VentureSwell

I pieced this together from Steve Blank's talk(s) on Customer Development at http://ecorner.stanford.edu/authorMaterialInfo.html?mid=2056. Thanks to Nivi at Venture Hacks (http://twitter.com/venturehacks) for passing this along. 

And what happened was I had time to think about,  finally in 1999, about what happened in the previous 21 years.  I realized that when I looked at what I had done and what  other entrepreneurs had done that there was a pattern.  And it was interesting thinking about  what was going on with startups.  I realized that traditional ways to  think about startups - have an idea,  raise some money, do product development, go through an alpha test,  beta test and first customership was the canonical model  of how entrepreneurs thought about early stage ventures.  In fact, in my class I kind of used to ask,  "Can anybody draw this model?"  And I used to say, "Even the waiters in  San Francisco could draw this model." 

So somebody raised their hand and said,  "Well, we're waiters in San Francisco because  we used to be CEO's of dot-com companies."  [Laughter]  So I no longer make that joke.  But the canonical product development model - idea,  raise money, develop, alpha, beta, first customership is  the one that the valley grew up in.  And when I looked back at it I realized that it  really set me up for a fundamental question.  As if you all follow that model,  why is it that some companies are opening bottles of champagne  at their IPO or at their act of rescission and others who almost  did the same rules are selling off their furniture?  What was the difference here?  And is there any way to predict other than you got  to work hard and all the war stories you would say?  Is there any way to predict success or failure?  And even more importantly,  was there any way to reduce risk in early stage  ventures because I had done eight of them?

 I had sat on a ton of advisory boards and  by the end of my career, I was investing.  And these patterns kept coming up.  And I'll tell you what the biggest pattern was for me.  An observation number one was that most entrepreneurs,  whether they were engineers or the CEOs or part of the founding team,  were focused on one thing.  And that was getting them to launch and first customership.  The board was driving them to that,  engineering was driving to that and everybody was focused on  "that's the box we're heading to." 

Well that's interesting.  Product development,  concept, develop, alpha, beta, first customership.  I started asking a heretical question.  Whose diagram was this?  What's the product development diagram?  Well, that's interesting because in the product development diagram,  we seem to be hiring and staffing sales,  marketing and business development people based on the  timing of events on a product development diagram. 

Turns out if you look at canonical model,  around alpha or beta time you start hiring marketing people,  start getting ready for launch.  Start doing positioning and data sheets.  Maybe hire an ad agency, start writing blogs, start communicating,  all heading to that big launch event.  And if you're doing something not on the Web - how many  of you were doing or will do non-Web products?  How many are you going to do Web products?  Assume that there's a world outside the Web. 

You start thinking about maybe it's time to hire sales people,  start staffing, getting early customers.  And if you're thinking about business development,  you start thinking about deals.  The observation that I had was all heading for a party.  What do we do at first customership? Anybody know?  What?  Party. It's a party. Who's having the party?  Anybody know?  Anybody ever been to a company that had a first customership party?  Anybody?

Trust me, the whole company parties, OK?  If you really think about it, who should be having the party?  Who's done at the end of product development? Is sales done? No.  Marketing done? No.  Is dev done? Not really.  The only people who are actually done is engineering,  yet somehow we end up inviting everybody else to engineering's party.  While that might seem a little facetious,  it actually drives a lot of company culture - spending, burn rate, etc.

 I realized that this might actually be a  fundamental problem with most startups.  Is that actually startups were burning money by starting sales  and marketing and business development activities either on  the Web or physically in the real world way too early.  Because if you think about it,  here we are in Silicon Valley,  at least those of us physically here not those watching on TV.  But for those of us here in the Valley,  we take technical risks with products all the time.  Investors put huge bets on all of you to do innovative things. 

Anybody have any idea what percentage of startups  fail because their technology fails? Any idea?  How many think it's over 50 % of startups fail  because their technology didn't work? Over 50 %. Over 25 %  How many think it's over 25 % fail because technology doesn't make it?  Turns out less than 10 % of startups fail  because the engineers were just wrong.  Turns out most startups,  and I'll leave Life Sciences aside for a second.  Most startups in every other field other than Life Sciences,  over 90 % fail because they didn't find  a market and customers - full stop. Big idea. 

Well, if that's the case and go ask your favorite  venture capitalist or next guest and you'll have a  lot of them in this class and every other class. Go ask them.  If that's the case, why is it that we have tons of methodologies,  some measure and help us getting the product right,  but no methodologies to help us get the  stuff about customers and markets right?  Anybody ever been in a company with - in product management  or know what product management people do?  Some people in the back of the room. 

