Failure as a Tool - Vinod Khosla, Founder of Khosla Ventures

I'm gonna start where we started the conference how many people remember Regina's first slide and what I'd say yes it's a great way I didn't know she was going to use that slide but I'm mostly gonna be the bookend for that I've never understood why they do podiums where you can't put a bottle of water anywhere on the podium so many other people have covered in fact some of the slides you've seen before from other people I had no idea they were planning and using it but it ties down so many of the themes and since I'm last I I get plenty of time it says I'm - eight minutes already that's nice I'll keep you through lunch I have 80 some slides I do honest to god you can't talk about failure simply no but you know condi brought it to the right place the valley is a undirected place when you see whether China and model of directed investment which seems powerful or the Russian planned economy of Fortune 500 companies five-year plans that does not work and I'm gonna talk about why and maybe how to use failure as a tool it may be the single most important tool we have this just came out Friday in the Wall Street Journal speaks to this issue on the other side are all the cynics people like Bob Perry that condi talked about who will tell you all the things that won't happen why nobody would need a pc in fact when I was starting son there was a quote from Ken Olson the CEO of Digital Equipment Corporation that said something similar like nobody needs a computer of their own that was great for us and I think he was right nobody needs the computer they need a hundred different computers they are all around us but if you cynical you can always convince yourself things can't be done let's run my favorite tweets give you some my favorite quotes and in it's amazing the diversity of people who've said the same thing Robert F Kennedy Martin Luther King unless you're socially maladjusted unless you do something odd and different you're not going to do anything important for those of you who have kids I have a talk on our website I gave it Sharia school that I'd highly recommend my first six slides went something like don't listen to your teachers disobey your parents color outside the lines literally each was a slide and and then Rao who's here his son who was 9 years 9 years old I think raised his hand and said my mom's always trying to get me to improve my handwriting because I got a C in handwriting should I practice it and I said not if you have something more important to do this this sort of I wasn't very popular but the kids all loved it but I actually feel it's a really important way to raise your kids and so I this and I'll come back to this but Martin Luther King is right if everybody stuck to being proper and well-behaved and I may be the only parent who at the age of 5 was throwing food at my kids because they were sitting there to properly it's honest to god true and actually my daughter had at 24th I just remember 24th 24 4th birthday party a few weeks ago and she said they had my friends think these are all 20-something or year old kids you're really cool because I actually threw some food across the table and they thought that was pretty cool so if you want to be cool human progress depends on unreasonable people trying to do unreasonable things I fundamentally believe in the power of unreasonable people he people won't accept the status quo will not compromise and will try things but in the premature guts in confidence and this is really important because if I knew all the problems we'd run into when we started son any rational person would have convinced me not to do it and because I didn't know it allowed me to get started and I'll come back to think these things are self-evident but it's amazing how many people ignore them the other thing is when you trying these things it's not easy it's not comfortable it is uncomfortable because you're trying to do things that others don't believe can be done and you are scared if you're smart you scared about all the things there is no courage if you're not afraid and so keep that in mind as you vacillate think about panic sometimes and in Pierre asked me to move this up from way back in my slide deck to right in front it is true failure is allowable but it's not desirable and you have to have a reason to fail you should have goals why you might try something that could fail failures in good company I know how many people know I'm Stein was considered dumb he actually didn't learn to talks until he was four he didn't learn to read till he was seven the Wright brothers and Regina had this court from Lord Kelvin that said heavier-than-air flying machines are impossible he was the president of the Royal Society I actually had that slide I took it out that court experts have all these opinions a few years later and through lots and lots of innovations the Wright brothers kept building one Fair on each failure to come up with the first flying machine Edison went through ten thousand iterations of the lightbulb and his view was I I have not failed I just found 10,000 things that won't work or ten thousand ways things this won't work Michael Jordan was the other guy who there's some beautiful Michael Jordan quotes on failure I'd recommend reading his courts I didn't know how many to include but he says I've failed over and over and over again in my life and that is why I succeed I'm not afraid to try the game-winning shot that I might miss and embarrass myself with think about it personally I