April 2007(This essay is derived from a keynote talk at the 2007 ASES Summitat Stanford.)The world of investors is a foreign one to most hackers—partlybecause investors are so unlike hackers, and partly because theytend to operate in secret. I've been dealing with this world formany years, both as a founder and an investor, and I still don'tfully understand it.In this essay I'm going to list some of the more surprising thingsI've learned about investors. Some I only learned in the past year.Teaching hackers how to deal with investors is probably the secondmost important thing we do at Y Combinator. The most importantthing for a startup is to make something good. But everyone knowsthat's important. The dangerous thing about investors is thathackers don't know how little they know about this strange world.1. The investors are what make a startup hub.About a year ago I tried to figure out what you'd need to reproduceSilicon Valley. I decided the critical ingredients were rich peopleand nerds—investors and founders. People are all you need tomake technology, and all the other people will move.If I had to narrow that down, I'd say investors are the limitingfactor. Not because they contribute more to the startup, but simplybecause they're least willing to move. They're rich. They're notgoing to move to Albuquerque just because there are some smarthackers there they could invest in. Whereas hackers will move tothe Bay Area to find investors.2. Angel investors are the most critical.There are several types of investors. The two main categories areangels and VCs: VCs invest other people's money, and angels investtheir own.Though they're less well known, the angel investors are probablythe more critical ingredient in creating a silicon valley. Mostcompanies that VCs invest in would never have made it that far if angelshadn't invested first. VCs say between half and three quarters ofcompanies that raise series A rounds have taken some outsideinvestment already.[1]Angels are willing to fund riskier projects than VCs. They alsogive valuable advice, because (unlike VCs) many have been startupfounders themselves.Google's story shows the key role angels play. A lot of people knowGoogle raised money from Kleiner and Sequoia. What most don't realizeis how late. That VC round was a series B round; the premoneyvaluation was $75 million. Google was already a successful companyat that point. Really, Google was funded with angel money.It may seem odd that the canonical Silicon Valley startup was fundedby angels, but this is not so surprising. Risk is always proportionateto reward. So the most successful startup of all is likely to haveseemed an extremely risky bet at first, and that is exactly thekind VCs won't touch.Where do angel investors come from? From other startups. So startuphubs like Silicon Valley benefit from something like the marketplaceeffect, but shifted in time: startups are there because startupswere there.3. Angels don't like publicity.If angels are so important, why do we hear more about VCs? BecauseVCs like publicity. They need to market themselves to the investorswho are their "customers"—the endowments and pension funds andrich families whose money they invest—and also to founders whomight come to them for funding.Angels don't need to market themselves to investors because theyinvest their own money. Nor do they want to market themselves tofounders: they don't want random people pestering them with businessplans. Actually, neither do VCs. Both angels and VCs get dealsalmost exclusively through personal introductions. [2]The reason VCs want a strong brand is not to draw in more businessplans over the transom, but so they win deals when competingagainst other VCs. Whereas angels are rarely in direct competition,because (a) they do fewer deals, (b) they're happy to split them,and (c) they invest at a point where the stream is broader.4. Most investors, especially VCs, are not like founders.Some angels are, or were, hackers. But most VCs are a differenttype of people: they're dealmakers.If you're a hacker, here's a thought experiment you can run tounderstand why there are basically no hacker VCs: How would youlike a job where you never got to make anything, but instead spentall your time listening to other people pitch (mostly terrible)projects, deciding whether to fund them, and sitting on their boardsif you did? That would not be fun for most hackers. Hackers liketo make things. This would be like being an administrator.Because most VCs are a different species of people fromfounders, it's hard to know what they're thinking. If you're ahacker, the last time you had to deal with these guys was in highschool. Maybe in college you walked past their fraternity on yourway to the lab. But don't underestimate them. They're as expertin their world as you are in yours. What they're good at is readingpeople, and making deals work to their