There's entire tool sets on how to manage technology risk. Tons.  But there are almost no tools to manage customer and market risk.  Well, think about this;  you're an entrepreneur; I have a great idea, good! Let's go build it.  Oh, good, let's go raise money. Oh, let's go sell it.  Do you know what the next step it is? Oh, we're out of business.  Because most of us grow up reading these wonderful stories about  all the people who made a ton of money at Google and Facebook  and other companies that like - these are great examples.  Do you ever notice they don't bring in  the people who you didn't hear about? Because you wouldn't have come. 

How about bringing in the people who said,  "You know I cratered my last seven companies. Let me tell you why."  You'd go, "Well, I don't want to be one of those. I want to be Google  and Facebook. That's whose presentations I want to hear."  Any idea what the ration is between the Googles  and Facebooks and other startups?  Know how many companies fail every year in Silicon Valley?  Anybody want to guess? Do you know what the ratio is?  I'm sorry? Over a thousand to one.  Now, one of the nice things about human nature is every one of you is  going to be convinced that you're going to be Google and Facebook.  You have to be, right? I mean, that's the passion of an entrepreneur.  You have to be convinced. You know those other 999 companies?  They were just idiots. I am much smarter than them because  I have better idea and we're cool.


  Unless some of you have been working in a specific  domain for the last 20 years or so,  the odds are that anything you're thinking about customers  and markets are nothing more than a guess.  And you go, "Oh no, the buddies in the dorm. They liked it! And look,  you know I put it up and I got 300 hits!"  That's not a business.  In fact, the heuristic is, if you can't get 10,000  hits like in Silicon Valley, like your server is not connected.  But it's not a business.  The real notion is, how do you take your idea,  size the opportunity and figure out whether you're hypotheses,  your guesses about customers, about the problems your solving,  about your product are correct? 

And the first thing I teach students and coach young  entrepreneurs is while your hypotheses are great,  have all of them you want,  you got me excited, the next step is get the  hell out of the building and test them.  On a Web company, you could do that virtually.  You can get users to click on your page.  If it's a physical product,  you physically leave the comfort of your dorm or  office and you go out and you talk to people. 

And the first step you want to do is leave any notion of your slides  or website or whatever at home and go out and test some of the  fundamental hypotheses that you have about your business.  And you have two that are just absolutely essential.  One is, in any company that you're making a product or service you  believe you're solving a problem or need that a customer has.  Great! I'll believe you. Now show me that there are customers  out there who have agreed with you. 

Who have said, "Why yes, this really is a problem and it's the most  important problem we have. And in fact,  it is so important we'd actually pay you money or use  your service. Or if it's not a problem you could convince  me that it's a need I didn't even know I had. And I'll  rush and go use your site or buy your product."  And while you're out there, by the way, you can find out millions  of other things about customers.  And this is hard because if you're a passionate entrepreneur,  you look at somebody like me and go,  "Steve, you don't get it. I'm right. I'm passionate.  I got it! I nailed it. I've talked to 20 people. They're  all over me. I'm done! We're launching!" 

And I'll usually go, "Great. So this ought to be a very quick  exercise. In about 20 minutes,  you could go out of the building or get some people to  click on your website and confirm just what I said."  For those of you thinking about companies and particularly  those of you building web companies,  you can run this exercise in a week.  And trust me, it would be very sobering.  The goal is not to convince you not to do a business.  The goal of a customer development process is simply to take  the hypotheses about your business and product as-is and see  if there's customers and a market outside the building.

 The most radical thing I make companies  do is actually sell their product.  You know Web product, you could decide you're going for eyeballs,  for users, you could go for ads or you can actually decide to  do something radical and actually charge money for something.  Or for a physical product, that's almost a given.  Before we do anything,  let's see if the founders could go out - not a sales team but the  founders - can go out and generate some revenue with this company.  That's what I call "customer validation".

 I want you guys to actually sell something.  How many of you were in engineering? OK, all right.  How many of you have actually sold something?  OK. Not just a pencil, not Girl Scout cookies.  You know, Tina asked me earlier,  what was the best advice I think I've ever gotten or given.  And that was even if you are in engineering,  if you decide you're going to be part of an entrepreneurial startup,  if you've never been out trying to sell  your idea or be part of a sales team,  you will never ever be a great entrepreneur, ever. 

I really want to get all of you angry who  are great engineers or great architects.  That's great, but you'll always be dependent on someone else.  Great engineers understand what customers need.  I don't mean you need to learn how to  go to dinner and drink or whatever,  but you really do need to understand how people buy,  why they buy, how they process information  and how they think about your product.  And also, how they think about the problem you think they have.

  Any of you read TechCrunch? Any of you read some other blogs?  What's another good technology blog?  Venture and Gadget, etc. Isn't it cool  if like you start your company,  the first thing you do is like, TechCrunch mentions you?  Wouldn't that be the coolest thing ever?  I got to tell you, if I'm on your advisory board or board,  I'll break your arms.  And you go, "Whoa! Whoa! What's that about?"  Why would I say something like that? 

