对一个运动的反思 | Eric Ries(精益创业方法论的创立者)
Reflections on a movement | Eric Ries (creator of the Lean Startup methodology)
Eric Ries: People act like having a startup fail is the worst thing that can happen to you. And man, that’s not even in the top 10. It’s bad, I’ve done it, it’s awful. It’s really bad, but far worse is to be in a company that won’t die, a zombie, undead company that you hate, but you can’t leave. Oof, have I met people like that, and we’re having a mental health crisis among founders that’s not talked about enough. People have started to talk about the downside mental health risk, obviously the stress of being a founder is hard, but look what’s happening to the people that are so-called successes? When you build a company and you sell it out and it becomes something that you find abhorrent, man, maybe you get rich, but it’s not good. And so building a company you hate, that becomes a maligned force in the world, that you have to go pretend you weren’t involved with or like you feel complicit in… That’s way, way worse, and I wish more founders would take that more seriously early on.
Introducing Eric Ries
Lenny: Today, my guest is Eric Ries. If you’re not familiar with his work, I would be shocked. He is most famous for creating the Lean Startup methodology and movement and also his incredibly influential book, The Lean Startup. He also coined more terms and concepts that are part of the tech culture than anyone I could think of. He currently spends his time advising founders and startups. He was a former founder and CTO, and currently is the founder and executive chairman of the Long-Term Stock Exchange.
Forward-thinking companies like Figma, Amplitude, Loom, Ride Games, Linear, and more use Sanity to build content growth engines that scale, drive innovation, and accelerate customer acquisition. With Sanity, your team can dream bigger and move faster. As the most powerful headless CMS on the market, you can tailor editorial workflows to match your business, reuse content seamlessly across any page or channel, and bring your ideas to market without developer friction. Sanity makes life better for your whole team. It’s fast for developers to build with, intuitive for content managers, and it integrates seamlessly with the rest of your tech stack. Get started with Sanity’s generous free plan, and as a Lenny’s Podcast listener, you can get a boosted plan with double the monthly usage. Head over to sanity.io/lenny to get started for free, that’s sanity.io/lenny.
You fell in love with building products for a reason, but sometimes the day-to-day reality is a little different than you imagined. Instead of dreaming up big ideas, talking to customers and crafting a strategy, you’re drowning in spreadsheets and roadmap updates, and you’re spending your days basically putting out fires. A better way is possible. Introducing Jira Product Discovery, the new prioritization and roadmapping tool built for product teams by Atlassian. With Jira Product Discovery, you can gather all your product ideas and insights in one place and prioritize confidently, finally replacing those endless spreadsheets. Create and share custom product roadmaps with any stakeholder in seconds, and it’s all built on Jira, where your engineering team is already working, so true collaboration is finally possible.
Great products are built by great teams, not just engineers. Sales, support, leadership, even Greg from finance, anyone that you want can contribute ideas, feedback, and insights in Jira Product Discovery for free, no catch. And it’s only $10 a month for you. Say goodbye to your spreadsheets and the never ending alignment efforts. The old way of doing product management is over. Rediscover what’s possible with Jira Product Discovery. Try it for free at atlassian.com/lenny, that’s atlassian.com/lenny. Eric Ries, thank you so much for being here. Welcome to the podcast.
The Journey After Lean Startup
Eric Ries: Thanks for having me.
Lasting Impact on Startup Culture
Lenny: This just feels very surreal to have you on the podcast. It feels like a surreal experience. You’re such a legend in product and in startups, and I’m just really excited to get to learn from you and for listeners to get to learn from you. And my first question, what are you up to these days? I know you work on the Long-Term Stock Exchange. What else is going on in Eric Ries’ life?
Two Elements of First Principles Thinking
Eric Ries: If you can believe it, The Lean Startup came out in 2011, so it’s been a little while now and it’s been such a crazy ride. And basically every year since it came out, I keep thinking, “All right, that’s the end. Surely that’s enough.” And just each year it gets crazier and crazier, and new things come my way. So, I’m building the Long-Term Stock Exchange, LTSE, I’m sure we’ll talk about, and I spend a lot of time still with founders and with big company execs trying to answer their questions about how to build companies in the right way.
Obviously, been sucked into a lot of AI stuff of late, as everybody is, and one of the interesting things that actually surprised me is I spent a lot of time the last 10 years working on the issues of governance and corporate governance and how companies should be structured, not really with an eye towards technology in particular, but just with long-termism and humanism really at its core, and it turns out that those questions have just become really acute for AI companies. So because of my work on governance, I wound up meeting with and getting to work with some of the top AI companies, so that’s been really interesting, but I feel like I go where the issues that are on people’s minds go, and it kind of pulls me into new and really interesting things all the time.
The State of Lean Startup
Lenny: Amazing, so I’m going to want to talk about all those things as we get into this. Something I wanted to mention is I feel like people don’t give you enough credit for how many core concepts of startups and product building you’ve either invented or popularized. So what I know, obviously Lean methodology, then there’s the term MVP, then there’s the term pivot, and then also I think A/B testing, customer development, continuous deployment, vanity metrics as a term, I’m guessing there’s many more, and then the Long-Term Stock Exchange. And I think the theme I get from this is there’s a lot of first principles thinking that you’ve done, and so just a couple of questions here. One is, did you think you’d have this much impact on startup culture and language? And then just, where do you think this comes from, this first principles thinking that you seem to be really good at?
The Responsibility of Ideas
Eric Ries: Well, first of all, thank you for saying all that. That’s giving me too much credit, but I like the idea that it’s going to balance out the too little credit I ostensibly get somewhere else, that’s great. To answer your question, no, I didn’t know this was going to happen. This is so embarrassing to admit now, but just to give you a sense of how different the world is now than at the time of the financial crisis, I was so embarrassed to be blogging, that I didn’t put my name on the blog. I published it anonymously because it wasn’t a thing that people did and people told me it would be career suicide. And there were so few startup bloggers at that time, that the top three all reached out to me personally because they saw it in their referrer logs on their website. That’s how little traffic they got, that when a new person started writing about their thing and linking to them, they were like, “Who is this guy? What’s going on?” It was a very different time.