feel like having done this for as long as I have I have a bigger library of failures and ways to fail than almost anybody I now work with more entrepreneurial companies and I hope I'm smart enough not to make the same mistake over and over again just discover new ways to make mistakes to me no failure means no risk means nothing new it's that simple this is why failure is important and so as I say if you're not failing if you're exactly unplanned exactly on schedule you are not stretching so I basically said when we started son if I knew everything that could fail or that I didn't know I would not have started it jumping into some things that you believe in something you have passion for without a parachute is great because necessity not safety is the mother of invention it is amazing how often you see an impossible task and when you keep banging your head in the wall because your company is gonna fail if you don't solve this problem it gets solved many of them you may have noticed how many of your breakthroughs happen the week before the board meeting it's amazing how but necessity is the mother of invention so let me go through some facts those five-year business plans Steve Blank talked about say Steve still in the room now he's that they create they're really the illusion of knowing don't have a clue and so stop pretending now that's not to say don't make plans because frankly to me plans do two things they tell me the quality of your thinking and the things you've thought about and the things you know you need to worry about but the most important thing is what we don't know we don't know those are the things we can only understand by being in full body contact with our problem this quote was used earlier this was his quote about pre fight fight plants that people have and I also had a quote that C Blanc used this morning all of us know this most people doing business plans pretend this is not true it is way more true and that's why Steve said get out there and engage the other thing I want to say is failure is not fun so don't you know we remind to the size failure but my goal isn't to do that I'm essentially saying it's an undesirable in the SAS t because we can't completely know it's never easy it's not a party and we should recognize that but I will say as our problems get bigger that freedom to fail is extremely liberating when I finished graduate school I was talking to my way my first VC is about my first company called dayz system and I said this is great because I don't own anything and I said it's a great deal if we win we Ben if we lose you lose because all I had to lose was my student loans I had nothing else to lose huh it's very liberating and while others were worrying about careers I said hey since I'm nothing to lose and I can bet with other people's money it's a great deal and I hadn't seen I didn't even know this was possible but think about Apple apples board brought Steve Jobs back in a failing company imagine if they were doing well if they had something to protect if they would have led Steve experiment the way he did with whole new ideas that no analyst or expert believed in think also for the fire of the fact that Steve failed a couple of times he succeeded sometimes DreamWorks kinds of experiments next didn't really work but he got the opportunity I'll come back to this the other thing is field strategies or field tactics are different things people confused the tail many people said well Friendster didn't work so social networking must not be right or lots of people did mp3 players so it's not it's a bad market the difference between the iPod and other mp3 players and there was probably dozens of them including the Microsoft Zune was the way they approached it for the tactics tactics was what was different the other thing people don't realize if there's ten things that are important to your success you control three of them your competitors control three there were so many things that could have done to completely destroy son no matter how well we execute it and then some of it is just luck you run into the right person the right idea the right timing I was on the border of picture tell the first video conferencing company and developing the market was such a bear it was so hard and then the Iraq war happened and suddenly videoconferencing took off we had nothing to do with it we didn't start the war but we'd sure took advantage of it to keep this in mind and that leads to this issue of persistence but let me this first time I'm going through this deck so I thought I had a different slide next so why fail you really want to test things hypotheses explorations better findings careful validations take every possible risk or opportunity you can find and in isolation try and test it but that's a price point whether it's a product feature whether it's a market risk it's a comparative risk that's what Lean Startup movement is about it isn't about they're not let me wait till I get to that the goal is to find out as much as possible on as little money as possible that is why failure is allowable it's to reduce the magnitude of the failure later when you've done a business plan execution based on your assumptions and you run into that first contact with customers or that first first punch that Mike Tyson talked about in what's the startups goal it's about fine buying insurance it's creating protecting against downsides its Risk Reduction but more than anything it is about creating option value there are things that just take off I am almost certain whether he