Startup Investor School Day 4 Live Stream

Galatians, you've made it to the very last day of start-up investor school. Thank you all again so much for being here and for being part of this. I am excited to provide the last day, so finally you guys can get some of your questions answered about ICOs.

So can anyone tell me where you're not supposed to park? There okay, there's parking, there's plenty of parking. There’s fewer of you today, so there shouldn't have been less of a parking issue, I hope. I hope, I hope it wasn't too bad, and I apologize for the issues with that. For people online, again #YCSIS for questions during Q&A, and there's still the check the Slack channel, of course. Today we'll try to stay more on time than we did yesterday. I trust we'll be better, but the good news is, as you know, there's wine and beer. Wine? Do we have wine or just beer? Wine and beer and pizza afterwards for those of you who are going to say yay, right? No sushi, just nothing, nothing. There will be nothing healthy, just we're investors; we can take it.

So there was one note that someone had that I thought was a little interesting. He pointed out that, Karl, thank you, that we haven't really talked about taxes. I can't imagine why we haven't talked about taxes. Well, taxes are an interesting issue when you're doing angel investing. Actually, if you're new to this, you actually realize it's a pain in the butt because when there's an exit, it's not like a simple thing; you have to do this thing called an installment sale because usually there's an escrow and you get paid. You have to figure out what your basis is; it's a pain in the butt. So I'm not going to talk about it.

But there's also this other thing that's worth investigating called qualified small biz or section 1202 which can have a significant impact. In past years, it had a huge impact on your potential tax liability, so look it up. I won't say any more than that.

Was there a question on taxes? Because I don't know anything about taxes. Oh, okay, high five.

Just a quick summary of yesterday before we get going. So yesterday was our day of angelic advice. Whether you realize it or not, those people up here are amazing investors, so hopefully you paid close attention to their words of wisdom. I always do; I've learned a ton from all of them. Alaud talked about how he finds billion-dollar companies, which, you know, if the game you're into is to make money here—actually raise your hands if the main reason you're angel investing is to make money. I think only two or three of you raised your hands. Raise your hands! Like, none of you are here to make money! What are you doing here? Because maybe because you want to change the world and you care about things, which is great.

But, you know, investing in billion-dollar companies gives you more fuel to do that, so it's still a good idea. And he talked a lot about how he helps companies. One of the things you hear, if you ask people again, it is a dear very tight network, if you ask people about Alaud, they'll all tell you he's incredibly helpful. You know, why does that matter? Well, that matters because everyone knows he's incredibly helpful, so you want him on your cap table. Page Mon is kind of special, and if there's one thing you should take away from Page Mon, other than the fact that he's really helpful, it's that anybody can get deal flow, right? Anybody can figure out how to get deal flow, and getting deal flow matters because you have to have choice.

Jeff talked about portfolio construction and asset allocation, and I think the takeaway for me there, what I've really learned from Jeff is that even, you know, most of us who are angel investors are not pros; we're amateurs. We’re blurred in the game; we’re doing it for personal reasons. That doesn't mean you can't be professional and rigorous about how you go about doing this, and that's the right way to do it—to be organized, to think about your portfolios, to remember that it's a long game, and to pay attention to what goes right and wrong.

He pointed out, like, you know, his portfolio construction was ex post facto, right? He looked back and said, "Oh yeah, I have a portfolio now; that's awesome." That doesn't mean you have to do it that way. Budgets matter, check size matters, sectors matter, etc. And then Andreia talked about personal brand and how much that matters. Do not ignore that. Do not think that every one of those folks isn't successful, in particular, because they have fantastic personal brands.

And I liked her laugh—Swizz's think like a boss—that's back to what I was saying before about being professional and thinking through how you're doing this. I think Ali had a little bit of a different perspective on things, which is interesting. He points out that so many of the great companies, huge companies, have been built on the backs of software founders, who were software engineers. That's kind of interesting, right? I mean Microsoft, Facebook, Amazon, Google, Yahoo, Netflix—all of those. There's a quite a bit of market cap if you add those up together.

So it's something to think about. I guess we came up with our little aphorism at the end, which is you want to look for the three B's, right? Brilliant founders in big markets with ideas that seem bad.

Right, so today's the conclusion of start-up investor school. I think it's gonna be a great day. We're gonna start with Andy Bromberg, who's going to talk about early-stage investing and the past, present, and future. He is the CEO of CoinList, the person who will answer every single one of your questions about the past, present, and future start investing. He has said for the first time he can give someone a product without them being there; they want it in advance. The best guerilla marketing ever.

After Andy, we're going to have Aaron Harris, who's going to talk about being good. You'll want to pay attention to that one. And then our final guest is a special guest, my friend Ron Conway, who's probably the most famous investor, angel investor, ever. So please welcome Andy Bromberg.

Thank you!

All right, good morning everyone! See if we can get this clicker to work! Nice! All right, my name is Andy Bromberg. I'm the CEO and co-founder of CoinList. We are the platform where the best digital asset companies are on their token sales, and we're investors like hopefully you all find high-quality deals in the space. But that is mostly not what I'm talking about today.

We're talking about the past, present, and future of early-stage investing, and we're gonna start with some trends that we'll see throughout the entire history of early-stage investing, move on to some history—the early days of venture: 1940, 1950, 1960, 1970, 1980—the Middle Ages, up to 2000, and then kind of recent history, and going all the way to the present, and then go back to some trends and talking about that.

I think it's really important to talk about the history of venture to understand what's happening today. A lot has happened in this space, and it is a relatively new space relative to other industries—talking 70 or 80 years old. And what has happened in the past really informs what's happening today, and you'll see a lot of the same trends surface throughout all these decades.

So some of the trends that we'll see over and over again talking about early-stage investing: first, decreasing costs to start companies. 1940 was very expensive to start companies. As we'll see today, it's very cheap. That cost is constantly going down. Second, a decrease in cost to invest for investors like you all are for funds. The cost to invest and the barrier to entry there has gotten lower and lower over time, and we see that happen over and over again in this ecosystem.

And then the last is the arc of the market bends toward liquidity, and we'll see this over time starting early and going all the way to now—that liquidity and faster liquidity is what some markets are always pushing towards. Now there's a caveat here, which is there are speed bumps in all these trends, and certainly in the last few years, we've seen the speed to liquidity go down. We think that's going to revert back, and we're gonna see that come faster again.

And then the kind of mega trend that we'll see through all of this—in all of these inform—is more and more capital being available to startups as it becomes easier to start companies, and you see more and more companies being started. As it becomes cheaper to invest in companies and easier to invest in companies, this availability massively increases, which means more startups, more people trying to change the world, all the things we've been learning about for the past few days.

So let's get into it. 1940 to 1960s. Jeff told me actually that when he talks about the history of venture capital, he starts even earlier than this—back around 1492—when, as he says, it was Spain VC with partners Isabella and Ferdinand who invested in Christopher Columbus’ labs on a venture to find a new world. They got 90 percent of the equity in that company, which was a great deal for them. Unfortunately, for you all, that does not happen anymore.