And I had just finished a startup. I had left a startup that was being successful, and I had this reputation among VCs that I could make engineering teams be supernaturally productive. I had the magic pixie dust. And so, they would ask me to come in and work with their teams and I would be like, “No, listen, honestly, I’m not that special. It’s nothing to do with me. It’s just this framework. I have these ideas that are helpful on a team when they adopt them.” And they would pat me on the head and be like, “Yeah, kid, I’m sure that’s great, but can you please bring the magic pixie dust to my company anyway?”
And I would go and have these meetings, and I’m not making this up, people would yell at me in the meeting. I would get yelled at. In fact, I would be asked not very kindly to leave, and I’d be like, “I don’t understand. What are you yelling… Why are you mad at me? You asked me for this meeting.” And they’d be like… I’m just telling you what I witnessed with my own eyes. We at IMVU, we shipped product 50 times a day on average at a time when people were lucky to be doing it monthly if they were really advanced. Remember, it wasn’t that long ago that we put the year the product came out in the name of the product. That tells you what cycle times used to be. That was considered normal, and people were like, “That’s impossible. That could never work.”
And at that time, I wasn’t even propounding a theory or anything, I was just explaining what I had seen and people couldn’t believe it. And so I had this great idea, if I write some of these stories down, then when someone asked me for these… I was doing these meetings all the time. I didn’t understand the way Silicon Valley worked at that time, I didn’t understand why this was happening to me, having these meetings, and I said, “I’ll write the story down and then I’ll send people the essay, and if they think it’s crazy, then we don’t have to have the meeting and I won’t get yelled at.” That was really as far ahead as I was thinking at that time. And then it basically took over my life, and I was being asked to speak, but people really wanted to unmask the anonymity. Who is this person? And so I came clean, I identified myself and started writing and speaking and working with companies, and it became a whirlwind.
The first principles part of it, I don’t know, I’ve always been wired that way. I feel like maybe I’m a failed scientist. I wish I was actually good at chemistry or something. I could have actually been a real scientist, but I always really had a respect for the truth for scientific thinking. That was a big cultural value that I was raised with. My parents are doctors, it was just definitely part of my upbringing, and I was really into reading stuff that worked that way. So, it was very natural for me. I felt like as I had success as an entrepreneur, I wanted to know, why does this work? And people gave me advice. All the advice used to be of the form, “Steve Jobs once did this thing, so if you do it too, you’ll be like Steve Jobs.” And it’s like, well, he also wore the black turtleneck. Is that also part of the… Help me understand, why does it work and is it still appropriate? Does it make sense for that industry or not? I didn’t like these just so stories.
And I realized looking back, I didn’t have this understanding at the time, but looking back, I feel like first principles thinking, it really has two components that people pretend are one thing, and it’s more one than the other. A lot of my work is just descriptive. It’s not actually telling people what to do, it’s simply giving a name and a concept to a thing that already happens. That’s the funniest part. In the early wave of criticism against Lean Startup, people were very angry at the idea that I was saying that a startup is an experiment like, “You can’t treat a startup like an experiment.” I was like, “Well, you could treat it however you want, but it is an experiment.” People would like, “Well, people don’t like the concept of pivot.” This anti-pivot people, anti-MVP people every once in a while come out of the woodwork still to complain about it.
It’s like, “well, I’m not telling you that you should pivot. I’m just saying that is what it’s called when you change the strategy, but try to have fidelity to the vision.” It needs a name because we do it all the time and even when we don’t do it, we talk about doing it. We can’t reason about it, we can’t… And so this is descriptive part, and then that’s the real work, honestly, is just saying what is… To me, the big question from the beginning was, what is a startup really? Why is it so different from a big company? Why are none of the best practices that I learned in my career work? It was like a survival mechanism, I want to know. And then once you’ve identified what it is and laid out the original principles of what you’re trying to do, the prescriptive parts are just a matter of deductive logic.
It’s like, “Well, given that we’ve made this hypothesis, what are the implications of that?” And of course, the prescriptive parts are important because that’s how you learn whether your hypothesis is correct. So, that’s the other thing I think people don’t appreciate. People call me a first principles thinker, which I love that, it’s a big compliment, but I feel like it conjures up for people this idea that I sit in a cave for years and years and conjure up the magic idea. It’s like, no, I wish people had all the outtakes of all the concepts I’ve tried. Lean Startup was not my first try, the conceptual vocabulary. It’s not the first time I got it right. It came through a lot of trial and error, a lot of trying to figure out what works and doesn’t work, first in my own career, in my own work, then trying to give advice to other people and trying to translate what worked for me to see what would work for them, and then to answer the questions that I thought were the natural ones.
Under what circumstances does it work and when would it not be appropriate? What predictions does the theory make that we could then go see if are true or not true? And then the bigger picture thing is, what are we really doing here? What does success even look like? That, I think, is the thing we all really grapple with is it’s very easy to come up with a theory or a method or a process that produces the maximum of something, but then it’s like, “Is that thing good? Is that actually helping us accomplish some kind of goal?” And so, I think that first principles thinking is very helpful for that too, but I don’t know. I guess to really answer your question, the true answer is I don’t know. I’ve just always felt like I was wired that way. I have a love of ideas and want to know why things work.
Misconceptions About the Lean Startup
Lenny: So to me, a couple of takeaways from what you just shared. One is just you’re wired to think in this specific way of thinking very logically, scientifically, I have a similar feeling where someone tells me a story… The way I started writing partly is people kept asking me questions about how Airbnb did stuff because they’re building their marketplace. I’m like, “Here’s what they did, but who knows if that was the way to have done it?”
Eric Ries: Exactly. Did they succeed because of that or in spite of that?
If I Could Rewrite the Book
Lenny: Let’s dig into the Lean Startup method and methodology. It feels like we haven’t heard an update on just how things are going, where it’s going, how popular it continues to be. I’m curious just, what’s your sense of the state of the Lean methodology in the Lean Startup movement?