For me, any type of press,  any type of PR, any type of talking about  your company is not done over here.  Because on day one, your hypothesis about what your company is about,  what service you're offering, what pricing, whatever,  trust me, will be radically different after  you have some contact with customers.  And while it might make you feel good being able to show your  mother or father a picture of you in the press or a mention,  I'll do that for free.  I'll take your picture, we'll hold up the New York Times and  then we'll send it to your parents. 

But I rather any press you get for a company is part of a strategy,  not a random tactic.  And that's typically done after you've  understood what business you're in, who your customers are,  and how do you need to scale demand for your company. 


  The last step in customer development is about company building.  And for me, company building is the Shakespearean  tragedy of an entrepreneur.  Anybody ever hear stories of entrepreneurs whose company  have gone public and they've been thrown out of it?  Anybody ever read - oh! Couple of hands. Oh! A lot of hands.  It's almost a war in Silicon Valley.  Gee, the VCs took their company away from them.  Oh, they lost their company.  Oh, they were great entrepreneurs but then they brought in the suits.  Right? Almost a war in Silicon Valley. 

Turns out that what most entrepreneurs don't understand  is when companies are becoming successful,  all of the skills that made them great entrepreneurs may  or may not be useful in the next phase of a company.  And in turn, scare the heck out of investors.  Because while you might be great at  starting and finding new businesses,  typically when your company gets to this phase,  you're growing the business.  And you're doing something repeatable and scalable.  Might not be the same skills that worked early on. 

And I call it a Shakespearean tragedy is because world class entrepreneurs  here are not necessarily the same people who build the company.  And yet, another irony is the largest technology  companies are run by their founders - Microsoft, Oracle, Jobs at Apple.  Interesting conundrum. But then there's a gap in between.

  When I first started at Convergent Technologies,  the founder, Alan Michaels,  had left Intel and he believed he was going  to be making a single board computer.  Computer that fit on one circuit board back  in the early 1980's was a radical notion.  And what Alan did was he went around to computer companies and said,  "Look at what I got. How many would you like?"  And they all kind of knew Alan from his work in Intel and said,  "Wow! That's great! We'll take three for our R&D lab." 

And Alan went, "No, no, no. I'd like you to take 3,000."  The first company he went to said,  "Alan, we're looking at a new type of computer but it doesn't  even have a case. It doesn't have an operating system. It doesn't  have any applications. It's a nice building block but maybe we'll  take three and maybe we'll license the design from you."  So Alan went to another company, this time NCR.  And he gave them the same pitch and they said,  "Alan, we'll take five!"  He said, "5,000?" "No, we'll take five."  Gave him the same story.  "It doesn't have a box" and whatever. 

And finally, Alan went to the third company,  a company called Burrows at the time.  And they said, "Alan, we'd take 10."  He said, "Oh, oh, oh. You mean you'd take 10 boards? Absolutely,  but would you like to see my box?"  And they said, "Well, you didn't mention a box when you..."  "Oh! Let me draw the specs of the box. It has an enclosure."  And they said, "Oh, great! Does it have an operating system?"  "Absolutely! It has an operating system." "What else does it have?"  "Well, what else do you need?" "Oh, does it have a word processor?"  "It has a word processor!"  "What else, what were you thinking about? Oh, it has one of those!" 

Right there in discovery Alan finally realized that  what he was hearing was that what he had was good enough  to get three but was missing a couple of zeros.  Now of course he had to go back to the rest of his company,  which had all of seven people and said,  "Good news! I just got an order for 10,000  computers from Burrows. Bad news is we need an  operating system and a set of applications."  And his co-founder, Bob Garrow,  looked at him like he had been - somebody hit this  guy with a brick because he was clearly... 

He said, "Alan," you know this was like 1987,  "you just don't like off the shelf get an operating  system and a set of applications."  And so Alan, being a great entrepreneur said,  "Well, where do you get them?"  And Bob said, "Silicon Valley, I don't know. Xerox PARC,  what are you going to do, hire Xerox PARC?"  He hired the head of the Advance System Division  of Xerox PARC who came with an operating system  and the guys who wrote the Bravo word processor,  became the Convergent Machine.  He went from a company that potentially would have sold  25 boards in the first year to one that actually sold its  company four years later to Burrows for $400 million. 

Now that's an interesting story about discovery  was that Alan was good at listening.  But what's even more interesting is what happened  at his next company called Ardent.  Alan was now a successful CEO.  In the early 80's Alan Michaels was well known in Silicon Valley,  bad boy of Silicon Valley.  Kleiner Perkins, everybody wanted to fund his next company,  anything he wanted to do.  Alan decided he's going to build a graphics super computer.  A cross between a Cray and a Silicon Graphics which  in hindsight turned out to be the - exactly if you were  making a floor wax and dessert topping in one.  It's the null set, but we only found that out later. 