advantage. Think twicebefore you try to beat them at that.5. Most investors are momentum investors.Because most investors are dealmakers rather than technology people,they generally don't understand what you're doing. I knew as afounder that most VCs didn't get technology. I also knew some madea lot of money. And yet it never occurred to me till recently toput those two ideas together and ask "How can VCs make money byinvesting in stuff they don't understand?"The answer is that they're like momentum investors. You can (orcould once) make a lot of money by noticing sudden changes in stockprices. When a stock jumps upward, you buy, and when it suddenlydrops, you sell. In effect you're insider trading, without knowingwhat you know. You just know someone knows something, and that'smaking the stock move.This is how most venture investors operate. They don't try to lookat something and predict whether it will take off. They win bynoticing that something is taking off a little sooner than everyoneelse. That generates almost as good returns as actually being ableto pick winners. They may have to pay a little more than they wouldif they got in at the very beginning, but only a little.Investors always say what they really care about is the team.Actually what they care most about is your traffic, then what otherinvestors think, then the team. If you don't yet have any traffic,they fall back on number 2, what other investors think. And this,as you can imagine, produces wild oscillations in the "stock price"of a startup. One week everyone wants you, and they're begging notto be cut out of the deal. But all it takes is for one big investorto cool on you, and the next week no one will return your phonecalls. We regularly have startups go from hot to cold or cold tohot in a matter of days, and literally nothing has changed.There are two ways to deal with this phenomenon. If you're feelingreally confident, you can try to ride it. You can start by askinga comparatively lowly VC for a small amount of money, and then aftergenerating interest there, ask more prestigious VCs for largeramounts, stirring up a crescendo of buzz, and then "sell" at thetop. This is extremely risky, and takes months even if you succeed.I wouldn't try it myself. My advice is to err on the side of safety:when someone offers you a decent deal, just take it and get on withbuilding the company. Startups win or lose based on the qualityof their product, not the quality of their funding deals.6. Most investors are looking for big hits.Venture investors like companies that could go public. That's wherethe big returns are. They know the odds of any individual startupgoing public are small, but they want to invest in those that atleast have a chance of going public.Currently the way VCs seem to operate is to invest in a bunch ofcompanies, most of which fail, and one of which is Google. Thosefew big wins compensate for losses on their other investments. What thismeans is that most VCs will only invest in you if you're a potentialGoogle. They don't care about companies that are a safe bet to beacquired for $20 million. There needs to be a chance, howeversmall, of the company becoming really big.Angels are different in this respect. They're happy to invest ina company where the most likely outcome is a $20 million acquisitionif they can do it at a low enough valuation. But of course theylike companies that could go public too. So having an ambitiouslong-term plan pleases everyone.If you take VC money, you have to mean it, because the structureof VC deals prevents early acquisitions. If you take VC money,they won't let you sell early.7. VCs want to invest large amounts.The fact that they're running investment funds makes VCs want toinvest large amounts. A typical VC fund is now hundreds of millionsof dollars. If $400 million has to be invested by 10 partners,they have to invest $40 million each. VCs usually sit on the boardsof companies they fund. If the average deal size was $1 million,each partner would have to sit on 40 boards, which would not befun. So they prefer bigger deals, where they can put a lot of moneyto work at once.VCs don't regard you as a bargain if you don't need a lot of money.That may even make you less attractive, because it means theirinvestment creates less of a barrier to entry for competitors.Angels are in a different position because they're investing theirown money. They're happy to invest small amounts—sometimes aslittle as $20,000—as long as the potential returns look goodenough. So if you're doing something inexpensive, go to angels.8. Valuations are fiction.VCs admit that valuations are an artifact. They decide how muchmoney you need and how much of the company they want, and those twoconstraints yield a valuation.Valuations increase as the size of the investment does. A companythat an angel is willing to put $50,000 into at a valuation of amillion can't take $6 