But we're gonna skip, you know, four or five hundred years to the 1940s and 1950s when the first venture capital firms emerged. These are kind of the two that we talk about most often as the earliest ones—J. Wolff Whitney & Co. and ARDC emerging in the '40s. ARDC really had what we consider to be the first venture win—they invested in Digital Equipment Corporation and returned about 500x in 11 years there. And that was the beginning of this trend toward our first kind of recognizable venture funds—what we think of as venture funds today.

In the 1950s, in 1958 we saw small business investment companies, which was legislation introduced by the U.S. government, which effectively pushed this forward. They said if you want to start a company that's a venture fund as we would call it today, we’ll actually give you leverage. We'll loan you money to invest in startups, and allow that to increase. And so that really pushed the industry forward way faster than it could have otherwise because there just wasn’t capital available at that point to invest in startups.

And then, you know, as we look towards the early days, some names you might recognize funded around this time: Venrock, Greylock, Sutter Hill, Draper, and Johnson. And at the same time—and we won't in this history—the Valley was coming into its own. So Soft Fairchild Semiconductor, some of the first Silicon Valley companies were started around this time.

A couple other noteworthy things about this period: one, this is the first time we saw the two-and-twenty structure emerge. So 2% management fee, 20% carry—that hadn't really existed up until this point in private equity firms for the first time, started using that structure. We still obviously use that today for a lot of venture funds.

And then briefly, you know, talking about what those deals looked like in this period, what we saw was that hundreds of thousands of dollars—which is obviously even more in today's dollars—was invested into these companies by the principals, by friends, by family, before it ever got to a venture fund. So we're not talking about having a product at the seed stage or having a really good deck or anything like that—we're talking about something being out in the market and making money on the basis of having a lot of friends and family and principal investment going in.

So very high barrier to entry, and what we'll see as we go through this is that that didn't change for 40 years. We saw that for basically the first 40 years of venture, you had to invest hundreds of thousands of dollars of your own money or scrounge it up from someone before a venture fund would invest in you, and that, you know, feeds into the trends that we were talking about earlier.

So we move into the 1970s, you know, some more names that you might recognize here: KP, Mayfield, CRV, Sequoia, NEA—all fun founded in this time. They were different than they are today. In 1970, CRV was a five million dollar fund, and KP was a seven million dollar fund—those are smaller than most investments those funds make today in raw dollars, but that was the entire fund size at that point in the 1970s.

You know, we also saw the emergence of angels in the '70s. As a historical note here, we think the term "angels" actually came about in the 1920s, but it was obviously not in start-up land, it was for Broadway. The first angels were theater angels who were investing in, you know, shows on Broadway to get them off the ground and allowed them to start to make money.

But we saw technology angels emerge in the 1970s for the first time. And then some major companies in the '70s that you will certainly recognize: Apple, Genentech, Tandem, Cray, Compaq, EA—'70s to early '80s—all of these companies were founded and obviously still massive success stories today. And we started to see more and more funds being founded, as you can see here.

And a lot of that was because angels, which were emerging, were becoming funds for the first time. But that could only happen because there was capital available. And what was happening in the '70s to make that trend start was that institutional capital was investing in venture for the very first time. Endowments and big corporations created a new thing called alternative assets that they were willing to put money into, and they were investing in non-traditional assets for the first time—that included venture.

And so angels that had successful track records in the early '70s went on to found some of these funds on the basis of, for the first time, institutional capital being available to them.

Moving into the 1980s, I've got a few charts here, actually courtesy of a great website called Reaction Wheel, that I recommend everyone check out. Some really good articles about the history of venture on there. The 1980s was a boom time for venture capital. If you look at this number of funds by vintage year on the left side, you've got the '60s and '70s, and then that first big jump, the third bar is 1980, and then it's year by year after that. There were a few dozen funds in the early '80s; there were more than 650 at the end of the '80s.

So that was really caused by this trend of more and more institutional capital being available and by this idea of just venture being an attractive asset class. It hadn't been around for that long, and people were realizing that there were crazy returns available to them if they started these venture funds.

And so this was certainly boom times. What's interesting here is that if you look back at kind of the mid-'70s, not a lot happened. So we saw this early boom in the '50s and '60s, and then '72 to '78, there weren't a lot of good deals—it was kind of a low time for venture. And then as we moved into the '80s, some of those companies that I showed on the last slide started to get funded—the Crays, Apples, and Genentechs—they got really interesting again, and we saw literally hundreds of firms started through the decade of the '80s.

But at the same time, here's a chart of tech IPOs through the '80s—1980 on the left, 1990 on the right. Early '80s things looked really good; 1983 we had more than 50 technology IPOs, and 1984 we were down at less than a third of that, and it stayed down there for the rest of the '80s. Deals were changing, and these IPOs skyrocketed. That fueled the rise in the number of funds, and then as things trickled down towards the late '80s, the market went down again. Certainly, the stock market crash affected that, and the number of IPOs went down really sharply.

The secular trend in the '80s was that you needed to invest earlier. So this is what I was talking about before—that up until this point, you had to invest hundreds of thousands of dollars of your own money to start a company. But now there were 650 firms to get money from; it was competitive. Those firms needed to beat each other to the deal. Eugene Kleiner had this quote that it's now a matter of weeks or even days to make up their minds because if they don't, someone else will. That was not true for the first 30 or 40 years of venture. They could take their time—they could take months to sign on a deal because there was just not a lot of capital available.

But as we moved into the '80s, that capital skyrocketed, deals became competitive, and funds had to invest earlier and earlier to get the returns they wanted, because otherwise they would lose out on deals and be left with the losers. In the '80s, you'll also see the market bounce around a whole lot. There was the rise of leveraged buyouts—a lot of venture investments were actually made in slow-growth consumer brands in the '80s, which was kind of an odd trend that happened—but there was just so much capital coming into the ecosystem; it had to be allocated somewhere, and we saw it going to slow-growth brands in addition to fast-growth startups.

And then, of course, at the end of the '80s, the stock market crashed, and that kept things down for a few years. This is, you know, tech companies formed in the United States—'80 to '84, '85 to '89, and '90 to '93—boom times in the '80s. And as we got towards the '90s, the stock market crash dampened everything down very aggressively, and we did not see a lot of companies get funded in the early '90s.

But in the '90s, we did get software and services and jondura funding Netflix and Amazon, and for the first time, we're seeing some of the recognizable software companies that we know today and the movement towards funding that from venture capital. To give you a sense of what happened in the '90s, in 1996, the venture industry as a whole had an AUM of about 12 billion dollars. In the calendar year 2000, LPs put 120 billion dollars into venture. So that’s not that it went from 12 billion AUM to 220 billion AUM in four years, but the AUM was 12 billion, and then in a single year, they put in ten times that four years later.

So massive boom times in the '90s in terms of capital availability, going back to that trend of more and more capital always being available to these startups over time. As we know, in the '90s, companies were getting backed because they could go public. We saw this path to liquidity start to increase, and people trying to get liquid very quickly—companies starting, and we’ve all heard the stories. Months or a year or two years later, going public, and that was a big part of the driver of this capital, because they saw these immediate returns that they could put in the structure. Changed a lot; carry changed. Funds were getting 30, sometimes even 40 percent carry on these deals because there was liquidity so quickly, and people couldn't put enough money into the space so the venture funds could start to do whatever they wanted and take a lot more of the economic upside.