What Is Missing: Getting Better
Eric Ries: Oh, it’s an interesting question. We used to do a Lean Startup conference every year, and then I stopped because of COVID and we haven’t resumed yet. So, it was like we used to have an annual opportunity to bring people together, assess the state of things, hear the latest updates and stories, and then I really feel the lack of it, I miss it. We tried to do online versions, but it didn’t feel right to me. And that’s partly COVID and the thing that our world has been going through the last few years, but it’s also partly I didn’t understand what it would be like to start a movement until I’d actually done it. And I remember the first conferences we did, the first events we did had the feeling of a religious revival. People were like, “We’ve got the new thing. We are going to stick it to the man.”
And I really thought, “My job is going to be to lead these hordes into battle.” I really had a martial metaphor. I used it just a second ago, combat of ideas. I was ready for, the old ideas are going to fight us to the death and we’re going to win it, and it’s not like that at all. We charge onto the field and no other army ever came to charge us back, just we won by default because the people who were doing it the old way didn’t really like it or want… It was very few defenders of the old way. There were very few people who came out and said, “Actually, no, Stage-Gate is actually still the correct product idea. You’ve got it wrong.” It never happened, and in fact, the fancy startup people went from dismissing it as totally pointless to complaining that it was overhyped and over before ever passing through the intervening stage of finding out what it was. That was really interesting. I’ll never forget, TechCrunch had an article, this is years and years ago about how Lean Startup is overhyped, and I was like, “You haven’t even talked…” That was the first entry point to talking about it. I was like, “Could you at least learn to use the word pivot correctly and then criticize it for being overhyped? Is that all possible?” So, it went from insurgent revival to just the default thing that people did. Even the people that disagree have to carry the meme in order to criticize it. So, it was fascinating to find myself on the other side of this thing, and I would meet young founders coming into the industry for the first time, and they’re like, “It seems obvious,” and I’m like, “It was only five years ago that people thought this was so crazy they’re yelling at me, now you think it’s obvious?”
It’s really interesting to have something go from controversial to obvious, and of course, tons of people who were really opposed to it at the beginning came back. Now they say that they knew it all along and they were my big… You see how it changes people’s minds. So, it’s hard to generate that religious energy and excitement around something that everyone views as obvious, and it really was hard for my… Just as an ego matter, I want the validation of the winning and the this, and I realized at a certain point that the victory is not that this fancy company used it or whatever, the victory actually is in the stories that people tell me about how the book was helpful to them.
That’s the whole ball game. No one had to lose in order for us to win. People don’t do this as much anymore, but occasionally someone still who wants to make a fight between Lean Startup and Agile or Agile and design thinking or this and that… And I tried really hard. If you look at Lean Startup, it’s full of citations and connections to Six Sigma and lean manufacturing and design thinking and customer development and DevOps and software craftsmanship and everything I could find. I really wanted to show how these things work together, and then we went through a phase where people wanted me to make it into a religion for real. People used to write me and say, “So-and-so on the internet is right about Lean Startup in they’re wrong. You need to make them stop.” It’s like, “What power do you think I have to make someone stop writing? I think they can write whatever they want.”
And then the other thing that people said was, “Well, Lean Startup is about this particular set of practices. If someone finds a new practice, they’re at heretic and need to be excommunicated.” And it’s like, “Guys, no, no, no. I wrote extensively in the book… Did you read? It’s a scientific theory, which means that whatever works is Lean Startup. So if someone comes up with something new and it works, we can’t be threatened and upset about it, we have to adopt it.” Well, when you have that attitude that whatever works is the thing that the truth is what matters and people’s quiet moments of helpfulness, that’s our metric of success, it kind of takes the hype and drama and stuff that drives press coverage and what takes it all out of it, and I don’t miss it at all. I feel much happier doing that than I ever did when… I used to travel a lot and do a lot more public stuff, and I think partly, again, one of the changes of COVID and conferences went away for a while, so it was a quiet time, but I feel like it’s actually been really amazing. That’s what victory looks like. They don’t throw you a parade and say the thing… And even though people who are your critics will still grumble about it, but at the end of the day, the new entrepreneurs who come in just take for granted that this is the conceptual vocabulary of entrepreneurship, and then they do with it what they will, and I’ve learned to really appreciate that. As much as my ego, of course, likes to get into fights with people. It’s like, “No, that’s actually better, and that’s what winning looks like.”
Audience Question: Is MVP Still Relevant?
Lenny: It’s so interesting that it’s just you’ve done so well in communicating these ideas that it’s just become part of the culture, and people don’t even think about, “Oh, Eric Ries came up with all these things.” I also feel like people push back at you on the things they disagree with, and then they never give you credit for the things that work great. It’s only like, “Oh, no, MVPs don’t work anymore.”
Establishing the Baseline
Eric Ries: Once a year at least, someone still writes to me and says that we should have used a different term for MVP, that the misconceptions around MVP are driven by the choice of the term. And I always write them back and say like, “Please tell me the better term and popularize it and I’ll use it. It’s a great idea.” I’m more aware of the limitations than anybody else. And I’ll tell you another thing that I think people don’t appreciate enough, everyone wants to be a thought leader these days. It’s like a disease. When Clay Christensen died, I knew him and admired him a great deal, and his endorsement of my work was one of the most important…
For me, from the old guard of management thinkers, for him to say that it was worthwhile was really a big deal to me. And yet there was this big debate, it’s now a little bit older now, about whether he’s responsible for all the bad things people have done in the name of disruption. And the first part of me is like, “That’s an outrageous slander on Clay. That’s so unfair, a bunch of tech bros like use disruption as a code word for all kinds of dumb stuff has nothing to do with his theory, has nothing to do with his work.” It’s like, “Have they even read the work, man?” I was one of his defenders, and especially after he passed, I was very emotional about it, it’s not fair.