million from VCs at that valuation. Thatwould leave the founders less than a seventh of the company betweenthem (since the option pool would also come out of that seventh).Most VCs wouldn't want that, which is why you never hear of dealswhere a VC invests $6 million at a premoney valuation of $1 million.If valuations change depending on the amount invested, that showshow far they are from reflecting any kind of value of the company.Since valuations are made up, founders shouldn't care too much aboutthem. That's not the part to focus on. In fact, a high valuationcan be a bad thing. If you take funding at a premoney valuationof $10 million, you won't be selling the company for 20. You'llhave to sell for over 50 for the VCs to get even a 5x return, whichis low to them. More likely they'll want you to hold out for 100.But needing to get a high price decreases the chance of gettingbought at all; many companies can buy you for $10 million, but onlya handful for 100. And since a startup is like a pass/fail coursefor the founders, what you want to optimize is your chance of agood outcome, not the percentage of the company you keep.So why do founders chase high valuations? They're tricked bymisplaced ambition. They feel they've achieved more if they get ahigher valuation. They usually know other founders, and if theyget a higher valuation they can say "mine is bigger than yours."But funding is not the real test. The real test is the final outcomefor the founder, and getting too high a valuation may just make agood outcome less likely.The one advantage of a high valuation is that you get less dilution.But there is another less sexy way to achieve that: just take lessmoney.9. Investors look for founders like the current stars.Ten years ago investors were looking for the next Bill Gates. Thiswas a mistake, because Microsoft was a very anomalous startup. Theystarted almost as a contract programming operation, and the reasonthey became huge was that IBM happened to drop the PC standard intheir lap.Now all the VCs are looking for the next Larry and Sergey. Thisis a good trend, because Larry and Sergey are closer to the idealstartup founders.Historically investors thought it was important for a founder tobe an expert in business. So they were willing to fund teams ofMBAs who planned to use the money to pay programmers to build theirproduct for them. This is like funding Steve Ballmer in the hopethat the programmer he'll hire is Bill Gates—kind of backward,as the events of the Bubble showed. Now most VCs know they shouldbe funding technical guys. This is more pronounced among the verytop funds; the lamer ones still want to fund MBAs.If you're a hacker, it's good news that investors are looking forLarry and Sergey. The bad news is, the only investors who can doit right are the ones who knew them when they were acouple of CS grad students, not the confident media stars they aretoday. What investors still don't get is how clueless and tentativegreat founders can seem at the very beginning.10. The contribution of investors tends to be underestimated.Investors do more for startups than give them money. They're helpfulin doing deals and arranging introductions, and some of the smarterones, particularly angels, can give good advice about the product.In fact, I'd say what separates the great investors from the mediocreones is the quality of their advice. Most investors give advice,but the top ones give good advice.Whatever help investors give a startup tends to be underestimated.It's to everyone's advantage to let the world think the foundersthought of everything. The goal of the investors is for the companyto become valuable, and the company seems more valuable if it seemslike all the good ideas came from within.This trend is compounded by the obsession that the press has withfounders. In a company founded by two people, 10% of the ideasmight come from the first guy they hire. Arguably they've done abad job of hiring otherwise. And yet this guy will be almostentirely overlooked by the press.I say this as a founder: the contribution of founders is alwaysoverestimated. The danger here is that new founders, looking atexisting founders, will think that they're supermen that one couldn'tpossibly equal oneself. Actually they have a hundred differenttypes of support people just offscreen making the whole show possible.