These were boom times, and there was a bust, and I actually won't dwell on the bust here. We all know about it; it's really not that interesting for the sake of thinking about the history of venture capital. These markets are cyclical, and we see that in early-stage fundraising and late-stage fundraising, but every once in a while, there’s a bust. And there happened to be a really big one at the end of the 90s, early 2000s, and then we moved on.

And as we get to the 2000s, things were okay, but not great in the early 2000s. 2002 to 2009 was about 205 billion invested in venture, about 220 billion returned—so not really the venture returns that we were looking for, although certainly some of those companies have since returned more capital there, but things cooled off for a little bit there in the 2000s.

And then for the first time in a while, we saw the rise of some new structures, some new methods of funding and a new wave of models. 2005, Y Combinator, this fine institution that we were standing in right now. And then in 2006, TechStars accelerators started to rise, and this contributes again to this trend of the decreasing cost to start a company, making it easier and easier to start a company. And there was certainly capital available for those companies.

Obviously, around the dot-com bubble bursting, Angels went away in large part because most people lost a lot of money that they would have otherwise invested. But we saw their re-emergence in the mid-2000s leading up to 2008, and the recession, when of course that got stamped out a little bit more again.

And this is a really important one for, and I think underrated in the history of venture capital, which is the rise of convertible notes. Actually, in large part due to Y Combinator standardizing the terms and encouraging founders to raise money on convertible notes. Historically, all of these deals have been done with just equity—straight equity, selling preferred stock, but there are downsides to that, as I’m sure you’ve talked about this week. You have to close all at once; you have to go through this extensive legal process. You can't do these rolling closes and get things out there quickly.

And Y Combinator worked to standardize these terms, and we saw the rise of convertible notes from 2005 to 2009 as more companies started raising faster than the cost to start a company again went down thanks to that. And then super angels and micro VCs, SoftTech—I know you heard from Jeff, SV Angel. You hear from Ron—a little bit—lowercase leases, even seed funds like First Round Capital really came into its own in the mid- to late-2000s, and this was again just more capital early on in startups' lives, making it easier and easier to start a company early on.

Some more trends in 2010, after the recession, more angels, more seed funds investing smaller and smaller amounts. At the same time, we saw seed rounds crest a million dollars for the first time, and startups started to be able to raise really meaningful amounts of money from early-stage investors, often before they had real traction in the market. We saw a little bit of a reversion to preferred stock, likely due to this increasing institutional capital at the early stage.

In the earlier 2000s, most of the early-stage seed capital was coming from angels who didn't really want to deal with legal docs on their own—they were just investing their own money. Now we started to see those funds that I mentioned on the last slide investing, and their preference was often for preferred stock early on.

So we saw this move back to preferred stock, really just about 2010 to 2013 there. And then as the markets pushed towards driving this cost to start a company down, platforms emerged for the first time. Really two categories here: you've got platforms like AngelList that connected investors to—and at least at that point—connected investors to startups and made it easier and easier to raise capital as well as easier and easier to invest in startups.

AngelList really democratized that process to a large degree, as well as perks-based funding—so IndieGoGo and Kickstarter—making it so that without actually raising equity, you could raise early money. Obviously, Pebble is a phenomenal example from around here that was able to really get this business off the ground without selling equity and raising enough money to do things like build hardware in a time when that really wasn't being funded effectively with equity rounds.

And even a little bit more recently, going to the 2010s in 2012, this is another underrated item in the history of early-stage fundraising—the JOBS Act and subsequent legislation over the next couple years led to some really massive changes in the early-stage funding ecosystem. After 2008, all of securities law and investment advisory law got much, much tighter and much harder to invest in things as the government imposed regulation to avoid or attempt to avoid recessions like that happening again.

Except for venture. Venture is really the one class that got the exception to that. You can argue whether or not that's because it was common sense in the government or if it's because the venture industry lobbied really effectively, but at the end of the day, the JOBS Act came out, and subsequent legislation... and a few things came out of that. One, venture funds got this exemption from being a registered investment advisor, which made it much easier to operate a venture fund and avoided going down the path of that becoming really heavily regulated.

We got general solicitation—what we call 506 C offerings—where you can publish that a startup is raising on the internet and people can invest in that. Obviously, AngelList took advantage of that in a pretty meaningful way. 506 B got further solidified, which is private, non-general solicitation fundraising and allowing that to be done in an easier fashion. And eventually, crowdfunding—what we call reg CF equity crowdfunding—so the Kickstarter, IndieGoGo model, where anyone can invest in a startup but small amounts of money and really lowering the bar to becoming an angel investor or startup investor.

And again, making it easier for startups to raise money around here as well. We got Angels got a no-action letter from the SEC, which made it much easier to run syndicates and do this general solicitation. But really the early 2010s and going all the way to now have been a boom time for Angels and making it easier and easier for Angels to invest in startups.

We did see this speed bump of liquidity times increasing as startups got some leverage and realized they could stay private longer. We certainly think that's gonna revert as we move forward. Continuing forward here, see here in 2014, Y Combinator created the SAFE, which I'm sure you've heard about—the simple agreement for future equity—and we got back to convertible structures. So we'd seen that rise of convertibles, largely thanks to YC, from 2005 to 2009.

Reversion back to preferred stock and now with the standardization of this document—a real move back to convertible structures—and the vast majority of startups today are raised with either a SAFE or convertible but using a convertible structure. And the core idea there is that—and the core difference, or at least one of them, between a SAFE and a more traditional convertible note is that SAFEs don't have interest on them.

So most convertible notes that startups raise on say that you get a percentage point or two of interest every year, and if your debt gets paid back or whatever, you get a little bit more interest if you were investing in a start-up on the basis of a couple points of interest. You were in the wrong game.

Y Combinator realized that said this is creating accounting nightmares; there's no reason for this—let's remove that term from this; let's build a structure. It's convertible without it being debt and be able to move back to this much easier, faster, lower-cost solution for raising early-stage money.

And then the latest and greatest in 2017, we saw ICOs—initial coin offerings—and I'll touch on them briefly here. We can certainly talk more in the Q&A after, but ICOs are an incredible trend for early-stage fundraising. It's this proliferation of companies going out that can raise money without meeting with people, and that's such a new concept. You heard yesterday about how to run effective founder meetings. This is a world where companies don't even need to do that.

The cost to start a company is so low now; you can create something, put it out on the internet, and have people invest in it online without ever talking to those people. And that's not true in every case for every ICO. And certainly as many of them still meet with high-quality funds and Angels, but it's this real move towards absolute democratization of investing in early-stage startups.

And then I would argue that there's an even more recent trend than ICOs in 2017, which is everyone an Angel. And there are a couple components here that I would talk about. One is ICOs—so certainly ICOs contribute to this idea that anyone can invest online in these companies and get access because you don't need the company to give you access anymore. For a lot of these deals, they just fundraise, and you invest, and you get tokens or you get shafts or whatever you might be purchasing.