agrees or not Mark Zuckerberg didn't imagine Facebook the way it was he had an option he took those options he did the explorations he followed the ones that worked you know Pinterest is a very successful site and Ben Silbermann told me once in there's a video of this on YouTube he went through three hundred iterations of that product before it took off this is also my favorite most people are limited by what they are willing to try not what they're capable of I think Regina in some ways talked about this issue so just try epic it is worth doing so in the startups like these you don't have strong operating history you don't have static environments you can't plan you can't even manage the way you normally manage you can have a rigorous schedule and hold people to it but if they're working on the wrong thing it doesn't matter so getting to some do's and don'ts this is absolutely key in the culture of a company it may be the single most important thing that I see missing in well executing environments but Stan Stanley McChrystal said just because you allow failure don't let people feel like they're their failures and then be really intelligent about what you're going to fail on which means what are you going to try and what you're going to experiment with there's and you probably can't see it these slides will be available there's a blog post by Brad Feld who I tried to have come here that's title of the blog post is called small experiments often as opposed to pretending you have a plan and you have the illusion of knowing what you're building so I think some of the people in this room have heard me say what are you gonna raise money on in fact Nutanix when it raised too much money I said okay to raise the money but tell me where you're gonna take 20% of the excess money and waste it on experiments experiments that have small downside you can lose one times your money but have 10x upside and opening up new markets you can't do it with a hundred percent of your cash balance and this Brad Feld force talks about this you know when Square recently raised a couple of hundred million dollars I said to Jack I said Jack okay to raise that much money but you gotta define and I gave exactly the same advice to John Haring at Lookout who just raised a lot of money tell me ten things in which you can raise five million dollars each where the five million is the downside but five hundred million on your market cap is the upside what experiments can you run but the same is true if you have a hundred thousand dollars to waste you can run ten ten thousand dollar experiments it is not a consequence or being large or having lots of money at every level it's a very good blog post this is obvious it's amazing how often people don't examine their failures you want to get into your failures what failed why should I try this experiment again coming at it from five degree different angle the other thing is when things fail we want to rationalize and then we go talk to others and I especially hate hearing from people who have no experience with trying things in the field none of that first punch that first contact with customers board members will have opinions about I told you so the price was too high the price was too low whatever whose opinion you trust on what topic and some people trust everything somebody says some people trust everything everybody says some people trust everything some supposed experts sighs I'll come back to that be critical about whose opinion you take and on what topic and ask the question why are they qualified to answer this question and I'll come back to if an expert has qualified how you take their opinion most startups should plan not to execute a plan but plan to build a plan I'll come back to this many of you have heard this analogy for me think of being in a one of these British roundabouts there's six roads you can take you can take the road you think you want to take and go down but you will only get to where that road is leading it's ok especially when your burn rate is really really low to go around in circles trying to scope out these roads not enough people do that especially since most board members and VC's want you start executing on a plan exploration is okay in fact exploration from my point of view is desirable as long as you spending tiny amounts of money your burn rate is small and wasting time I talked about wasting money wasting time isn't a bad thing depending upon your market if you're in a race to something you have to take a different approach none of this is always applicable it is often applicable you can see why I advise kids the way I do race money waste I'm not advice you most likely to get so you can discover a road worth taking as opposed to the road you thought was the right turnoff and you went down that and executed really well and ended up in an uninteresting place so don't plan flexy plan and create this culture of experimentation for those of you who remember the talk Scott cook gave his talk was titled culture of experimentation he's trying to change the culture in a large company like interior to stop planning and doing a lot of experimentation I think it's a really important phrase there are people who make this develop a culture of make no errors plan well execute well I'd rather you develop plans to recover really rapidly to evolve and iterate as opposed to develop and again I want to emphasize it's not one or the other but these are biases it's not black and white these are biases of where I find people are leaning the wrong way you want