[3]11. VCs are afraid of looking bad.I've been very surprised to discover how timid most VCs are. Theyseem to be afraid of looking bad to their partners, and perhapsalso to the limited partners—the people whose money they invest.You can measure this fear in how much less risk VCs are willing totake. You can tell they won't make investments for their fund thatthey might be willing to make themselves as angels. Though it'snot quite accurate to say that VCs are less willing to take risks.They're less willing to do things that might look bad. That's notthe same thing.For example, most VCs would be very reluctant to invest in a startupfounded by a pair of 18 year old hackers, no matter how brilliant,because if the startup failed their partners could turn on them andsay "What, you invested $x million of our money in a pair of 18year olds?" Whereas if a VC invested in a startup founded bythree former banking executives in their 40s who planned to outsourcetheir product development—which to my mind is actually a lotriskier than investing in a pair of really smart 18 year olds—hecouldn't be faulted, if it failed, for making such an apparentlyprudent investment.As a friend of mine said, "Most VCs can't do anything that wouldsound bad to the kind of doofuses who run pension funds." Angelscan take greater risks because they don't have to answer to anyone.12. Being turned down by investors doesn't mean much.Some founders are quite dejected when they get turned down byinvestors. They shouldn't take it so much to heart. To start with,investors are often wrong. It's hard to think of a successfulstartup that wasn't turned down by investors at some point. Lotsof VCs rejected Google. So obviously the reaction of investors isnot a very meaningful test.Investors will often reject you for what seem to be superficialreasons. I read of one VC who turneddown a startup simply becausethey'd given away so many little bits of stock that the deal requiredtoo many signatures to close. [4]The reason investors can get awaywith this is that they see so many deals. It doesn't matter ifthey underestimate you because of some surface imperfection, becausethe next best deal will be almost as good. Imagine picking outapples at a grocery store. You grab one with a little bruise.Maybe it's just a surface bruise, but why even bother checking whenthere are so many other unbruised apples to choose from?Investors would be the first to admit they're often wrong. So whenyou get rejected by investors, don't think "we suck," but insteadask "do we suck?" Rejection is a question, not an answer.13. Investors are emotional.I've been surprised to discover how emotional investors can be.You'd expect them to be cold and calculating, or at least businesslike,but often they're not. I'm not sure if it's their position of powerthat makes them this way, or the large sums of money involved, butinvestment negotiations can easily turn personal. If you offendinvestors, they'll leave in a huff.A while ago an eminent VC firm offered a series A round to a startupwe'd seed funded. Then they heard a rival VC firm was also interested.They were so afraid that they'd be rejected in favor of this otherfirm that they gave the startup what's known as an "explodingtermsheet." They had, I think, 24 hours to say yes or no, or thedeal was off. Exploding termsheets are a somewhat dubious device,but not uncommon. What surprised me was their reaction when Icalled to talk about it. I asked if they'd still be interested inthe startup if the rival VC didn't end up making an offer, and theysaid no. What rational basis could they have had for saying that?If they thought the startup was worth investing in, what differenceshould it make what some other VC thought? Surely it was theirduty to their limited partners simply to invest in the bestopportunities they found; they should be delighted if the other VCsaid no, because it would mean they'd overlooked a good opportunity.But of course there was no rational basis for their decision. Theyjust couldn't stand the idea of taking this rival firm's rejects.In this case the exploding termsheet was not (or not only) a tacticto pressure the startup. It was more like the high school trickof breaking up with someone before they can break up with you. Inan earlier essay I said that VCs were a lot like high school girls.A few VCs have joked about that characterization, but none havedisputed it.14. The negotiation never stops till the closing.Most deals, for investment or acquisition, happen in two phases.There's an initial phase of negotiation about the big questions.If this succeeds you get a termsheet, so called because it outlinesthe key terms of a deal. A termsheet is not legally binding,but it is a definite step. It's supposed to mean that adeal is going to happen, once the lawyers work out all the details.In theory these details are minor ones; by definition all theimportant points are supposed to be covered in the termsheet.Inexperience and wishful thinking combine to make founders feelthat when they have a termsheet, they have a deal. They want thereto be a deal; everyone acts like they have a deal; so there mustbe a deal. But there isn't and may not be for several months. Alot can change for a startup in several months. It's not uncommonfor investors and acquirers to get buyer's remorse. So you haveto keep pushing, keep selling, all the way to the close. Otherwiseall the "minor" details left unspecified in the termsheet