But then the more that I’ve sat with it, the more I thought, “Gosh, maybe we are a little bit responsible for the things we put into the world,” and it’s terrifying because of course, you can’t control what people do, you can’t control… A lot of people are willfully ignorant. Whenever someone says something where I’m like, “Gosh, five seconds on Wikipedia, and they could have found out the truth about the thing they’re talking about.” So, we live in an era where ignorance is optional, that’s cool, but also a little bit scary. So I think taking on that responsibility, this is true for anyone who makes a product. I think it’s not actually unique to authors at all. When you make something, put it out into the world, on the one hand, you don’t have any control about what happens next, and yet I do think we bear a certain responsibility about, was it in fact a net positive? Was it helpful to people?
It’s part of what motivates me to try to really get it right and not just get it right for my own satisfaction, but get it right as evidenced by what happens when people use it. And that’s why testing and experimentation is such a big part of my life and my theory. To me, that’s how we square the circle of our responsibility in a highly uncertain world is you go find out what happened when people use the idea and if it’s not right, even after their misunderstandings, it’s still not right, then you’ve got to make it better and find a way to communicate it more clearly, so it’s kind of an eternal challenge, but I think an important one.
Why Less Is More
Lenny: Along those same lines, is there a misconception of Lean Startup or any of the other things you put out that just really frustrates you, that just is a recurring misconception that people just continue to get wrong?
Minimal Does Not Mean Low Quality
Eric Ries: I wrote an article, God, five or 10 years ago. God, I can’t even remember how many years ago now. It was like Top Misconceptions About Lean Startup, and it was like, “Number one, lean means cheap, so if you’re doing Lean Startup, it means you’re not raising money.” And it’s like, “That’s not true.” And then kind of I remember what they are… And they’re all still prevalent, nothing’s changed. I could write that same article today. It’s like, “Lean is opposed to having a vision. If you were really visionary, you wouldn’t do experimentation. Experimentation leads to local maxima,” and whatever.
I read it in somebody’s book that that’s what Lean is about. And I was like, “Man, did I forget to write it in the…” I was so gaslighted that I actually went back and cracked open a copy of the Lean Startup, and I was like, “On what page do I address this?” It was like on page nine in the introduction, one of the first things in the book, I was like, “Phew, I did say it,” so that’s pretty common, that turn the crank thing about optimization, that’s pretty common, and then pivots, people perennially misunderstand what a pivot is, and again, whether that’s a descriptive or a prescriptive thing is really tricky.
I feel like at this point people are just being willfully ignorant. It’s okay to criticize Lean Startup, that’s no problem. It doesn’t bother me. I think it’s good to do, it’s healthy to do so, but to just trot out those same old things, how does that advance the state-of-the-art? And how does that help entrepreneurs in any way? And I’m not even really sure what they’re trying to accomplish by doing that. That’s the thing that I’ve grown confused about is, what is the purpose of this discourse, other than just to kind be reflexively contrarian? I don’t know.
Lenny: And get some likes on Twitter/X.
Lessons From the Data Center
Eric Ries: Maybe you’re right.
Understanding the Pivot Concept
Lenny: So along those same lines, so you wrote the book 12 years ago. If you could go back and change something in the book, is there something that you would want to change?
Advice for Building an MVP
Eric Ries: Oh man, I would change everything about it. It’s actually why I won’t let myself do it. When I was a consumer of books, I hated it when authors would come out with a second edition that would destroy what was beautiful about the first edition, including the mistakes because in their attempt to put disclaimers on everything and make everything more nuanced, you’d totally lose the thrust of the argument of the thing they were trying to do, and their own becoming famous really messes with you psychologically. So, you don’t want to expose your readers to that. Go get therapy, don’t destroy the thing that made your book amazing. So as an author, of course, I think so many elements of it could be improved. The one that is the most glaring to me… Well, I guess there’s two. I’ll mention two just because they’re relevant.
One is I’m a little embarrassed, there’s a chapter towards the end of the book where I talk about the scaling of Lean Startup. So it’s like, “Okay, you’ve had success. Now, what are the principles for how you scale up a big enterprise now that has many Lean Startups within it?” And stuff like that, and I don’t think I wrote anything wrong in that section, but I really cringe to read it now because it’s so blithe and just like, “Step one, do this. Step two, do that.” It makes it sound so easy and so simple.
And of course, I’ve subsequently written a whole book longer than Lean Startup just to illuminate those principles and explain how they work in real life because I actually had the privilege of getting to go do that with companies at really extraordinary scale. So, I’m a little embarrassed now to be like, “I made it sound so easy.” I hate to mislead any entrepreneurs into thinking any part of entrepreneurship is going to be easy because it’s just setting people up for disappointment. And the other funny bit is, you know that meme on Reddit, the Star Wars meme where Anakin Skywalker says, “I’m going to change the world for the better,” and then pause, and then she’s like, “For the better, right?” I feel like that is the meme of our time for the tech industry. Every company was about changing the world and we forgot to ask, for the better? I always just took it for granted that our goal was to make the world a better place. That’s the values I was raised with. That’s what I thought we were doing, and I was railing about it for a while. I would get depressed. Remember, don’t be evil? And there’s been all these moments in the tech industry where it seemed like we were going to really do something extraordinary, and then just the things that these companies become is so depressing. And I was like, ” Well, I at least made it clear.” And then I was like, “Did I?”
Again, I actually went back and reopened the book and I was like, “Did I tell anybody that they should change the world?” And if I did, I assume I specified for the better, and it’s actually right there in black and white. Right in the introduction of Lean Startup, one of the… It’s like a throwaway line, “This is the technique that will give the entrepreneurs of tomorrow the tools they need to change the world.” And I’m like, “No, I should have said for the better. I didn’t know I had to say it. I thought it was obvious. I thought we all agreed.” So, I kind of feel like that is also a major oversight and I’m making light of it now, but it’s had catastrophic consequences. Not really, I don’t think Lean Startup has led people in that way. I think it has much more to do with other social forces, but I feel bad that I missed my chance to take a stand on something that has turned out to be quite important.
Lenny: So, you saw me tweet a call for people to ask questions.
Learning From Failure
Eric Ries: Yes, I’m really excited.
Lenny: So, I have a few questions from the audience. The first is around the concept of MVPs. So Karri Saarinen, who was actually on the podcast recently-
The Value of Failure Experiences
Eric Ries: Oh, great.
