October 2010After barely changing at all for decades, the startup fundingbusiness is now in what could, at least by comparison, be calledturmoil. At Y Combinator we've seen dramatic changes in the fundingenvironment for startups. Fortunately one of them is much highervaluations.The trends we've been seeing are probably not YC-specific. I wishI could say they were, but the main cause is probably just that wesee trends first—partly because the startups we fund are veryplugged into the Valley and are quick to take advantage of anythingnew, and partly because we fund so many that we have enough datapoints to see patterns clearly.What we're seeing now, everyone's probably going to be seeing inthe next couple years. So I'm going to explain what we're seeing,and what that will mean for you if you try to raise money.Super-AngelsLet me start by describing what the world of startup funding usedto look like. There used to be two sharply differentiated typesof investors: angels and venture capitalists. Angels are individualrich people who invest small amounts of their own money, while VCsare employees of funds that invest large amounts of other people's.For decades there were just those two types of investors, but nowa third type has appeared halfway between them: the so-calledsuper-angels. [1] And VCs have been provoked by their arrivalinto making a lot of angel-style investments themselves. So thepreviously sharp line between angels and VCs has become hopelesslyblurred.There used to be a no man's land between angels and VCs. Angelswould invest $20k to $50k apiece, and VCs usually a million or more.So an angel round meant a collection of angel investments thatcombined to maybe $200k, and a VC round meant a series A round inwhich a single VC fund (or occasionally two) invested $1-5 million.The no man's land between angels and VCs was a very inconvenientone for startups, because it coincided with the amount many wantedto raise. Most startups coming out of Demo Day wanted to raisearound $400k. But it was a pain to stitch together that much outof angel investments, and most VCs weren't interested in investmentsso small. That's the fundamental reason the super-angels haveappeared. They're responding to the market.The arrival of a new type of investor is big news for startups,because there used to be only two and they rarely competed with oneanother. Super-angels compete with both angels and VCs. That'sgoing to change the rules about how to raise money. I don't knowyet what the new rules will be, but it looks like most of the changeswill be for the better.A super-angel has some of the qualities of an angel, and some ofthe qualities of a VC. They're usually individuals, like angels.In fact many of the current super-angels were initially angels ofthe classic type. But like VCs, they invest other people's money.This allows them to invest larger amounts than angels: a typicalsuper-angel investment is currently about $100k. They make investmentdecisions quickly, like angels. And they make a lot more investmentsper partner than VCs—up to 10 times as many.The fact that super-angels invest other people's money makes themdoubly alarming to VCs. They don't just compete for startups; theyalso compete for investors. What super-angels really are is a newform of fast-moving, lightweight VC fund. And those of us in thetechnology world know what usually happens when something comesalong that can be described in terms like that. Usually it's thereplacement.Will it be? As of now, few of the startups that take money fromsuper-angels are ruling out taking VC money. They're just postponingit. But that's still a problem for VCs. Some of the startups thatpostpone raising VC money may do so well on the angel money theyraise that they never bother to raise more. And those who do raiseVC rounds will be able to get higher valuations when they do. Ifthe best startups get 10x higher valuations when they raise seriesA rounds, that would cut VCs' returns from winners at least tenfold.[2]So I think VC funds are seriously threatened by the super-angels.But one thing that may save them to some extent is the unevendistribution of startup outcomes: practically all the returns areconcentrated in a few big successes. The expected value of a startupis the percentage chance it's Google. So to the extent that winningis a matter of absolute returns, the super-angels could win practicallyall the battles for individual startups and yet lose the war, ifthey merely failed to get those few big winners. And there's achance that could happen, because the top VC funds have betterbrands, and can also do more for their portfolio companies. [3]Because super-angels make more investments per partner, they haveless partner per investment. They can't pay as much attention toyou as a VC on your board could. How much is that extra attentionworth? It will vary enormously from one partner