will beinterpreted to your disadvantage. The other side may even breakthe deal; if they do that, they'll usually seize on some technicalityor claim you misled them, rather than admitting they changed theirminds.It can be hard to keep the pressure on an investor or acquirer allthe way to the closing, because the most effective pressure iscompetition from other investors or acquirers, and these tend todrop away when you get a termsheet. You should try to stay as closefriends as you can with these rivals, but the most important thingis just to keep up the momentum in your startup. The investors oracquirers chose you because you seemed hot. Keep doing whatevermade you seem hot. Keep releasing new features; keep getting newusers; keep getting mentioned in the press and in blogs.15. Investors like to co-invest.I've been surprised how willing investors are to split deals. Youmight think that if they found a good deal they'd want it all tothemselves, but they seem positively eager to syndicate. This isunderstandable with angels; they invest on a smaller scale and don'tlike to have too much money tied up in any one deal. But VCs alsoshare deals a lot. Why?Partly I think this is an artifact of the rule I quoted earlier:after traffic, VCs care most what other VCs think. A deal that hasmultiple VCs interested in it is more likely to close, so of dealsthat close, more will have multiple investors.There is one rational reason to want multiple VCs in a deal: Anyinvestor who co-invests with you is one less investor who couldfund a competitor. Apparently Kleiner and Sequoia didn't likesplitting the Google deal, but it did at least have the advantage,from each one's point of view, that there probably wouldn't be acompetitor funded by the other. Splitting deals thus has similaradvantages to confusing paternity.But I think the main reason VCs like splitting deals is the fearof looking bad. If another firm shares the deal, then in the eventof failure it will seem to have been a prudent choice—a consensusdecision, rather than just the whim of an individual partner.16. Investors collude.Investing is not covered by antitrust law. At least, it better notbe, because investors regularly do things that would be illegalotherwise. I know personally of cases where one investor has talkedanother out of making a competitive offer, using the promise ofsharing future deals.In principle investors are all competing for the same deals, butthe spirit of cooperation is stronger than the spirit of competition.The reason, again, is that there are so many deals. Though aprofessional investor may have a closer relationship with a founderhe invests in than with other investors, his relationship with thefounder is only going to last a couple years, whereas his relationshipwith other firms will last his whole career. There isn't so muchat stake in his interactions with other investors, but there willbe a lot of them. Professional investors are constantly tradinglittle favors.Another reason investors stick together is to preserve the powerof investors as a whole. So you will not, as of this writing, beable to get investors into an auction for your series A round.They'd rather lose the deal than establish a precedent of VCscompetitively bidding against one another. An efficient startupfunding market may be coming in the distant future; things tend tomove in that direction; but it's certainly not here now.17. Large-scale investors care about their portfolio, not anyindividual company.The reason startups work so well is that everyone with power alsohas equity. The only way any of them can succeed is if they alldo. This makes everyone naturally pull in the same direction,subject to differences of opinion about tactics.The problem is, larger scale investors don't have exactly the samemotivation. Close, but not identical. They don't need any givenstartup to succeed, like founders do, just their portfolio as awhole to. So in borderline cases the rational thing for them todo is to sacrifice unpromising startups.Large-scale investors tend to put startups in three categories:successes, failures, and the "living dead"—companies that areplugging along but don't seem likely in the immediate future to getbought or go public. To the founders, "living dead" sounds harsh.These companies may be far from failures by ordinary standards. Butthey might as well be from a venture investor's point of view, andthey suck up just as much time and attention as the successes. Soif such a company has two possible strategies, a conservative onethat's slightly more likely to work in the end, or a risky one thatwithin a short time will either yield a giant success or kill thecompany, VCs will push for the kill-or-cure option. To them thecompany is already a write-off. Better to have resolution, one wayor the other, as soon as possible.If a startup gets into real trouble, instead of trying to save itVCs may