Balancing Craftsmanship and MVPs
Lenny: … Founder of Linear, had a question around MVPs. And his question is essentially, If you still believe that MVPs, essentially shipping very basic versions and iterating is the way to go, given that user expectations have risen a lot in any markets, iOS apps, for example, web productivity software, things like that.
Craftsmanship Is Not Against Experimentation
Eric Ries: I love that question because first of all, I get it a lot, and also it’s important for us to really understand. People think that MVP is about a specific tactic. So, an MVP is like a bare bones, stripped down the thing that will crash your computer, but it’s not anything to do with MVP. MVP is simply for whatever the hypothesis is that we’re trying to test, what is the most efficient way to get the validation we need about whether a hypothesis is true or not? And so, part of a successful test is to understand what are the demands of customers in our market. Now, the funny part is people have a real fear of failure in product we all have this deep, deep, deep fear. So the first answer is, you think user expectations are high in the iOS App Store, you should try the purchasers of battery backup systems for data centers or people who build deep sea oil drilling machines or jet engine.
I’ve worked with companies where the standards are actually really high and it makes iOS apps seem like pretty basic. It’s like, “Well, nothing blows up like your app,” but even in these really high end industries in really high stake situations, the cost of offering someone to buy something that they don’t like, it’s really not that high. They just say no, and a lot of MVP is just about containing the liability of making a mistake. So yes, one of the things I really encourage people to do is go find 10 customers, not 10 million, but find 10 customers. Get them to use the product and have them tell you what’s awful. That’s okay, and I used to be so upset… When I used to launch a product if customers would argue with me, people used to be like, “You idiot, I can’t believe you launched this product without features X, Y, Z. They’re so essential.”
And I would be like, “You’re wrong…” I’d got to get mad at them, but then I got a little smarter. I’d be like, “X, Y, Z, you say?” And I’d be like taking notes, “Oh, interesting.” Now, sometimes X, Y, Z would be the exact next three features of my roadmap, and again, I would want to argue with them as, “I know knew that…” And I eventually used to be like, “Oh, thank you for that wise feedback.” And then when we launched X, Y, Z, they would feel a sense of ownership. “Wow, I’m the one who told him to do that. He wouldn’t have even done it.” But of course, the thing we’re really afraid of as entrepreneurs, I think if we’re really honest, it’s not that people say, “You idiot, you didn’t launch without X, Y, Z,” people will say, “You idiot, you launched without A, B, C.”
And you’re like, “What the hell is A, B, C?” Never heard of that. That’s the much more common response that I get. So, let’s grant the premise of the question that expectations are high, and if we launch an app and it doesn’t meet people’s expectations, they won’t like it. First of all, what really are the consequences of that? In an early part of Lean Startup, we had to do a lot to explain to people the difference between a product launch and a marketing launch. And you can launch the product to customers without telling anybody that you did that. It’s okay. No one’s going to find out, it’s not going to be big news. And of course, once you’re a big successful company, that’s a totally different issue where now you put your corporate brand on stuff, but if you’re an actual startup, nobody cares. So, it’s okay to try it.
But the other thing that I think people don’t appreciate is that MVP is very flexible. So, if you tell me… When I work with entrepreneurs, I’m like, “Listen, I’m sorry, our customer is… The bar is here and the thing just has to be that good.” I’m like, “Excellent, and it can’t have very many features, I guess, can it?” They’re like, “Well, what do you mean?” “So, you just told me the bar is way up there, so we better do just the one critical feature at that level of quality, and then if that works, we’ll do the next one.” And then of course they’re like, “Oh, no, no, no, I need to do all the features at all the quality.” It’s like, “Well, then what are we testing?” It’s like, “By the time you find out whether that’s any good…” And if you look at the pivot stories, I’m sure you found this too.
Very often 90% of the work in the first version is thrown away no matter how high quality it was. In fact, as an engineer, that was part of my original disillusionment with the Agile software movement then ultimately with Lean Startup is that people promised me that if I followed these best practices, I would eliminate waste from my work, but if I build something that customers don’t want, that’s the ultimate waste and nobody protected me from that. So, I don’t really care how high quality the initial MVP is because quality is defined by the customer. And if we don’t know who our customer is, we literally don’t know what the word quality means. So, that’s where I think founders really run into trouble and everyone’s like, “Well, I’m Steve Jobs, I’m this, I have the magic taste.” And it’s like, “Well, first of all, do you have the years and decades of failures and experience from which that taste derives?”
If you do that then I’m like, “I’ll give you the benefit of the doubt.” And if you don’t then, “Definitely don’t rely on that, definitely do some experimentation. Go get those scars.” But even if you have a taste, if you’re completely right, what can it hurt to double check? My view was like, “Oh, let’s have 10 customers look at it and have them throw up about how it’s so horrible, that at least confirms that we’re on the right track.”
And then we call it establishing a baseline when we talk about innovation accounting and the metrics side of Lean Startup. Let’s go find out how bad the situation is right now, so that we can then know that we improve from it. Because the thing I actually see from people who in the name of quality will wait a long time to ship a product is they ship it with really high quality and nobody uses it, and now we don’t know why. Is it because our idea of quality is wrong? Is it because the value prop is fundamentally… I feel like building a high quality wrong value prop is nothing to be proud of. There’s no craftsmanship in that. I once had a company that built a really beautiful product, amazing land, everything was awesome, but there was a JavaScript error where when you clicked the download button, the signup button, whatever, it didn’t click. So, they launched this thing and it had 0% signups.
Getting Fired for Doing Right
Eric Ries: … click. So they launched this thing, it had 0% signups and they’re like, “It’s a catastrophe. We made the wrong value prop.” It’s like, how do you know? From one experiment you can never know. You have to be willing to do a series of experiments. Again, the logic of MVP is just, products take time to build. The best products in the world, everything takes time. It takes a month, a year. AI will mean I could build any product I want instantly. But it’s not instantly, it’s going to take time.
