a note to the network saying, my sequel is is freezing and so on, does anyone know how to help me and Bob Hayes, who was the partner was trying to find people and so on, and I, who was the first editor of the Database Journal was thinking about it. But one of the other founders said, Well, here’s the phone number of Marten Mickos, and the founder of MySQL, maybe he’ll know what to do. So Aaron picks up the phone in Palo Alto at 6pm and calls Marten, not realizing it’s 3am in Germany, which is where Marten was and, wakes Marten up, and I’ve talked to him about it. But he said it wasn’t the first time he’d been called at that hour. And he said to Aaron, oh, you know what, call these guys in Palo Alto, they just tuned Facebook’s MySQL implementation, they’ll probably be able to help. So he called them and they basically asked him a couple of questions, the main one being what size buffers are you using? And he said something like, this isn’t literal truth, but he said something like, I’ve got four megabyte buffers. And they said, Oh, no, no, no, you need four gigabyte buffers. And he made that change, and all sudden signed up 20,000 people. But it didn’t come from me as a database expert. It didn’t come from Rob, the partner, we probably would have gotten him a consultant the next day, you know, to help solve the problem. But the network got him that solution in two hours. And that was just so powerful. 36:18 Through differentiation, clearly. Yeah. You know, Howard, in the past few years, we’ve seen some massive seed stage funds raised, yet you and Josh, were adamant in your opposition to expanding the size of the fund at First Round? Why the decision not to expand at the rate of your peers? And why did you move on for First Round? 36:35 So two things, the first of all, the decision was, we didn’t want style drift, but the big institutional LP’s called style drift, you start off saying we’re gonna invest 250 to 500,000 in the company, and then all of a sudden, you get $100 million, or you get $500 million. And you say, well, that doesn’t scale, because we only have so many partners, we better be investing $2 million, or 5 million, and all sudden you go from being a seed investor, to an A investor, to maybe a later stage investor. So we really try to keep the number of dollars per partner, as we grew in the 30 to 35 million that’s grown, because rounds are bigger. Now, I understand the newest First Round funds gonna be much bigger, but we stayed under 200 million for a long time and just went over that last fund or two. When I started, I was 60 years old with First Round in 2005. I said to Josh, I’ll give you 10 years, I love helping start things. That’s what I do best. I’m not the kind of operator that that Josh is a spectacular operator. And in year eight or so I said, “Okay, it’s time to plan for transition”. Generational transition is very difficult to funds. And so we started out thinking about how to create a glide path to sort of smoothly glide out and we did that, and I started executing on it, and I have a blog post that generational transition, which you can still find this up on Medium somewhere in the First Round thing. It’s called Moving On, I think. And it worked out very well, except that my inbox got flooded with “Would you join my seed fund? I see you’re leaving First Round, or stepping back from First Round?” And the answer to that was No. First of all, I’m still doing things that First Round, I’m still advising them. But no, that’s the best seed fund, I don’t want to hurt it, and I’m an LP and and so I want it to succeed even more. No. And then I got a call from Raj Ganguly, who I met over the years and he said he was coming to New York with Eduardo Saverin, and the Facebook co-founder, and they were doing something new and would I talk to them? And they came, told me what they were doing. And I said, that’s great. I said, I’ll give you a half an hour, sort of two hours later, I said, that’s great. Let me become a big LP. And they said, no, no, no, no, we want your expertise, even part time, and I am a third time with with B. And what your role will be is to help us make new mistakes, and not make the mistakes you made building Renaissance and Idealab and First Round. And really let us hit the ground running by putting in processes and procedures and policy and stuff that you learned over the decades, and that we would painfully learn otherwise, but this way we can short circuit that. And then they said first of all, we’re not going to compete with First Round. We’re B stage and later. Secondly, we’re global, First Round was mostly US and Canada. There were a couple of exceptions, but like one or 2%. And we’re mostly B2B, not B2C. And I had worked for the last 14 years with John Deere as a member of their Strategic Global Technology Advisory Board. And I understand the power of large companies, and the fact that they had to change and digitize and I watched John Deere going in 2008, when I joined their council from being a giant tractor manufacturer and doing a brilliant job of building big engines, and the stock trading at 20 bucks a share to becoming a precision agriculture AI driven big data company and the stock trading at 300 plus a share. And I saw that that transformation was gonna come to other industries. And that’s what B Capital was going to drive. So all that was exciting. And then they said, and by the way, we are partnered in an exclusive partnership with BCG, the Boston Consulting Group, worldwide, where they will be the platform team for at least a big part of the platform. What you built at First Round as a platform team, we’ll have some platform people internally but we’ll also have the 24,000 people at BCG on call as platform team, and the rest is history as they say we now have about 6 billion under management and we’re in our fund 3 series. And finally after about five years, we found somebody Gabe Greenbaum out in LA who had run Pritzker Ventures to do an earlier stage fund that I went talk to Josh and I said, you know, I’m gonna stop being a special advisor I, it’s five years and we’re gonna do some competing, not more cooperating, we actually co-invest with First Round quite a bit. But we started a an early stage fund, so I had to give up my Howard @ First Round email address. Two email addresses that I regret losing over the decades. And remember I was on the internet in ’72, ’73. So I’ve had been getting emails for literally 50 years, almost 50 years, H at Z, when we had Z.com. I was firstname.lastname@example.org, which was very easy to type. But when we sold Z, I had to give that up, and Howard at First Round back up. 40:41 So Howard, Bill Gurley recently tweeted that many founders that run short of cash will try to irrationally maintain their past market capitalization with dirty term sheets. We have increasing cost of capital and the threat of recession is looming, if it’s not already begun. Howard, aside from down rounds and flat rounds, what else should we expect in this asset class? And what advice would you have for your fellow investors? 