So anyone can do it without having access to those companies. Second, platforms like Republic and equity table—equity crowdfunding platform—allow actually anybody to invest, whether you're accredited or unaccredited, in security sales online again thanks to the JOBS Act and subsequent legislation. And then things like this that you all are sitting here right now, the idea that Angels are getting educated and taught how to invest and learning and meeting each other and engaging in the community is a really new trend. Historically, that's been just word of mouth.

You've known someone who's an Angel; they've given you the tips. But we're seeing this move towards education I would argue probably no better than here. And one more program to mention that I think is interesting is a program from a fund called Maiden Lane called Spearhead, which is giving founders education about how to become an Angel investor and also giving them capital to invest leveraging their investment.

And I see real real similarities between what's happening now with everyone being an angel and the earliest days of venture capital which we were talking about earlier in the '40s and '50s when now there's this flood of new money coming in. There's education becoming available and there's leverage being offered to people that are good at their jobs.

And so, you know, moving towards this and here the ICOs and everyone in Angel process really speaks to this societal trend of the co-creative process. That as companies are starting, they need help getting started; they need capital, they need advice, they need connections. And moving towards a place where Angels—and anyone—can be helpful in that endeavor is certainly a big macro trend.

One thing I would note here, just going back to the regulation, is that in the same way we saw the JOBS Act in 2012, I wouldn't be surprised if the ICO industry needs something similar—if there's a non-exempt safe harbor or some legislation around codifying some of the rules and guidelines around ICO fundraising.

So where are we now? What does the early-stage fundraising ecosystem look like today? There's normal seed fundraising—which you've spent most of this week talking about—mostly SAFEs or convertible notes raised from accelerators, seed funds, Angels like yourselves, larger funds with seed programs, usually call it a million to two and a half million dollar rounds. Obviously, we see outliers on either side of that as well.

And this is what we call what I would call normal seed fundraising. But, huh, also this other thing—token fundraising, ICOs—mostly raised on what are called SAPs (Simple Agreements for Future Tokens). You will note that that bears a real similarity to SAFE—Simple Agreements for Future Equity. It’s a document really modeled off the YC SAFE structure, except for tokens instead of equity—mostly raised from hedge funds, venture funds, institutions, and again Angels who are putting money into these ICOs, but they're raising ten to two hundred million dollars.

And so there's a question here of how we could be talking about these two early-stage fundraising mechanisms, where one is an order of magnitude or more money than the normal seed fundraising process. So we'll talk briefly about this and again happy to answer questions about ICOs in the Q&A and a little bit more detail.

But what is going on here? It's very different from equity investing, and you have to remember this: you own part of a network, not part of a company. And I'll talk about what a token is in a second here, but this is a really important distinction for a few reasons. One, governance—when you're an equity holder in a company, you're a shareholder. You can vote on what that company does, and you can be a part of that decision-making process. Not so for token holders, except in token governance mechanisms, but they don't own a stake in the company.

Valuation is wildly different. We can talk a lot about valuation—how do you value these tokens? It is not the same as seed fundraising. Obviously, if you're if the normal seat valuation is 5 or 10 million dollars, and a token company is raising 10 to 200 million dollars, those valuations cannot be done in the same way.

The investment processes I alluded to is very different. All of a sudden, you're not going and meeting with the company a few times, going back and forth, sitting down, coming to terms and doing it—most of the time, you're reading information online and then choosing to invest and sending the money. And so that's a very different investment process from normal seed fundraising.

And the future interactions are really different. In the same way if you're an angel, you probably get pretty personal investor updates, and the founder might come to you and say I need help with this or ask for connections. In a token world, they often have hundreds or thousands of investors. They aren't doing that for most of their investors; maybe they do if you know them already and you've been a helpful angel to them kind of personally, but if you're a normal ICO investor, you are not getting that level of interaction with the founders of the network that you invested in.

And then last of all, exits. So you talked about exits a lot this week. You know, you wait five or seven or ten or more years, and then you get a liquidity event. But with tokens, you're often liquid almost immediately. Maybe there's a little bit of a waiting period; maybe the network's not quite live yet when you invest, but at some point, you're just liquid. And now you're holding a liquid asset, often months or a year after you invested.

And the decision-making there as to whether or not you exit at that point or hold even longer-term is very different from normal seed investing, where you're locked up, and you don’t have a donor of a choice.

We could take ours on all these differences. One last thing I want to say on the ICO front is people break down different categories of tokens differently, but I tend to break it down into three categories: one, protocol tokens—this is an example; this would be Ethereum, where it's a platform that other tokens are built on top of. So its biggest value comes from being a place where other tokens use the protocol to build their own networks.

There are application tokens, which are those tokens built on top of platforms like Ethereum. An example would be Numerai, which is a distributed network of data scientists solving market optimization problems, but a very specific use case there. And then there are securities tokens or asset-backed tokens, where you take an existing security—real estate, or startup equity, or something else—and you put a token on top of it that represents that ownership.

I would say with ICOs, a word of caution—be careful; it’s a new industry—it's a year old really, and prices are a little bit irrational; maybe norms aren't really set yet, and so, you know, the market's young, and it should be treated as such.

I won't talk about tokens really quickly—what is it token? And I'll speed through this here, but I think it's important to understand what you're buying when you're buying into an ICO. Tokens are incentive layers on top of a network. So we like networks—Facebook is a network, right? It's a social network; it's a series of nodes that are connected together.

And now we can build an actual economic incentive layer on top of that network and create markets on these decentralized networks. But what does that mean? So let's talk about an example—Filecoin, which was created by Protocol Labs, which is a YC company, and it was the first sale that ran on CoinList. We were actually birthed out of the Filecoin sale, and the combination of Protocol Labs and AngelList.

So we have some YC DNA in our company. So Filecoin tries to solve this problem of storing files, and the obvious solution to storing files is Amazon S3. Which is “I’m gonna go; people want to store files; I'm gonna go buy a bunch of servers; then I will allow those people to pay me for storing their files on my servers.” Great!

There's a couple things there—one is, you might not want a centralized party storing files. They can censor what you're doing; they might go down; there are downtime possibilities; there are all sorts of reasons not to like centralized systems. And on top of that, we've got an Airbnb situation here, where there are actual abundant unused resources for storing files—there’s empty hard drives everywhere that aren’t stored with files and could be used.

So how could we try and fix that? How do we connect those two sides—the market? The simple and naive solution to that is put spyware on everyone's hard drives and monitor how much space they have and allow people to pay this centralized company to store files on unused hard drives. But again, you run into these centralization problems—there’s censorship possibilities, and there are other potential downsides of having a central party, and people just don't want spyware on their hard drives.

So the question is: is there a way, if Jeff has files to store and I have storage space in my computer, to connect us in a trustless way without a central party involved? And that's a hard problem, because if Jeff has files to store and he sends them to me, he needs to trust that I'm storing those files, and that's not something we want to build in.