just sell it at a low price to another of their portfoliocompanies. Philip Greenspun said in Founders at Work that Ars Digita's VCs did this to them.18. Investors have different risk profiles from founders.Most people would rather a 100% chance of $1 million than a 20%chance of $10 million. Investors are rich enough to be rationaland prefer the latter. So they'll always tend to encourage foundersto keep rolling the dice. If a company is doing well, investorswill want founders to turn down most acquisition offers. And indeed,most startups that turn down acquisition offers ultimately do better.But it's still hair-raising for the founders, because they mightend up with nothing. When someone's offering to buy you for a priceat which your stock is worth $5 million, saying no is equivalentto having $5 million and betting it all on one spin of the roulettewheel.Investors will tell you the company is worth more. And they maybe right. But that doesn't mean it's wrong to sell. Any financialadvisor who put all his client's assets in the stock of a single,private company would probably lose his license for it.More and more, investors are letting founders cash out partially.That should correct the problem. Most founders have such low standardsthat they'll feel rich with a sum that doesn't seem huge to investors.But this custom is spreading too slowly, because VCs are afraid ofseeming irresponsible. No one wants to be the first VC to givesomeone fuck-you money and then actually get told "fuck you." Butuntil this does start to happen, we know VCs are being too conservative.19. Investors vary greatly.Back when I was a founder I used to think all VCs were the same.And in fact they do all look the same. They're all what hackerscall "suits." But since I've been dealing with VCs more I've learnedthat some suits are smarter than others.They're also in a business where winners tend to keep winning andlosers to keep losing. When a VC firm has been successful in thepast, everyone wants funding from them, so they get the pick of allthe new deals. The self-reinforcing nature of the venture fundingmarket means that the top ten firms live in a completely differentworld from, say, the hundredth. As well as being smarter, theytend to be calmer and more upstanding; they don't need to do iffythings to get an edge, and don't want to because they have morebrand to protect.There are only two kinds of VCs you want to take money from, if youhave the luxury of choosing: the "top tier" VCs, meaning about thetop 20 or so firms, plus a few new ones that are not among the top20 only because they haven't been around long enough.It's particularly important to raise money from a top firm if you'rea hacker, because they're more confident. That means they're lesslikely to stick you with a business guy as CEO, like VCs used todo in the 90s. If you seem smart and want to do it, they'll letyou run the company.20. Investors don't realize how much it costs to raise money fromthem.Raising money is a huge time suck at just the point where startupscan least afford it. It's not unusual for it to take five or sixmonths to close a funding round. Six weeks is fast. And raisingmoney is not just something you can leave running as a backgroundprocess. When you're raising money, it's inevitably the main focusof the company. Which means building the product isn't.Suppose a Y Combinator company starts talking to VCs after demoday, and is successful in raising money from them, closing the dealafter a comparatively short 8 weeks. Since demo day occurs after10 weeks, the company is now 18 weeks old. Raising money, ratherthan working on the product, has been the company's main focus for44% of its existence. And mind you, this an example where thingsturned out well.When a startup does return to working on the product after a fundinground finally closes, it's as if they were returning to work aftera months-long illness. They've lost most of their momentum.Investors have no idea how much they damage the companies theyinvest in by taking so long to do it. But companies do. So thereis a big opportunity here for a new kind of venture fund that investssmaller amounts at lower valuations, but promises to either closeor say no very quickly. If there were such a firm, I'd recommendit to startups in preference to any other, no matter how prestigious.Startups live on speed and momentum.21. Investors don't like to say no.The reason funding deals take so long to close is mainly thatinvestors can't make up their minds. VCs are not big companies;they can do a deal in 24 hours if they need to. But they usuallylet the initial meetings stretch out over a couple weeks. Thereason is the selection algorithm I mentioned earlier. Most don'ttry to predict whether a startup will win, but to notice quicklythat it already is winning. They care what the market thinks ofyou and what other VCs think of you, and they can't judge