41:05 Well, we’ll see merger m&a growing, we’ll see you know, sort of private to private mergers where you declare a victory by taking stock in something else, someone else’s stock, we’re going to see a lot more of that, obviously, we’re going to start to see the pretty term sheets, we’re starting to see term sheets with things like full ratchet protection, you know, as opposed to weighted average or anything like that. We’re starting to see 2x preference 3x preference, we’ll get to 5x, which is what we got to in the global financial crisis, we’ll see things like that. And what we’ve done to be capitalist tell all our companies, you know, what’s your plan to get to cash neutral with the current capital, and if not, what’s the least additional capital, you need to get the cash neutral? When I started in the business, really, in the early 80s, I visited Fred Adler, who was one of the early people in venture, and above his desk was a sampler that read “happiness as positive cash flow”. You know, if you don’t need to raise money, you’re in a much stronger position than if you do need to raise money, and so we’re telling our companies, growth is great, but not at the expense of running out of money. So grow slower, but stretch your cash. And I think we’re seeing more and more of that. But we will see definitely, first of all, we’ll see companies close, and companies that should close, and we’ll see companies that have raised too much money but aren’t doing well, do the mergers with other companies that are doing better, so they can get their hands on the cash, and give everybody a face-saving exit in those companies. 42:29 Howard earlier in the interview, you mentioned the early days of YC and the intense focus they had, do you think YC has changed for the better or the worse? 42:36 I think for the worse. We see a thousand companies, we see the companies are not just in the U.S., but because we invest globally, we’ve done some YC companies in India and so on. And I think socially, it’s a good thing, that they’re helping entrepreneurship around the world and helping lots more people get the benefit of it. But when you had 20 YC companies, you could go and look at each of them and think about how they were doing. When they have 1000 companies, it’s no longer very much of a stamp of approval, it’s too much of an index, they still have a high quality bar. And I respect that. But it no longer has the cachet for us as investors than it had in the early years. And their mission is a little different. They also have capital, they’re also a fund. They’re doing other things. So I kind of respect that. But it does make it you know, we used to fight to say what companies have gotten, then how can we find out about those companies before Demo Day? Which ones do we approach? You don’t do as much of that anymore. 43:28 Interesting. Howard, in the past, you’ve talked about how part of your purpose at First Round was to kind of teach the younger generation, and now it sounds like you’re contributing, you know, lessons learned to the team at B Capital, do you think venture can be taught? 43:41 Oh, I do definitely think some of the key elements can be taught, I think having great taste, and founders can’t really be taught. I mean, you either connect with founders and know how to connect with them and realize which are the right ones. But I think a lot of the other aspects of venture can be taught and are being taught in lots of colleges and universities and business schools. You know, what do term sheets look like, what are these various levels of financing, all those kinds of things can be taught – how to do due diligence, very critical set of skills. We do a lot of teaching on how to do due diligence, but that’s teachable. How do you do reference calls? We had, during the First Round days, we one of our annual meetings, I interviewed Scott Cook. And we talked about how do you do reference checking on people? What are some real tips for doing it? And he talked about not just obviously doing off books things we know go on LinkedIn, see who you know, that knows the person you were trying to talk about. But asking questions that get at their weaknesses as opposed to just giving the person you’re talking to the ability to say, oh, yeah, that Nick Moran. He’s wonderful. I, I can’t say anything bad about Nick. And then you ask and you say, you know, what was the dumbest thing Nick ever did when he was working for you, or working with you? 44:43 You may need a list, Howard. 44:48 So that’s a skill that can be taught and it’s an important skill that most people don’t really have innately. 44:54 Howard if we could feature anyone here on the show, who do you think we should interview and what topic would you like to hear them speak about? 44:59 Well definitely Josh Kopelman. Obviously, I think that he is the master of all this. He’s absolutely brilliant at doing it. You know, there are so many brilliant venture capitalists, but I think right now I would talk to some of the newer, more diverse managers. I have seeded 26 female and underrepresented minority managers over the decades. You should talk to Aaron Holiday or Nnamdi at 645 Ventures to see how they’ve gone about building a really great fund. 45:23 Amazing fund. Will do, he’s at the top of our list. And Josh has mentioned your voracious appetite for reading. If you could recommend one book to the investors and founders out there, what would it be? 45:33 Well, the book I recommend to everybody is a book by Daniel Boorstin, B-o-o-r-s-t-i-n, called The Discoverers, which kind of talks to the human innate need to learn and discover and master things. And it’s not a new book. It’s been around for decades. But it’s a spectacular book about how we discovered how to measure time and space and everything else that we do. And that’s the book I think everyone should read that book who is interested in learning how to master something. And when you’re doing a venture, you’re trying to master something. 46:01 Perfect. And then finally, here, Howard, what’s the best way for listeners to connect with you and follow along with B Capital Group? 46:07 Well, Howard@BCapitalGroup.com is my email, you know, I get 1000 emails a day. So I don’t promise to respond instantly to everything. B Capital does that, and then you can follow me as HLMorgan on Twitter, @HLMorgan on Twitter. And I tweet occasionally not not as much as I did in the early days, but that’s probably the best way and then you know, even though I’m no longer there, I would say if you don’t read the First Round review, you’re missing out on really great advice, and you should read the First Round Capital review. 46:34 100%. The man is Howard Morgan, co-founder of Renaissance technologies, First Round Capital, and B Capital Group. Howard, one of the OG’s of venture. Thank you so much for your insights today and your time, and special thanks to David S. Rose for making the connection. Thanks, Howard. 46:51 Thanks, Nick. Appreciate it. Bye bye. Transcribed by https://otter.ai