People don’t always do the right thing. It’s a quibble. The trustless network. And the way Filecoin has solved this—and I'll oversimplify a little bit here—actually a lot. It's a very complicated technical system, is with a really clever token and incentive model. There are two pieces of this: one is that when I offer to store files, I have to take some Filecoin and do what's called staking them or bonding them.

So I take Filecoin and I put it out on the network. I don’t give it to anyone, but it's basically an escrow, and I'm saying, "I'm putting these out there. If I get caught doing something bad, not giving Jeff the reliability guarantees he needs, not properly storing the files, then I lose those tokens." So I now have an incentive to do the right thing and actually act in the best interests of the network.

But now there’s a second problem—how do you make sure I’m doing the right thing without trusting someone to verify I’m doing the right thing? Because I would just always say, "Of course I’m storing Jeff's files; they’re always available." And so the second element of the system is what they call the verifiers and the miners.

And so we could take all of you in this room, and you could try and check if I’m storing the files correctly, but how do we do that without actually giving you the files? So one example you might do—and they have a way more complicated and interesting thing called proof of space-time, which they use—is you could all go to Jeff and ask him if we're storing his files: "What's the 40th, 100th, 172nd letter in your file?" And Jeff would say, "It's B; it's A; it’s C." And then you come to me, and you ask me the same questions, and I say, "I’ve got B, A, and C."

Now all of a sudden, when enough of you have done that, we can get confidence that I’m actually storing the files correctly without you ever seeing those files because we have this distributed network of verifiers. And you now need to be paid for this work, so the network mines new tokens; it's inflationary, and gives you to that, gives them to you for doing that work. And at the end of the day, we have this network of verifiers of people with storage space of people with files to store who are all connected—not of us actually trusting each other but all of us incentivized to do the right thing.

And we've added this incentive layer on top of this network, and we've built a market on top of it. So now there’s value for these tokens, and as a result of the work that's being done in the network, and you can value those tokens that way. Certainly, we can get into that more in Q&A, but that is an example of what you’re investing in when you buy into the Filecoin sale or bought into the Filecoin sale in August.

What you were buying is tokens on this network—tokens that will be minted and inflated over time for verifiers and tokens that will be used to transact on the network and tokens that will be used to stake as stores on the network. So really briefly, what's next? Where are we going? And I think we just go back to the trends on this.

First of all, continued search for liquidity, and I think the trends that we're seeing there are one—tokens, as we discussed, often instant liquidity; are very fast, and moving towards faster and faster liquidity there as well, as I think moves toward secondary trading and more and more equities secondaries being done in startups and the market pushing towards earlier liquidity.

Second, the trend is continuing—companies are going to be cheaper to start and cheaper to raise—for platforms, seed funds, and angels; legal automation and standardization across a great YC company that can help with that. And so we’re gonna see it be cheaper and cheaper for companies to start and then, as always, easier for investors to invest: tokens, reg CF and crowdfunding, 506 B and C, and education, like the one we’re seeing here.

So these trends will continue, as far as we can see: liquidity and cheaper and easier to start companies and raise for companies. That is it—early-stage fundraising, past, present, and future. Thank you!

[Applause]

All right, questions about ARDC and the investing in the '50s? Yeah! Yes! The question was, "Can you invest in both equity and coins at the same time?" The answer to that is yes! In the same way we saw structures evolve in the '40s, '50s, '60s, '70s, we’re gonna see the structures evolve in the ICO market.

I would just say the priors are so low—because we haven't figured out the perfect ICO fundraising structure already, we're a year into the market—there's no way we've hit the right thing. We're seeing a lot of investors push for equity in the company in addition to tokens. You’ll hear about the SAFE—the simple agreement for future equity and tokens—or people just raising on SAFEs or convertible notes or equity with conditions in the company governance that, you know, they will also get pro-rata shares of tokens.

And so it really... but at this point, it really depends on the company, and investors don’t have a ton of leverage at this point in the cycle right now—it's very, very heavily weighted towards the companies.

But more and more companies are allowing you to invest in their equity in addition to tokens, or investing in equity and at some point that equity converting into tokens as the network goes live. First of all, well, the question was: should you... you know, you were told earlier not to invest in companies without meetings, and even video calls are probably not enough.

I am not telling you to invest in anything, and in fact, I would argue you should exercise a real discretion when investing in ICOs as it is such a new market. But it's a very good question—it’s a trade-off, and at this point, the norms in seed and normal seed investing are such that if a company is not willing to give you a meeting or a video call at the very least, there’s probably something going on there—unless it's the most hotly contested round in the world; the founders don't have time—it's a whole separate decision.

If they're not willing to meet with you, that's a problem in the ICO market. If you want to invest in the ICO market, and you want to see some of the returns that are possible in the ICO market, it's just a fact of life at this point in the cycle. I think that'll change over time. I think we’ll see more bifurcation between really early ICO rounds and later ICO rounds, where in the early rounds it’ll look a lot more like traditional seed funding than it does today.

But you know, at some point, you just have to make a decision as to whether or not the upside of investing in the ICO market in the interest of doing so outweighs the downside of not getting the guarantees and the interactions that you want to have.

So Andy, the purpose of the in-person meeting more than anything else is to get a sense of the quality of the founders, and how are you to judge an ICO if there’s really, I mean, if there are thousands of investors—you're not getting your in-person meeting. How do you make that call if that's the main thing that's going to determine whether this company underlying the ICO is going to be successful?

Yeah, it’s the billion, ten billion, trillion-dollar question of how to evaluate an ICO. You know, the first thing I would say is some of the most successful, one of the most successful angel investments in the past decade was investing in Bitcoin. And if you invested in Bitcoin when the first markets came out for Bitcoin in around 2010, 2011, you have made returns unlike almost anything in the startup ecosystem.

You could not have met with the founder of Bitcoin because the founder of Bitcoin is an entirely pseudonymous. And so you had to make a decision there on the basis of the future potential of the product. You know, at CoinList, when we think about evaluating these ICOs, we look at things like technology; we look at the backgrounds of the team; we try and interact with them as much as we can.

And there's a different mode of interaction, so oftentimes these teams are available on their Slack channels or their Telegram groups, and you can try and interact with them. It’s not the same as an hour-long meeting or a couple of meetings or a video call, but you can get a sense for their quality and get a sense for their backgrounds. You can read and see if they've actually developed some meaningful technology; you can look and see how thoughtful they are about structuring the sale, and then you know in what's probably heresy to say here, follow-on investing in the ICO market is a real thing right now.

And you know, the biggest funds will get access to these teams. If one of the top crypto funds is investing in a sale and they're investing five or ten million dollars, they are not doing that sight unseen; they are getting access. And so you can either try to get access through one of those funds, or you can talk to those funds and get the sense for their diligence notes or, if you want—and I’m not recommending this—you can follow them.

And what’s maybe a little bit better in the ICO market about following good investors is that the time to liquidity is so much faster. So at the very least, you can get out quickly after the token gets liquid. That's not to say it's always going to be a successful investment, and the hype cycle is certainly high right now, but at the end of the day, looking at things like the technology, the team, their history of shipping product, how they're structuring the deal, as well as the people involved in the deal already that have gotten the chance to do really deep diligence—those are some of the key items that we look for.