to another. There'sno consensus yet in the general case. So for now this is somethingstartups are deciding individually.Till now, VCs' claims about how much value they added were sort oflike the government's. Maybe they made you feel better, but youhad no choice in the matter, if you needed money on the scale onlyVCs could supply. Now that VCs have competitors, that's going toput a market price on the help they offer. The interesting thingis, no one knows yet what it will be.Do startups that want to get really big need the sort of advice andconnections only the top VCs can supply? Or would super-angel moneydo just as well? The VCs will say you need them, and the super-angelswill say you don't. But the truth is, no one knows yet, not eventhe VCs and super-angels themselves. All the super-angels knowis that their new model seems promising enough to be worth trying,and all the VCs know is that it seems promising enough to worryabout.RoundsWhatever the outcome, the conflict between VCs and super-angels isgood news for founders. And not just for the obvious reason thatmore competition for deals means better terms. The whole shape ofdeals is changing.One of the biggest differences between angels and VCs is the amountof your company they want. VCs want a lot. In a series A roundthey want a third of your company, if they can get it. They don'tcare much how much they pay for it, but they want a lot because thenumber of series A investments they can do is so small. In atraditional series A investment, at least one partner from the VCfund takes a seat on your board. [4] Since board seats last about5 years and each partner can't handle more than about 10 at once,that means a VC fund can only do about 2 series A deals per partnerper year. And that means they need to get as much of the companyas they can in each one. You'd have to be a very promising startupindeed to get a VC to use up one of his 10 board seats for only afew percent of you.Since angels generally don't take board seats, they don't have thisconstraint. They're happy to buy only a few percent of you. Andalthough the super-angels are in most respects mini VC funds, they'veretained this critical property of angels. They don't take boardseats, so they don't need a big percentage of your company.Though that means you'll get correspondingly less attention fromthem, it's good news in other respects. Founders never really likedgiving up as much equity as VCs wanted. It was a lot of the companyto give up in one shot. Most founders doing series A deals wouldprefer to take half as much money for half as much stock, and thensee what valuation they could get for the second half of the stockafter using the first half of the money to increase its value. ButVCs never offered that option.Now startups have another alternative. Now it's easy to raise angelrounds about half the size of series A rounds. Many of the startupswe fund are taking this route, and I predict that will be true ofstartups in general.A typical big angel round might be $600k on a convertible note witha valuation cap of $4 million premoney. Meaning that when the noteconverts into stock (in a later round, or upon acquisition), theinvestors in that round will get .6 / 4.6, or 13% of the company.That's a lot less than the 30 to 40% of the company you usuallygive up in a series A round if you do it so early. [5]But the advantage of these medium-sized rounds is not just thatthey cause less dilution. You also lose less control. After anangel round, the founders almost always still have control of thecompany, whereas after a series A round they often don't. Thetraditional board structure after a series A round is two founders,two VCs, and a (supposedly) neutral fifth person. Plus series Aterms usually give the investors a veto over various kinds ofimportant decisions, including selling the company. Founders usuallyhave a lot of de facto control after a series A, as long as thingsare going well. But that's not the same as just being able to dowhat you want, like you could before.A third and quite significant advantage of angel rounds is thatthey're less stressful to raise. Raising a traditional series Around has in the past taken weeks, if not months. When a VC firmcan only do 2 deals per partner per year, they're careful aboutwhich they do. To get a traditional series A round you have to gothrough a series of meetings, culminating in a full partner meetingwhere the firm as a whole says yes or no. That's the really scarypart for founders: not just that series A rounds take so long, butat the end of this long process the VCs might still say no. Thechance of getting rejected after the full partner meeting averagesabout 25%. At some firms it's over 50%.Fortunately for founders, VCs have been getting a lot faster.Nowadays Valley VCs are more likely to take 2 weeks than 2 months.But they're still not as fast as angels and