thosejust from meeting you.Because they're investing in things that (a) change fast and (b)they don't understand, a lot of investors will reject you in a waythat can later be claimed not to have been a rejection. Unless youknow this world, you may not even realize you've been rejected.Here's a VC saying no: We're really excited about your project, and we want to keep in close touch as you develop it further.Translated into more straightforward language, this means: We'renot investing in you, but we may change our minds if it looks likeyou're taking off. Sometimes they're more candid and say explicitlythat they need to "see some traction." They'll invest in you ifyou start to get lots of users. But so would any VC. So all they'resaying is that you're still at square 1.Here's a test for deciding whether a VC's response was yes or no.Look down at your hands. Are you holding a termsheet?22. You need investors.Some founders say "Who needs investors?" Empirically the answerseems to be: everyone who wants to succeed. Practically everysuccessful startup takes outside investment at some point.Why? What the people who think they don't need investors forget isthat they will have competitors. The question is not whether youneed outside investment, but whether it could help you at all.If the answer is yes, and you don't take investment, then competitorswho do will have an advantage over you. And in the startup worlda little advantage can expand into a lot.Mike Moritz famously said that he invested in Yahoo because hethought they had a few weeks' lead over their competitors. Thatmay not have mattered quite so much as he thought, because Googlecame along three years later and kicked Yahoo's ass. But there issomething in what he said. Sometimes a small lead can grow intothe yes half of a binary choice.Maybe as it gets cheaper to start a startup, it will start to bepossible to succeed in a competitive market without outside funding. There are certainlycosts to raising money. But as of this writing the empiricalevidence says it's a net win.23. Investors like it when you don't need them.A lot of founders approach investors as if they needed theirpermission to start a company—as if it were like getting intocollege. But you don't need investors to start most companies;they just make it easier.And in fact, investors greatly prefer it if you don't need them.What excites them, both consciously and unconsciously, is the sortof startup that approaches them saying "the train's leaving thestation; are you in or out?" not the one saying "please can we havesome money to start a company?"Most investors are "bottoms" in the sense that the startups theylike most are those that are rough with them. When Google stuckKleiner and Sequoia with a $75 million premoney valuation, theirreaction was probably "Ouch! That feels so good." And they wereright, weren't they? That deal probably made them more than anyother they've done.The thing is, VCs are pretty good at reading people. So don't tryto act tough with them unless you really are the next Google, orthey'll see through you in a second. Instead of acting tough, whatmost startups should do is simply always have a backup plan. Alwayshave some alternative plan for getting started if any given investorsays no. Having one is the best insurance against needing one.So you shouldn't start a startup that's expensive to start, becausethen you'll be at the mercy of investors. If you ultimately wantto do something that will cost a lot, start by doing a cheapersubset of it, and expand your ambitions when and if you raise moremoney.Apparently the most likely animals to be left alive after a nuclearwar are cockroaches, because they're so hard to kill. That's whatyou want to be as a startup, initially. Instead of a beautifulbut fragile flower that needs to have its stem in a plastic tubeto support itself, better to be small, ugly, and indestructible.Notes[1]I may be underestimating VCs. They may play some behind the scenes role in IPOs, which you ultimately need if you want to create a silicon valley.[2]A few VCs have an email address you can send your businessplan to, but the number of startups that get funded this way isbasically zero. You should always get a personal introduction—and to a partner, not an associate.[3]Several people have told us that the most valuable thing about startup school was that they got to see famous startup founders and realizedthey were just ordinary guys. Though we're happy to provide thisservice, this is not generally the way we pitch startup school topotential speakers.[4]Actually this sounds to me like a VC who got buyer's remorse,then used a technicality to get out of the deal. But it's tellingthat it even seemed a plausible excuse.Thanks to Sam Altman, Paul Buchheit, Hutch Fishman, and Robert Morris for reading drafts ofthis, and to Kenneth King of ASES for inviting me to speak.Comment on this essay.