So, a question from out in cyberspace is around the realm of ICOs and token investing—is this more of a U.S. thing, an international thing, or how should we think about how global this new trend is?

Yeah, hello cyberspace! It’s a good question. Right now, it is a very global phenomenon. So about as much money has been raised by ICOs based in Europe as based in the United States was raised in ICOs in 2017. 2017 saw a couple billion raised in ICOs. It sounds like real money! It's real money—and the speed with which we've gotten to the real money phase of this ecosystem is incredible! Bitcoin, the original cryptocurrency, is less than a decade old—less than a decade old—and we’ve already seen hundreds of billions of dollars of value created, at least on paper, in this market and billions of dollars of real capital put into the market.

And so when we look at that, the scope—one of the beautiful things about crypto is that it is global, that anyone can start and anyone can send crypto anywhere. Now that has been truly accurate in the early Wild West days of crypto and ICOs. As regulation starts to increase, we will see a lot more restrictions put on the fundraising and on the activity by these tokens. We could spend hours talking about regulation here, but you need to follow it, and a lot of these tokens are securities, and you need to follow relevant securities law, but you know certainly it is still a global phenomenon.

And we're seeing a ton of activity in Europe, in Russia, in China, and Southeast Asia.

Awesome. More questions? Yes, are there... yes, are there venture capital lobbyists? Are there crypto lobbyists?

There are absolutely venture capital lobbyists—the National Venture Capital Association, as well as even kind of the Internet Association and some of the other entities out there are very active in lobbying the government. And I would make a very strong argument that the JOBS Act and all the good that it did for the ecosystem in 2012—and a lot of the subsequent legislation—came from really smart and aggressive lobbying efforts by very smart and connected people.

And on the crypto side, there are some, but not a lot. We have great organizations like Coin Center, which is not a lobbying firm, but it's an industry advocacy group based in D.C. that spends a lot of time educating our legislators and other folks in government, as well as a few lobbying shops popping up. But I think especially given the attention given to regulation in the space, we will see way more lobbying happening in crypto and in venture capital over the coming years.

Susan, so the question is whether Y Combinator has considered modifying the SAFE to somehow take into account a future ICO by the company.

So we're thinking a lot about the implications of ICOs, and we think about modifying the SAFE all the time. Unfortunately, every single modifier we ever think about would require us to take the "s" away and make it an agreement for future equity. But it will not be simple, and imagining trying to account for what an ICO might mean when nobody knows that—and that’s the hard truth—is what does it mean to own equity and tokens? How do the two different asset classes interact/clash? What happens in conflict? It is unknown.

So yeah, we think there’s going to be evolutions and changes here, but we don't know what they are, so it's premature for us to make any changes.

Actually, could I just touch on that really quickly to go back to the earlier question about equity and tokens? There's a question of, in a token world, what does owning equity in a company mean? If the network is built well, we would argue that no value actually accrues to the company itself. The purpose of the token is to remove the centralized party.

And so the company itself, the LLC of the corporation, shouldn't get transaction fees; they shouldn't get subscription fees; they shouldn't be rent-seeking on the network. So what that company becomes is an investment company that holds tokens. Protocol Labs holds a bunch of Filecoin tokens, and in theory, these companies, as they develop the networks and release these networks out into the wild, sure they may still help with ongoing maintenance of the network—they may help with contributing code to it—but they should not be getting any rent from the network; they shouldn't be getting fees in any way.

So owning equity in the company, if they do a successful token sale, may actually just be the same as owning tokens, because the company's only value will come from sure a little bit from a small team but then mainly from a holding of tokens that, you know, hopefully increases over time.

So if I can translate, you know, when Andy's talked about the future, he might bring maybe this means equity may not exist anymore, so you can just forget about these last four days.

Yeah, the question was, you know, where does the interest in the history of venture capital come from? And you know, I would really answer that as why is that interesting and important.

And I think it really goes to, if we think about the token ecosystem right now—and I mentioned this a few times in the presentation—there are a lot of parallels to how this is evolving over time. And as you look at the rise of cryptocurrency funds, we're seeing a lot of parallels to the early hedge fund industry. As you look at the rise of token fundraising, we see a lot of parallels to the early fundraising industry.

And I think the most important thing in all these things we're all striving to do in this startup ecosystem is to skate to where the puck is going. And if we sit and look at where things are right now, that's not that interesting; they're not going to stay there. That's not where the really successful products and investments are made; they're made by looking into the future and realizing how these things are gonna be structured in a year or two years, five years, ten years, twenty years.

And I don't think there's really a better way to think about the future than by looking at the past and seeing how those trends changed and trying to think about how those might apply in the future.

What an interesting conversation we've been having is about liquidity, and you mentioned George Torio’s of ARDC's investment in DEC, which in 11 years later yielded a 500x return. Interestingly, Dropbox is going public this year just 11 years afterwards, and it turns out that the slower liquidity is usually the bigger returns you get.

So is it a positive thing that we have this trend towards faster liquidity?

Yeah, I think it's a really interesting question. As with anything, there are pros and cons to the markets moving like that. And I would say there's even, setting aside returns, because I think that, you know, the longer liquidity may lead to greater returns at the moment of liquidity, but if the company continues to be successful in the public markets and you wait out for a longer liquidity timeline, you may make just as much money as if they had stayed private that whole time.

But there are other considerations here, too, and one of them, which is really big for token companies right now, is do you want your big-dollar investors to be liquid right after they invest? They might get out, and one of the greatest things about having early-stage investors in your company is them supporting you all through the lifecycle.

In venture, it just so happens that's forced—they don't have a choice once you invest; it's really hard to get out. And so you're stuck with the company through its ups and downs, and as everyone here has seen, companies go through ups and downs, and it's often worthwhile—not often in the right cases—that it's worthwhile to stick through those ups and downs.

Companies can be killed prematurely by early liquidity, and so what we're seeing in the token market is that even though tokens in general are liquid much faster, early investors are offering locked up now. And it’s probably not for long enough yet—it’s for a year or two years, or 18 months—but I think we’ll see a trend towards sure, the market trend is towards liquidity; the tokens will be live; people can buy and sell and trade and move around, but early investors will be locked up for a long time because you need that support, and you need that base of investors to stay with you and support the company.

And in fact, Filecoin did precisely that.

Absolutely, yep. A great example! Okay, we’ll take a couple more questions, and we will have a quick break yet—it's another trillion-dollar question! The question was, “What happens when you want to do a second round in the token world?!”

Not, I don't have a great answer for that question other than to say that the crypto world answer to that question, if I were to represent the whole community, is you shouldn't ever do one. And the reason for that is that—and this goes back to what I was saying earlier—the idealized kind of platonic form of an ICO is raise enough money to release this network, release the network, and then it is not yours anymore; it is the world's.

And you may be incentivized by, you know, Holdings that you have of that token to continue developing it as the network developer, but you're no longer in need of the company having money, and so the argument that people would make is that in this perfect world, you just don't need to raise again.