During that time that the product is under construction, there’s going to be some moment in time when you have half the product. Just logically speaking, you first have 10%. When you have half the product, let’s have customers use half the product. Well, it doesn’t cost you anything extra. You had to do it. You build a natural experiment anyway. Have them use it and tell you it’s only half as good or it’s unusable, because it’s so bad. Excellent, good.
The startup literature is littered with examples of people where less is more, where the crappy small thing they did turned out to be more valuable than the competitor. I’ll just give you one example, one of my favorites. I come from 3D graphics and avatars and metaverse stuff, which of course, now is bad. It’s one of these things goes through these hype cycles and I’m like, “Are we actually in the hype cycle or are we the drops?”
The Zombie Company Dilemma
Lenny: I think it’s trending up right now.
Eric Ries: Are we trending up again?. It’s really hard for me to keep track. Even while I was working on it, every few years it was either the future or nobody wanted to do it. I bet sales of [inaudible 00:35:25] crash, you’ll probably have that cyclical thing to them. When it’s hot, people read it, otherwise not. Anyway, we built a version, our first version of IMVU we built with 3D avatars in a cafe or in a scene. The avatars were all anchored in place because we didn’t have the technology called inverse kinematics. We didn’t have the technology that would allow an avatar to stand up and walk. When you do this 3D grab, it has to work perfectly. The feet have to actually glide along the surface and it doesn’t look right. You wind up in the uncanny valley and people find it horrible. So we’re like, “Can’t have that.” So we just have the avatar to sit there. People felt the product was claustrophobic. They would always be like, “The Sims lets me move around. Why am I stuck here in the chair?”
That was a very common thing in customer feedback. We were like, “We can’t afford to do the expensive thing. We couldn’t do it, couldn’t do it.” One day as a hack, as like a little demo, I was like, “Well, I’ll just change it so that when you click in the scene to a different spot, we’ll just stop rendering the avatar in the first place and we’ll put it in the second place.” We didn’t even do the proper cut. The camera would zoom over there. It was really ugly. I was really embarrassed about it. I was like, let me just play around with it and see what happens.
We shipped it to cut and I was like, this is below the quality bar. I felt bad as the engineer. Everyone involved felt really bad, but we had committed ourselves that we’re going to try to experiment with stuff. We’re going to see what happens. This is before Lean Startups existed. We didn’t even have the concept of MVP. We’re going to try it and just see what happens. We shipped it and nothing happened. No one cared. I was like, whew. Then we used to do these daily tracking surveys. We’d just sample people and ask them what they think of the product. In the tracking surveys, people started to say that IMVU was the most advanced avatar product they’d ever used.
People would be like, we used to be compared to the Sims unfavorably. Now people would be like, more advanced than the Sims. I was like, what does this mean? How did we get more of, did we improve our 3D graphics design? No one could understand why they were saying that. We used to do customer interviews. Finally, a customer was like, “It’s so great, man. In InView, when you want to go somewhere, you teleport. The Sims, you got to wait for the guy to get up and walk around. It’s so annoying.” We were just like, “Oh, my God. Embarrassing.” We didn’t do inverse kinematics for a really long time.
Pivot Frequency and Data
Lenny: I love that story. Something that I’m learning as you talk about MVP is that the idea of minimum is very fluid and very context specific. Essentially, the message isn’t, build something low quality and simple. It’s, figure out what is that minimum bar for your market and product. The message is just build as little as possible, whatever that is for you.
Vision Forms Through Tradeoffs
Eric Ries: Yeah, as little as necessary. It’s fascinating. I’m not joking, I’ve worked on jet engines and robotics and cancer therapies and all kinds, I’ve worked on stuff anyway, and LTSE, it took us 10 years to build the MVP for LTSE. So it doesn’t necessarily mean cheap low quality. It’s just whatever your hypothesis is, whatever you actually believe is going to work, let’s put that thing to the test. That’s all I really want. When I work with companies, first time founders especially are usually very stubborn about this and they don’t want to do it. It’s like, “Look, you tell me what you believe. I don’t have to be right and I want to just double check that you’re right by however we do it.” And then once people actually start to turn that idea over in their mind, they often will be like, “Well, you know…” A classic one is like, we can’t sell the thing until we build the physical equipment.
It’s like, is your product bought in a showroom where people see the physical equipment? No, they buy it from a brochure. So it’s like, “Okay, well how about we send them a brochure and ask them to pre-order it?” And then it’s like, what? I can’t do that. I can’t have a crappy brochure. It’s like, who said a crappy brochure? We can build a really high quality brochure in a tiny fraction of the time it takes to build the actual machine. And then it’s like, well then what are you talking about, it has really good fonts and it has pretty pictures? Maybe. What’s your theory, right? You tell me what do you believe about… And they’re like, “Well, actually what matters is the quality of the specifications of the equipment in the brochure.” It’s like, now we’ve got the leap of faith assumption. Let’s go find out.
And they’re like, but hold up. But we don’t know what those specifications are going to be yet. It’s like, well, I saw this 500 page plan over here. What do you want it to be? It’s like, well, it’d be really great if customers would accept this set of performance characteristics. Well, let’s go promise that because maybe they’ll be like, “That’s great.” And I’ll give you one example. I was mentioning data centers. There’s a team that sold equipment and data centers. They had a serious high-end team of PhDs working on research to make the thing way more efficient and they’re going to win a Nobel Prize in material sciences in order to make this product a reality. And I convinced them, instead of paying three years and $18 million, so you don’t make the thing, let’s go take a brochure, take it to some customers.
And I’ll never forget, they took the thing into the brochure, they took it into customers and customers laughed them out of the room and it was nothing to do with anything they cared about. It was nothing to do with efficiency where they’re like, we would never buy this. And they were like, “But you haven’t even looked to page two to see how efficient it is.” They were like, “We don’t pay for efficiency, we build data centers. How much they cost to operate is somebody else’s problem.” You’re like, “Oh my God.”