super-angels, the mostdecisive of whom sometimes decide in hours.Raising an angel round is not only quicker, but you get feedbackas it progresses. An angel round is not an all or nothing thinglike a series A. It's composed of multiple investors with varyingdegrees of seriousness, ranging from the upstanding ones who commitunequivocally to the jerks who give you lines like "come back tome to fill out the round." You usually start collecting money fromthe most committed investors and work your way out toward theambivalent ones, whose interest increases as the round fills up.But at each point you know how you're doing. If investors turncold you may have to raise less, but when investors in an angelround turn cold the process at least degrades gracefully, insteadof blowing up in your face and leaving you with nothing, as happensif you get rejected by a VC fund after a full partner meeting.Whereas if investors seem hot, you can not only close the roundfaster, but now that convertible notes are becoming the norm,actually raise the price to reflect demand.ValuationHowever, the VCs have a weapon they can use against the super-angels,and they have started to use it. VCs have started making angel-sizedinvestments too. The term "angel round" doesn't mean that all theinvestors in it are angels; it just describes the structure of theround. Increasingly the participants include VCs making investmentsof a hundred thousand or two. And when VCs invest in angel roundsthey can do things that super-angels don't like. VCs are quitevaluation-insensitive in angel rounds—partly because they arein general, and partly because they don't care that much about thereturns on angel rounds, which they still view mostly as a way torecruit startups for series A rounds later. So VCs who invest inangel rounds can blow up the valuations for angels and super-angelswho invest in them. [6]Some super-angels seem to care about valuations. Several turneddown YC-funded startups after Demo Day because their valuationswere too high. This was not a problem for the startups; by definitiona high valuation means enough investors were willing to accept it.But it was mysterious to me that the super-angels would quibbleabout valuations. Did they not understand that the big returnscome from a few big successes, and that it therefore mattered farmore which startups you picked than how much you paid for them?After thinking about it for a while and observing certain othersigns, I have a theory that explains why the super-angels may besmarter than they seem. It would make sense for super-angels towant low valuations if they're hoping to invest in startups thatget bought early. If you're hoping to hit the next Google, youshouldn't care if the valuation is 20 million. But if you're lookingfor companies that are going to get bought for 30 million, you care.If you invest at 20 and the company gets bought for 30, you onlyget 1.5x. You might as well buy Apple.So if some of the super-angels were looking for companies that couldget acquired quickly, that would explain why they'd care aboutvaluations. But why would they be looking for those? Becausedepending on the meaning of "quickly," it could actually be veryprofitable. A company that gets acquired for 30 million is a failureto a VC, but it could be a 10x return for an angel, and moreover,a quick 10x return. Rate of return is what matters ininvesting—not the multiple you get, but the multiple per year.If a super-angel gets 10x in one year, that's a higher rate ofreturn than a VC could ever hope to get from a company that took 6years to go public. To get the same rate of return, the VC wouldhave to get a multiple of 10^6—one million x. Even Googledidn't come close to that.So I think at least some super-angels are looking for companiesthat will get bought. That's the only rational explanation forfocusing on getting the right valuations, instead of the rightcompanies. And if so they'll be different to deal with than VCs.They'll be tougher on valuations, but more accommodating if you wantto sell early.PrognosisWho will win, the super-angels or the VCs? I think the answer tothat is, some of each. They'll each become more like one another.The super-angels will start to invest larger amounts, and the VCswill gradually figure out ways to make more, smaller investmentsfaster. A decade from now the players will be hard to tell apart,and there will probably be survivors from each group.What does that mean for founders? One thing it means is that thehigh valuations startups are presently getting may not last forever.To the extent that valuations are being driven up by price-insensitiveVCs, they'll fall again if VCs become more like super-angels andstart to become more miserly about valuations. Fortunately if thisdoes happen it will take years.The short term forecast is more competition between investors, whichis good news for you. The super-angels