You raise enough to launch it; you launch it; you put it out there; you develop it over time, or at some point, you might give up on developing it, and someone else takes it over, or someone might fork the network and start a new trend of it. But at the end of the day, you shouldn't need to raise a second round. The reality of the situation is that a massive number of token companies are going to crash and burn because they've run out of money, and their network will never really reach its potential.

So perhaps there are ways where we can start to build the entrenched fundraising, either scheduled or unscheduled, or allow for more support for it for later raises. But the ideal form does not involve ever raising again after your first raise.

Okay, do we have a last question?

Absolutely! Yes, the question is: ICOs might be similar in some ways, but with a nuanced perspective of the initial price matters. How can we use what we have learned in the IPO market to support the ICO market—not causing crazy fluctuations in price early on?

A couple pieces: one—and this is not what you were saying, but just to correct a misconception that other people have—I don't think ICOs are anything like IPOs. I think it's a really actually unfortunate naming convention that we have called them the same thing. They are much more similar to seed fundraising. I drew the parallels up there between those two things. IPOs happen when a company’s mature; they have a million disclosures to make; they've got revenue; they're supported. And that's just not the case with ICOs; they're much more similar to early-stage seed financing.

Recognizing that that's not what you said, there are still some similarities. And we've gotten really good as a market at stabilizing IPO prices for the most part—there are always exceptions—through a variety of things—green shoes and different mechanisms around the IPO. I think that one piece that will be critical here is the divide between really two types of ICOs.

There are sales of securities, which I would argue is almost everything that has been sold so far, and these things are illiquid. So most of these securities are offered under what's called a Reg D offering, which is the same way most of us raise money—which means that the security is actually locked up for a year. So once you buy it, you actually can't sell it for a year—the network might go live and you might have liquidity at that point, but that’s well after the ICO.

And so in those cases, it’s actually not a huge consideration. You buy once, you wait a while, and then a market gets created and you start to sell into that market. The other case is direct ICOs, where you're actually buying the token as it is live and then immediately starting to trade it. I don't think we figured out good mechanisms for that. I think there are smart things to do: reserving tokens, trying to support the market.

I think there will be investment banks built around concepts like this, but I also think that that latter class is going to be so rare because I think that what will happen is that the regulators are going to say these are almost all securities; you can't just sell these tokens to anyone on the public markets; you've got to treat them properly.

And when you treat them properly, there are built-in protections against price fluctuations early on because they're lock-ups on securities. Once you bought them, awesome! Thank you so much, Andy! Thank you!

Okay! Ten-minute break; grab some coffee! Arrest you! Or they're still eating. I can hear them in there. All right! Who thinks we should dedicate more time to Q&A in these things? Anyone? Is it about right? Okay!

I feel like that Q&A could have gone on for another 45 minutes easily—there are a lot of questions around ICOs and the future, but I guess that's natural!

Okay, people are filtering back in, so as we're concluding the course in start-up investing we're gonna focus on probably the most important aspect of being a great angel investor, and that's how to be good.

Interestingly, it's not a neutral question—it’s the most important question that you're going to have to ask yourself as you have decisions to make as you do investments and things happen. Things always happen; sometimes not great things—and you have to decide how you're going to act.

My good friend and partner, Aaron Harris, is going to walk us through what it means to be a good investor. He knows a lot about this; he has started companies, he has been an investor himself many times, and he's been a YC partner for five years now; is that right? Five years!

So he has seen thousands of companies, like the rest of us. Aaron, you should note, as you listen to him, has said that experts are generally right until they're wrong.

Aaron, thanks Jeff! Thankfully I don't consider myself an expert, so I don't know if that means I'm right or wrong more. But all right, let’s get this started.

So when Jeff and I talked about this idea of doing a session on how to be a good investor, I thought about it for a while and realized there's actually a definitional challenge here in questioning what the word good actually means in the context of being an investor.

And one of these is really obvious—you think of good investors; you think of people who make lots and lots of money, right? You generate returns. That’s the goal of being an investor; that's what we expect when we hear someone is a good investor. But in startups, particularly there’s another definition that I'd argue is actually more important than just generating returns, and that's in your reputation.

Now, it seems a little strange that reputation is something we think about because when we think about investing, we tend to think about money because you give money to generate returns—that's how markets work, right? That's taking risk. But reputation actually drives your returns when it comes to investing in startups.

And here's why: in public markets, it doesn't matter who you are; you can buy the stock. This is why a lot of hedge fund bosses are... doesn't matter—they make lots and lots of money no matter what they do. And so you hear ridiculous stories about the things that they do and the way that they mistreat people.

I know a little bit about this because I worked on Wall Street for four years, and I can tell you that inside of a standard investment bank where many of my friends worked, and where I worked, we never really thought of the people we worked for as people we liked or wanted to hang out with; they were simply good at doing deals and making money.

And the same is true across the hedge fund world, but it's not true when it comes to startups. And that's because this isn't an open market, and the supply and demand of investors to good startups aren't matched in an opening exchange. There isn’t nowhere you can go and buy access at your discretion in any startup that comes out of YC.

Now, obviously, this is changing a little bit when it comes to ICOs, but even in ICOs, if you want to get in on the presale or the pre-presale or the pre-pre-pre-pre-pre-sale, you're going to have to know someone.

And I've watched the way these networks have developed, even in that space. And what you see happening is people with good reputations get the best prices and get in on the best and most oversubscribed deals.

This is particularly true for hot companies—for the good startups. The good startups... this is a little tricky because, as people have talked about this week, you don't really know who the good ones are ahead of time, and often a rush of investors towards a startup is a contra-indicator of that startup's long-term success.

However, you have to go with what you can see at any given point in time, and if there is a deal that you want to get into, you need to find ways to advantage yourself to get into that deal. And that's why you need your advantage when it comes to investing in companies.

And I think that that advantage is primarily a function of how good your reputation is here! This is how this kind of works—being good is directly correlated to your deal flow.

So, let's say you get your first investment—what's your job with that investment as an angel? Well, because again, these aren't massive public companies that are immune to your help, your goal as an angel investor is to improve the likelihood of that startup's success. Can you help that company be really, really, really good?

Yeah! I forgot to advance that; sorry! If you can do this, if you can help companies do better, the founders will talk about you. There just aren't that many founders in the world who are completely unnecessary, so if there is a rumor out there that you're really helpful, then you'll get more deals.

And if you get more deals, then you'll have more opportunities to help companies succeed. And if you help companies succeed, then you'll see more deals. It's a recursive loop! As you do this more, you get better at it, and more companies talk about you.

Of course, the opposite is true—if you screw this pattern up, if you're seen as someone who hurts the companies, then your deal flow will dry up. I know this very, very well because I had investors who were not very good to me, who did things that I would consider to be bad—they don’t get many invitations to demo day.

I think L. Odd said yesterday that there was an investor who like tried to screw him on his acquisition or his sale of his company, and Jeff was his quote, "I'll never work with him again."

Yeah! Founders have long memories, and startups are a small world. So remember this: this thing reinforces on itself in either direction, and it can happen very, very quickly.

So if you want to do this thing—if you want to be good, if you want to increase the likelihood of success of companies and get more deals, how do you do it? In my mind, there are four places at which you have the opportunity to demonstrate that you are good to a founder through the course of and through an investment.