to focus early on unlearning learning as much as possible about all the risks and all the opportunities and all the solutions and problems in your market with the bias that every plan you do leads to rapid recovery while you failing before you fail before you try experiments ask lots of questions so Gold's a failure I've talked about this let me go a little bit deeper you notice I'm on slide 41 validate each element of your vision last year we handed out the Lean Startup book by Eric Ries that covers this in some detail this is the first goal of a startup find out what to build be in contact with that customer that's what Steve Blank talked about number of times they're seen startups build the wrong thing on budget and on schedule and in consequentially the second goal is to test the strategy and assumptions which are again different from what to build and then execution does become important it's not unimportant and I don't leave anybody with the message companies go through these phases and some of these phases run on parallel on some things you have to execute on budget and schedule and be very very tight while in parallel on other things you're experimenting the key goal in my view is optimal or maximal learning from each dollar you spend on a risk on an experiment that should be your early goal this is also something people forget I've never had a person come in and say here's our liabilities and here's our strategy for our liabilities I've never had a person come here's our competitors strengths here's how we gonna avoid them think about it they're just as important as here's my strengths and here's how we can leverage them this is like baseball the Giants winning or the Dodgers losing the game exactly the same time these are obvious I can literally talk to this slide for an hour this is sort of this collection of failures I I have in my library all the ways I've screwed up so I talked about circling the round up this there's also refers to a very good blog post by somebody from IDEO design this is how people think of business plans some talk about standard sequential pivots and Eric Ries talks about them this article actually talks about this approach highly recommend you read this blog it's all about testing all these experiments there are good failures and bad failures accepting failure or a line failure is not an excuse to be casual it is not an excuse to work less hard or be less diligent Lord lest heartful or less purposeful sometimes people mistake allowing failure in a culture of experimentation to be it's okay to be undisciplined it's not okay to be undisciplined and the goal every time you run an experiment there are some things you can just execute on yeah but the uncertainty is where you run experiments the goal almost always is small experiments where if they fail you lose very little a small amount of money but if they succeed they generate some large upside for you a new market a new segment a new marketing strategy a new product feature so we've heard a lot about the Lean Startup I won't spend time on it since I'm sort of running out of time but we talked a lot about it before you have to test your vision and then steer and then accelerate but nothing before it's time people also forget the Minimum Viable Product includes a V in it and they excuse it for a minimum product these are judgment calls you remember what ed is oh I have a great Edison code coming up you can't give up too easily I've seen people do it your whole startup process should be geared to accelerating this learning loop otherwise you might spend a lot of money doing the wrong thing how do you do that that's a seminar by itself maybe next year we can do that even if it's slower to your product vision or at least the illusion you have of your product vision and at least two people in this room in the last month or two have had meetings with me that I said oh yeah I said you're doing this you're not worrying about XYZ they said well we have money in both cases the answer was we have money through 2014 I said because you have money doesn't mean now isn't a time to panic there is a time when panic is the appropriate response and you shouldn't be too boneheaded about recognizing that moment if your visions wrong having 18 months of cash isn't gonna help you discovering what's right as quickly as possible or doing something different is important but you have to have the courage of your convictions because everybody outside believes what you're doing can't be done or shouldn't be done or nobody needs it you heard that from Airbnb nobody wanted or thought you need that product so this balance between the courage of your convictions on these scary decisions and that's why I say to people be obstinate about your vision be flexible about your tactics I also find people who are in the startup business and this is almost a religious belief because they want to make money always pivot too quickly always quick too quickly people who have passion about their vision stick with it longer sometimes too long but I'd rather stick to long than too short and so missionary founders just him to do better if you believe no matter what you do your plans not going to play out as orchestrated and you'll run into problems if you don't have belief system religious belief in what you're doing you'll give up and you have to have both optimism and paranoia if you aren't optimistic you're not going to attempt tying if you're not paranoid you're not going to be scared - enough to discover your problem so you do have to be schizophrenic