will try to undermine theVCs by acting faster, and the VCs will try to undermine thesuper-angels by driving up valuations. Which for founders willresult in the perfect combination: funding rounds that close fast,with high valuations.But remember that to get that combination, your startup will haveto appeal to both super-angels and VCs. If you don't seem like youhave the potential to go public, you won't be able to use VCs todrive up the valuation of an angel round.There is a danger of having VCs in an angel round: the so-calledsignalling risk. If VCs are only doing it in the hope of investingmore later, what happens if they don't? That's a signal to everyoneelse that they think you're lame.How much should you worry about that? The seriousness of signallingrisk depends on how far along you are. If by the next time youneed to raise money, you have graphs showing rising revenue ortraffic month after month, you don't have to worry about any signalsyour existing investors are sending. Your results will speak forthemselves. [7]Whereas if the next time you need to raise money you won't yet haveconcrete results, you may need to think more about the message yourinvestors might send if they don't invest more. I'm not sure yethow much you have to worry, because this whole phenomenon of VCsdoing angel investments is so new. But my instincts tell me youdon't have to worry much. Signalling risk smells like one of thosethings founders worry about that's not a real problem. As a rule,the only thing that can kill a good startup is the startup itself.Startups hurt themselves way more often than competitors hurt them,for example. I suspect signalling risk is in this category too.One thing YC-funded startups have been doing to mitigate the riskof taking money from VCs in angel rounds is not to take too muchfrom any one VC. Maybe that will help, if you have the luxury ofturning down money.Fortunately, more and more startups will. After decades of competitionthat could best be described as intramural, the startup fundingbusiness is finally getting some real competition. That shouldlast several years at least, and maybe a lot longer. Unless there'ssome huge market crash, the next couple years are going to be agood time for startups to raise money. And that's exciting becauseit means lots more startups will happen.Notes[1]I've also heard them called "Mini-VCs" and "Micro-VCs." Idon't know which name will stick.There were a couple predecessors. Ron Conway had angel fundsstarting in the 1990s, and in some ways First Round Capital is closer to asuper-angel than a VC fund.[2]It wouldn't cut their overall returns tenfold, because investinglater would probably (a) cause them to lose less on investmentsthat failed, and (b) not allow them to get as large a percentageof startups as they do now. So it's hard to predict precisely whatwould happen to their returns.[3]The brand of an investor derives mostly from the success oftheir portfolio companies. The top VCs thus have a big brandadvantage over the super-angels. They could make it self-perpetuatingif they used it to get all the best new startups. But I don't thinkthey'll be able to. To get all the best startups, you have to domore than make them want you. You also have to want them; you haveto recognize them when you see them, and that's much harder.Super-angels will snap up stars that VCs miss. And that will causethe brand gap between the top VCs and the super-angels graduallyto erode.[4]Though in a traditional series A round VCs put two partnerson your board, there are signs now that VCs may begin to conserveboard seats by switching to what used to be considered an angel-roundboard, consisting of two founders and one VC. Which is also to thefounders' advantage if it means they still control the company.[5]In a series A round, you usually have to give up more thanthe actual amount of stock the VCs buy, because they insist youdilute yourselves to set aside an "option pool" as well. I predictthis practice will gradually disappear though.[6]The best thing for founders, if they can get it, is a convertiblenote with no valuation cap at all. In that case the money investedin the angel round just converts into stock at the valuation of thenext round, no matter how large. Angels and super-angels tend notto like uncapped notes. They have no idea how much of the companythey're buying. If the company does well and the valuation of thenext round is high, they may end up with only a sliver of it. Soby agreeing to uncapped notes, VCs who don't care about valuationsin angel rounds can make offers that super-angels hate to match.[7]Obviously signalling risk is also not a problem if you'llnever need to raise more money. But startups are often mistakenabout that.Thanks to Sam Altman, John Bautista, Patrick Collison, JamesLindenbaum, Reid Hoffman, Jessica Livingston and Harj Taggarfor reading draftsof this.