And those four stages are: before you make the investment, which is in sourcing your deals; during when you're meeting the founders for the first time and negotiating the terms of your investment; as you transition from being a pursuer to an actual investor; or from an investor into a shareholder, which is closing the deal.

And then after the deal is closed, which is your ongoing relationship with the company. At each of these places, you have the opportunity to become helpful and to become good.

So let’s go through these one by one—how do you source deals? Well, what do you do? It's important to find out good places to locate investments. I highly recommend demo day. It will bring about 150 companies all in one day. You can do all of your sourcing for six months at one demo day!

But you know that you want to try to get an edge in that relationship. And so what a lot of people do is they'll start emailing founders. They'll find a list; they'll find a founder who knows a founder who knows a founder who knows a founder whose brother heard that a founder was in YC, and they'll email that person.

And maybe that founder won't respond because they're busy. And then the investor has a choice, and it's a pretty stark choice. They can either, A, respect the founder's decision that they don't want to talk right now and politely email back and say, "Hey, I know you're busy. Is it okay if I keep following up and find a time when you're not so busy?" To me, that's great.

Or they can start pestering the founder—they can start stalking the founder and showing up at coffee shops. They can drive around the circle here on Pioneer Way, hoping that a founder comes out of the building such that they can talk to them.

These things have happened. We've had investors literally hanging around outside the door in the hopes that they'll find some founder that they want to meet. Don't do that! Not only is it creepy, it's unlikely to endear you to the founder in a way that will get you the deal that you want.

And this is weird because you think, "Oh, I need to be persistent! I need to do everything! If I do everything, I'll win! I'll get the deal!" That's not how it works. When the company is in need of capital, they tend to go looking for it.

If you do end up sourcing a deal and you make that first contact and you go and have a conversation with someone, there are some pretty easy etiquette rules when you meet with the founder. The first is you need to respect their time. Do not ask for a three-hour meeting for a new company. If the founder says, “Yes, I can meet,” keep it to half an hour.

If the relationship is good, if things are going well, then you can make it more than that. If you go out for coffee or a meal, you pay! My friend Bubba Maracas, who was one of my angel investors and has also been a professional investor for quite a while, gave me a great rule of thumb: if an investor meets with a founder, the founder pays; if two investors go out, whoever has the higher AUM pays!

I think it’s a really great rule of thumb, and I use it all the time! And you should do this because it doesn't matter whether or not the founder is wealthy or poor. It doesn't matter whether or not the founder owns the coffee shop you're in. It doesn't matter if the founder bought really nice avocado toast and all you got was a cup of tea. You are trying to be good to this founder and show that you want to help them.

And if you're discussing the idea of giving that founder five, ten, fifteen hundred thousand dollars—what the hell is the price of a cup of coffee, right? Go the extra distance to show that you want to help that founder. They will remember it, and they'll tell stories about you.

If you don't, now meeting, and negotiating again? Respect the founder's time, right? So in that meeting, when you're talking to them, as that conversation develops, you're gonna ask questions, and they're gonna have the answers to some of the questions—and they're not going to have the answers to other questions. That's okay!

In all likelihood, it's a brand new company; they don't know all the answers. And if you expect them to know everything, you will be sorely disappointed. And if the founder says “No” to a question or says “I don’t know,” it’s okay to ask once or twice in slightly different ways to see if you can get the information, but don't ask 15 times.

This is a way to certainly annoy the founder and probably push on something that they might wish they knew but might be completely irrelevant. When you're speaking to that founder, focus on the founder! This is really hard to do in something like a cocktail hour after demo day where there's all these other founders going around! Any of them might be starting a multi-billion dollar company.

And so what you do is you're talking to someone; you're talking, you're talking—who's that over there? Can I... and you sort of start walking away as the person is talking to you. This is just basic human conversation stuff, which people often get wrong. Don't do that! Focus on the person you're talking to. Give them your time because they might be an incredible founder, and you might want to get into the deal.

And they will remember if you were a jerk to them. Now, when you're discussing what it is you want to invest, you need to understand the terms—understand what you're asking for, what it means, and what’s okay. One of the challenges here is that a lot of this stuff is pretty opaque, which is why we're doing this investor school.

We're talking about what each of the terms in a SAFE means, what a cap is, what a liquidation preference is. By the way, don't ask for those in angel investments! I saw an investor just last week agree to terms with a founder—a standard SAFE—they were going through the motions, okay, sending back and forth—oh, we send you a SAFE—okay, great, great, great.

And then the investor sent back an email, “Hey, just one last thing,” and listed ten demands, including a board seat, 2x participating preferred, dragged-along rights, a right to block a sale—I can't remember the other ones. And the founder came to me and said, “What do we do? We had an agreement, you know, for all this money which we do.”

I said walk away! There are other investors out there! And this is a bad investor!

And, your friends, right? It's pretty open secret that YC has an investor database. This information goes into the investor database, and those investors don't get a lot of deal flow afterward.

When you're in this negotiation, it is natural for one side to have more leverage than the other—that's fine; that's how negotiations work! Understand where the balance of leverage is. Use it wisely and don't push it too far.

This is something that will come up again and again in your relationships with founders. There are always unexplored pockets of probability that you could get if you pushed beyond what seems appropriate.

It seems like the right decision in the micro; it's the wrong long-term decision because you're trying to do more than one deal over time, and it is unlikely that any small thing you do—that a tiny, tiny bit more ownership that you get in one deal is going to be the difference between greatness and badness over the arc of your career as an angel investor.

Always negotiate in good faith. If you say you're gonna do something, do it! And if you are negotiating with someone, do it with the intention that you're actually negotiating for a reason.

One of the worst things that investors do—and we tell all of our founders about this—is to avoid investors who are negotiating to the point where they can find out if other people are investing.

And this is this weird thing: they'll say, “Oh yeah, you know, I want to invest, and that this term in this term—oh, who else is investing? Nah, it's cool—no, just tell me! Like, what else can you tell me?”

And they will only make a decision if they see a lot of other investors have come in. This is a terrible idea for two reasons: one, groupthink usually leads to bad decisions, and two—As I've seen again and again over the last few years—some of the best companies that have come out of YC’s demo day have been the ones that didn't attract all of the investor interest.

And it's the investors who think differently and go against the curve to find the things that they genuinely find interesting, regardless of what other people do. They find not only companies that are going to do incredibly well—they get in at better prices, and they get more ownership. And we remember that! Remember the smart founders!

I look back, and I say, "Oh wow, those guys are a lot smarter than I thought," and I start to send them things. I say, “Hey, you should look at this one; it's a lot like this other deal that you did.” Other people are not smart enough to look at this, but you are!

That's great! That's the position you want to be in. And finally, when you do finish a negotiation and you do agree to something, I don't care if it's a handshake; I don't care if it's verbal; I don't care if it's in an email or a contract—your word is your bond!

This is something we drill into our founders—if you agree to something, that agreement is final. I understand that there can be circumstances in which something wasn't revealed, which there’s material new information that changes the agreement; that’s okay,