no two ways about it both are absolutely essential you can talk yourself out of anything important by being pessimistic go back to my tweet about the cynics experimenting frugally is possible if you break down the experiment and always say now how could how could I test this thing with half the money or one-tenth the money it's often possible but people have this view of developing a product and the product vision that they are executing on where they'll spend a million dollars rate they could have tested something for ten thousand and then take six months or taking three six days not enough people focus on lowering the cost of experimentation so this is a chart from AI Netra I only got this in a review very recently we funded them ten months ago two hundred thirty one product iterations there are 50 some approx Hardware iterate distinct Hardware iterations of the product they've done in this short period of time because the cost of experimentation has been reduced by their startup process this is what I mean gear your startup process to accelerate learning that's a classic example of that also when you're looking outside people who have nothing to lose nothing to gain by our success will generally give you politeness they won't tell you what they really think see critique many of you feel uncomfortable because I'm brutally but it's what you want we tend to go away from people who criticize us or critique us and sometimes people don't do it well enough I often mix critique with criticizing or at least comes across and for that I apologize but it's because I'm passionate about surfacing things you should test or worry about and I've talked about that you want to walk down the risk of not up the expense Co the number of times I see plans this is our spend plan for the next two years when we raised our funding and people keep going down that plan without worrying about what the risk curve looks like if risks shows up something slips they don't adjust their plan what you want to do is walk down that yellow line of risk and at every inflection point of risk think about increasing your burn rate the lower you burn the more flexibility you'll have in taking more time to eliminate these risks and start with the biggest risks not the order in which you need to do product development something that might come much later you might want to test earlier because it's the biggest risk because if it fails or if it takes three times longer as often happens you've done it when your burn rate is lower and it's much easier also the fact is and remember I talked about normal management processes don't work if anybody here taking a course in f1 racing they teach you when approaching a curve that you accelerate or you brake as late as possible brake as hard as possible at the last possible moment so you don't skate out of the as you make the turn you accelerate as fast as possible that's how what I believe startup should be run and those curves are the risk curves that I talked about earlier while building an upside plan build a if-then survival plan up sides don't always happen you have to survive to build an upside so I'll come back to that with point this something people forget or not all the hard things are valuable but most of the hard things there's an occasional snapchat or Instagram and there are exceptions but most valuable things are hard if they weren't hard somebody else would have done them and the harder they are the bigger the comparative lead you build by solving those problems persistence is really important remember I said you control three things your competitors control three and then three things are just three or four things they're just lot if you don't give yourself time to get lucky by being persistent and this is where a burn rate management is so important give yourself a chance to be lucky the right things to happen people also optimize too much I actually think the most important thing for startups and frankly I believe about society about anything is not to optimize for flexibility but for for X or Y like throughput in a database but to optimize for flexibility it's what lets you make those adjustments those changes if you believe you can't plan then you have to optimize for flexibility and make explicit trade-offs in favor of flexibility and ability to change as you run into problems this is another good article lessons from failed startup I won't spend too much time on it since I'm out of time there is a trade-off between speed and stability each is different and appropriate at a different time preserve option value the number of times I see people say ok we agreed on this we are doing this that means less license this off well keep it let it hang around for a bit in case you have to pivot also knowing things isn't enough the gap between knowing and doing is a mile wide and you won't really know what you're doing till you actually do it this again this idea are few plans survive first contact with the customer very good blog also we hear a lot about lean startups Apple wasn't a Lean Startup lean isn't always possible if you're building a satellite system there are some things you have to do you can do lots of experiments but in the end there are some things that just take more time more money and a lot of the hard things take money persistence time patience a lot of things that go against some of the things I've talked about startup culture so there is an appropriate strategy for every startup and they can be different this is one of my favorite quotes I guess I have