That’s the thing people don’t appreciate, is these insights are so obvious in retrospect that everyone’s convinced that it won’t happen to them. When I teach students at Lean Startup, they’re like, “This is so dumb.” They do one case after another in business school and they’re like, “Dummy, dummy, dummy, dummy, dummy, right?” They’re like, “I would never make that mistake.” And I’m always like, “Look, you don’t have to argue with me. Just call me back when you get a job. And when you make one of these mistakes, remember we had these conversations and go read the book then.” And I guess by random chance there are some people in the world that just never made a mistake.
The dice, you roll enough dice, somebody had the experience that dice came up double sixes every time. Okay, so there are those people and they give a lot of the advice in the world, but I don’t know that it’s necessarily those of us who don’t have the double six luck should necessarily be following it.
Lenny: This will be a really good segue to talk about pivots because again, a big part of your message is a lot of the work you’re doing is just going to go to waste anyway. Why spend all this time building it? But just to put a final point on this idea of MVP, because a lot of questions came in around just like, what is an MVP? Ben Silbermann, the co-founder of Pinterest, has this quote, he didn’t ask a question in our thread, but he has this quote-
The Unreliability of Memory
Eric Ries: [inaudible 00:41:47]
The True Meaning of Pivoting
Lenny: Yeah. And his quote is, the hard part about MVP is that you don’t know what minimum is. You don’t know what viable is. And so just to give founders who are trying to figure out what that line is, some very succinct advice, how would you describe just helping them understand, here’s how far you should go with your MVP.
Vision Is Discovered, Not Chosen
Eric Ries: He didn’t say, you also don’t know what product is. That’s true. I can’t tell you how many times a service has turned out to be a product or vice versa. People embody their insight about customers in the wrong way all the time. That’s super common. That’s funny.
All the criticisms of MVP are really just people expressing their frustration with how uncertain startups are. It’s like, yeah, it really sucks. This is not a good way to make a living. If you want to retain your mental sanity, if cognitive dissonance, if you find that painful, this is not the right job for you. Because it’s really hard to hold in your head, I have this ambitious vision. I know what’s going to work. You got to do confidence, get everyone together. There’s no substitute in this world for a human being stepping forward and just saying, “I am going to make this thing happen.” And if you do that, you’re putting yourself away on a limb. It’s really difficult. It’s really hard to admit that. And I also do not know how to make it happen. That’s the truth. Whether you admit it or not, that’s the truth.
So that’s why I say let’s just double check. So simple, succinct advice. The first tip is, write out the list of features that are necessary in your MVP. Cut it in half and cut it in half again and build that. Honestly, if you just do that, that’s really not that bad. The nice thing about MVP is it’s one of those U-shaped curves where if you get even in the vicinity of optimal, you’re going to be fine. And most people’s natural idea of what is necessary is so laughably wrong. People are naturally off by usually one or two orders of magnitude, right? Very common for you to meet a team that is trying to do 100 times more work than is necessary to test their assumption.
So first thing is just, try being uncomfortably small in the thing that you actually make. And if you’re having trouble with that in the startup way and the SQL Lean startup, I really go through because it’s really more for a corporate large company audience. I really go through, okay, if you want to have a process for determining what to do here, you should first brainstorm your leap of faith assumptions and categorize them this way and select them like this. Okay, now for each assumption, brainstorm the metrics you might use. Select the metric you want, what is your learning metric, and then from there, brainstorm, MVP, score them according to this formula and then use that to help people.
Anyway, people can follow that if they want. The thing I learned is that if people are having trouble coming with MVP, the problem is usually upstream in the chain of deductive reasoning from the vision down to the MVP. Usually people are just not super clear on what do I want to learn, and the most common reason is because people don’t want to admit that they don’t know. So as an entrepreneur to say that we want to test to find out what customers want is really hard.
I’ll give you an example from my own experience. LTSC, I said, it’s taken us 10 years to get the thing approved and going. It’s a really very difficult product to build. One of the early iterations of it that didn’t work, we were building on partner infrastructure. It was a very different product back then. I don’t want to get into all the legal nitty-gritty, it’s kind of dry. But anyway, it took us six whole months. It took us time, we had to raise the money, build the team, do all stuff, get all the lobbyists and lawyers, everything in place, and then we had to go do this partnership agreement. The partnership agreement itself took six months to get in place, and then we had to go to the SEC and get approval. And so it was a very complicated multi-stage process. Probably took, I think I just described two years of my life end to end there.
And the whole time I was talking to the team about our goal with this whole thing is to learn what is actually necessary to get customers to list on our stock exchange. Our goal was learning. People were like, but I thought our goal was to get this contract done. I thought our goal was to, it’s like, well, yes, but the reason we’re doing those things is so that we can learn whether this is actually going to work or not. And so I remember, I’ll never forget during the contract negotiations, people were like, we got to take an extra week to deal with this point, and I’d be like, “We don’t have an extra week.” It’s like, but this thing’s taken two years. What’s a one week? I’m like, “No, every day is so [inaudible 00:45:54] we have to get the learning as soon as possible.” And I was constantly haranguing everybody involved.
We got to go faster, faster, faster. You’ve got to give on this point, people were walking all over us. We were getting just destroyed in the negotiation. I was the weakest lamest person, rather just like, we got to raise the money, anything to get speed. Anyway, long story short, we get all the way to the end. We get the thing approved by the SEC staff and then we get ambushed. And long story short, the SEC withdraws their approval and the whole thing is dead. The partner, we have to break up the partnership agreement, everything goes to hell. And I thought the company was going to die, and it was a totally miserable time. It was one of the worst things that ever happened to me in my life. I was totally upset. And in retrospect, I can laugh about it because it turns out, of course, it was the worst thing that ever happened to me, but it’s actually the best thing that ever happened to me, as a lot of startup things are.
Because from that, we were able to pivot into the thing that ultimately got us approved. The reason I tell this story is every bit of work we did that went into that partnership agreement for almost two years turned out to be a waste of time because in the end, the partnership agreement fell apart. So if we had won a few extra [inaudible 00:46:58], my margins would’ve be higher if we’d extracted more fees from the partner or we would’ve had less liability. People always like, well, we have the liability for this and liability for that. And I think about all the things, this thing was 100s and 100s of pages long to document. None of it mattered. In the end, the partnership was never consummated, no customer, nothing ever happened.