And this time, Alan Michaels was not  in the field talking to customers.  Where do you think he was? Any idea?  Second company, successful CEO, where was he? In the office.  Who was he talking to?  His investors.  Who was the people in the field talking to customers?  His VPs, particularly his VP of Marketing yours truly.  And customers were saying,  "Steve, this is the worst combination of product  features we've ever seen!  But if you've ripped off this piece and just gave us the  server the price performance of this stuff is really - and  if you change the vector unit to do this, it was great!" 

Oh, I went around. I went around to almost every national lab.  I went around to the NSA; I went around to a bunch.  A lot of excitement for something about 20%  different from what we were building, not completely but different.  So I went back to headquarters feeling really smart.  I actually found out what customers want.  What do you think happened when I went  to Alan Michaels' office and told him,  "Hey, guess what? Our baby's ugly. In fact, your baby's ugly." 

What do you think he said?  "You're fired." In fact, I was fired three times that day.  [Laughter]  So important heuristic here,  if you're going to do customer discovery at all,  and by the way, he never listened.  Company went out of business painfully.  One of the wonderfully enjoyable failure sagas at Silicon Valley.  If I had two hours, I would tell you.  But here's another heuristic,  if you're going to go out and discover whether  customers like your idea or not,  this is not an outsourceable problem.  The founders need to do this.  Particularly the people capable of changing strategy  need to be the ones hearing good news and bad. 

Why? Because it's like hiring some consultant agency.  "Oh, what do they know?"  Or worse, "Steve, you're just not explaining it well enough."  Anybody here that one? "We just need to explain it harder."  Getting feedback from customers is the most  valuable thing you will do as entrepreneurs. It is not outsourceable.  And if you are a technical founder, don't believe that  "Oh, I'll hire a professional salesperson and they'll sell it for me"  without you ever, ever leaving the building.

This is a great question.  One of the problems with entrepreneurship and  I think I'll kind of answer your question,  is that we treat startups as junior versions of large companies.  I mean just fundamentally that's wrong. Think about it.  In a startup, like a big company, we say,  "Oh, big companies have VPs of Sales,  we need VPs of Sales. Big companies have VPs of Marketing,  we need VPs of marketing. This is just a junior  version that if we water correctly,  will grow into a big company.

This is just like IBM  but smaller or just like Google but smaller."  It turns out that until you have your first customers,  all the tools and tricks that large  companies use are divided by zero.  They really don't work in small startups because  you're actually a different type of animal.  And so I'm just riffing on the should we use saleforce.com  or other CRM stuff, that implies that you actually have a sales  force that's just a smaller version of a larger company.  And my experience is no. Actually, the titles are wrong. 

If you really want to get into it,  one of the first tools I do now with startups are,  really, VP of Sales? No, you're not. VP of Marketing, nope.  You're all in the customer development team. Holy cow!  What's a customer development team?  Well, VP of Sales assumes that all we have to do is drop in an additional  salesperson and sales will go up at a fixed increment.  Customer development says,  "We don't even know who are customers or markets are.  So therefore the titles that says VP of Sales,  which implies we're scaling a sales organization, is just wrong.

  If you look at my career,  I was never in the same industry twice - maybe semiconductors.  There's nothing about a business that isn't rocket science  to understand well enough within a certain period of time.  I'm not asking you to become a heart surgeon,  but it's not that hard to understand  the tools a heart surgeon might need.  But you won't do it from inside your building. 

That is, the job of a great entrepreneur  is to become a domain expert,  enough to understand the day in the life of their customer,  their buyer and anybody else in their organizational chain.  If you can't do that,  you are just tying one hand behind your own back.  And if that sounds like too much of a burden,  don't do it and be like every other entrepreneur  that goes out of business.  Understanding why your customers buy,  what's important to them isn't passed. 

"Gee, I developed a neat idea in my dorm or company."  Does that answer your question?  Have you ever, from out of your personal experience,  ever experienced something where you've actually driven  a customer purely by understanding his business model?  Sure. It's not that I've driven a customer.  It's the other way around.  I've understood that I was talking in Greek  and they were speaking in Spanish.  That usually it's because I was talking about features  and they had a problem they needed me to solve and  I wasn't talking about reservoir simulation.  I was trying to sell super computers.  I wasn't talking about computational fluid dynamics  and I was trying to sell them super computers.  I wasn't talking about Quark, PageMaker and Illustrator,  and I was trying to sell them desktop publishing graphics boards. 

Usually, when you can connect on a level of what's their need,  you actually get a lot further about - let me tell you  about the bits or let me show you this great website.  Does that answer your question? 

Awesome.

Loading mentions Retweet