few successor pencils represents the 1% if you work which results from the 99% that is called failure this was digest in quote I was talking about it took him more than ten thousand iterations to come up with the lightbulb many of you and last year we talked about base camp get to a survival place before you sort of build the grand vision I like to say get to base camp where you have enough stability as a business then you're actually in contact with customers before you plan your scent to the peak but get your base camp not in a random place but close enough to the vision so you can actually use it to scope out the plan and learn that learning is very important I promised comment on experts we talked about Dan Gardner talked about experts last year their average accuracy is the same as dart throwing monkeys this was the result of a toniest study across 28,000 experiments a forecast by experts remember when I said I'll come back to how you treat experts treat their opinion especially the ones who are not directly in contact with problems on the ground not actually doing things as good learnings which are mostly biases many biases are good and you should use them that way add it to your risk matrix an expert's opinion is important it's not to be ignored but it isn't to be believed either talk about that so I personally believe and I talked about when I got started I had nothing to lose but my student loans that attitude my willingness to fail is what I fundamentally believe gives me the opportunity to succeed I know a lot of people who are a lot smarter but a lot more conservative they don't want to take risks they don't try things lots of people much better than me most of the things I do I don't mind failing and I've been critiqued sometimes for saying I don't mind a 90% probability of failure in a startup but most people like the fact when I say but if there's a 10% chance of changing the world that's pretty good odds I find what's important to optimize for your risk is down small but what you work at try should be worth succeeding at most investors and that's why I hate calling myself an investor in fact I never have reduce the probability of failure to the point where the consequences of success are inconsequential it goes back to doing epic so let me give you one example and then I'm gonna try and finish up with a few very fast quotes 1996 January I'm working with pradeep's Indo said what should we do decide the world and the Internet should be tcp/ip the Internet Protocol IP stands for Internet Protocol talked to a dozen customers they all said we'll never use this product so I called Cisco CTO and they said they'd never build a public tcp/ip router above or C 12 which is like your home service now they said every customers asking for a technology called ATM a me and Pradeep agonized over this and finally said let's just do it and we did and the internet grew so far fast and we got lucky and we found one guy guy called Michael Dell at a small company called unit small internet provider so Cisco focused on what AT&T in Pacific Bell and others were telling him we focused on what the market need was saying if the internet was going to grow this company's I haven't checked market caps recently is 15 or 20 billion dollars but what's scary is if he hadn't attempted it most likely the internet would be an ATM network it just gives me the shivers to think what would have happened because Cisco wasn't going to develop it Lucent and the Alcatel didn't have programs the hardest startup was a company called Newbridge and this is where the courage of your convictions becomes really really important I didn't even tell my partners at kleiner what feedback we were getting from customers let's say you crazy because nobody wanted it but we believed they needed it and we had conviction around their need and so we built that and they came and the same is true that some people tried to convince me by the way I was telling you AT&T 1996 AT&T said they would never use that product five years later AT&T was essentially going out of business people forget five years later it was sold for a song to Cingular Wireless in fact only the brand really survived this massive company disappeared in five years the same thing happened at Sun everybody tried to convince me the best thing for Sun to do was build graphics terminal for deck boxes we started in February of 1982 in 1987 that got sold to another startup compact which was the number two computer company in the world so I could go through all these examples they're just so many epic is worth doing what would you do if you weren't afraid to fail or if you were sure you would succeed and in fact you can look at the history of innovation where did it come from did it ever come from a big company GM should have owned the electric car so try and fail but please don't fail to try my other Michael Jordan quote so let me stop there I in the interest of time won't take questions so this quote was a quote I put in 1982 on a slide deck I did though the we did overheads back then not slide decks and it was on a movie my first day of learning how to hang glide and this movie was dedicated to those who dare to dream the dreams and a foolish enough to try and make them through the absent rate for hang glider's was really really high back then and i said this is what entrepreneurship is about so please go at it thank you [Applause]

Channel: Khosla Ventures

Views: 77,934

Keywords: 2013kvsummit

Id: MLtw31CGTHc

Length: 52min 48sec (3168 seconds)

Published: Sat May 17 2014