And the interesting thing and reason to the story is that the people who went through that experience with me were so much faster on the next thing that we did. It was really amazing to see because they were like, “Now I understand what you mean.” Most people live very coddled lives that have not actually really dealt with this kind of failure in their business career. Obviously in their personal lives, they’ll have all kinds of trauma. And you know, they’ll work at a crappy company and have a horrible experience, but they haven’t experienced this specific thing where you are the hero of the hero’s journey. You are the protagonist of the story. And the thing is not just a setback, but you just die.
It actually fails. It’s so psychologically damaging that even people that have had that experience, a psychological defense mechanism is to tell themselves that that didn’t happen. And you know it better than anybody, how many startup founders are adamant that they didn’t ever fail, they didn’t pivot. Everything’s exactly the way they saw it at the beginning. And it’s like, I’ve been around enough startups to know how often that is just not true, but we need to tell ourselves that story. But anyway, the number one lesson of the scientific method is if you can’t fail, you can’t learn.
And so if you set yourself up in a position where psychologically you can’t admit that something didn’t work, you can’t learn. And that really is the essence of MVP. It’s just like whatever the catastrophic thing that’s going to destroy your company is going to be, we just want to find out as soon as possible so that there’s still time to do something about it. We don’t just die and move on.
Lenny: I’m actually seeing this a lot right now. I’m an investor in a bunch of startups, and so many of them are failing right now, and they built such beautiful products, such amazing products that they spent years just like crafting, making so great. And then it’s all for naught. And there’s actually been a theme on the podcast recently around the importance of craft and design. And I’m curious just to get your thoughts on just when it makes sense to invest deeply into the craft and experience of a product versus the minimal viable product approach. Do you start very simple and then invest, anything’s there?
Setting a Fixed Time Window
Eric Ries: Yeah, well, I think that’s a great question. People use the word craft to refer to so many different things. I personally find a tremendous beauty in simplicity. So I would much rather people do one thing extraordinarily well than 20 things half-assed personally. And how many products do you use? You’re like, God, there’s like 92 buttons on here of which I only ever pressed one. Could you have made that button have a really nice feel? I would’ve sacrificed all this other stuff. And I’ve actually worked on microwaves and refrigerators and stuff with actual physical buttons. And even your crappy kitchen microwave that you paid $60 for at Walmart or whatever, if you’ve never seen one of those things manufactured, actually the amount of craft that goes in behind the scenes to wire up every one of those buttons that are never pressed by a human being, it’s actually really painful if you start to think about, wow, that’s a lot of human energy being wasted.
So the elements of craft I think are critically important. First of all, part of craft is perfection. Perfection is an attribute of quality. And we have to go back to the first thing. In order to know what perfection is, you have to know what the customer believes. You have to know, you have to be perfect for some purpose for some person. Some people will just say, “Well, I’m just going to build for myself and that way I solve the problem.” That’s totally fine, if that’s true, if that’s you, go nuts, you might only have a market of one, but you’ll be happy. So be it. And there are startups like that where it’s like, well, I don’t care how big my market is. I’ll just build for all the people like me and I have a huge tribe, then I’ll be hugely successful. If not, I’ll just be an artist doing my thing. I have total respect for that.
That’s like the ends and the methods are totally aligned. Where people get into trouble I think, is when they want to use craft as an excuse to not get any feedback. So what I tell people is first of all, there’s an element of craft that is just like whatever you do, how are you going to go about it? Whether you’re building an MVP or a fully featured whatever thing you’re doing, I’m a software engineer by training. My code is a certain way, I have a certain craft in the way that I write software because I think that’s the best way and that produces the best outcomes. I will never compromise that. And compromising that in the name of going fast is stupid. Write crappy code. You spend more time firefighting than you do writing code in the future.
So a certain amount of craft is just about keeping your options open. And actually it’s very much favorable to speed, using the right primitives, having the right abstraction layers, using the right tools in the right way, having high quality people in your team, all those things that you go faster, not slower. And a big part of Lean Startup is about helping teams get into that optimal configuration by acting as a speed regulator. Because there’s such a thing as optimal speed, which we can get into if people are interested.
But another part of craft is just an excuse for just avoiding feedback, making things pretty in the way that I want them to be because I want them to be that way. And even there, when I meet people like that, people are just like, “Sorry, my artistic ethos is this and it has to be this way.” Then I’m like, “Excellent.” Then that’s not one of your leap of faith assumptions that you’re willing to test. So be it. Don’t pretend to test it, just be like, I’m not, we’re just going to do it this way because that’s how we do things and it’s fine. And I think when company has a core value, I’d rather go out of business than violate my core value. I also respect that too.
But then that to me, that means having a respect for craft, for design, for perfection, that increases the need for experimentation and testing for all the other stuff, because we’ve already invested a huge amount of our energy and force into this thing that we really care about. Well, then the other stuff, and do we also have to do this fancy press campaign? Do we also have to go put the founder’s face on billboards and put them up around town? Are those all part of your craft? I just feel like people use that as a catchall excuse to do all kinds of stuff.
But I think the idea that craft is a signal to your customers about the things that you actually care about and that you’re going to take care, it is part of making a promise to customer that they’re going to believe. If that’s your thesis, that doing an MVP that lacks craft doesn’t test the thesis. So I think people want to set up a dichotomy here that doesn’t really exist. I’m totally in favor of that when people have that authentic desire. But I also meet a lot of people that are just faking it. They don’t have taste, they don’t have a design ethos, they don’t have… So if you don’t have that stuff, where are you going to get it from? Well, good news experimentation will get you there if you let it, because you’ll eventually learn the habits of mind, the patterns of the things that work, that make sense.
Lenny: I really like that advice of, as a founder, build the company you want, if you want to build, why create a business and a company you’re not excited to be working on? And if you end up failing, it’s on you. That was a big mistake it turned out, but at least you tried and you’re not building a company you don’t want to work at. And then as a product leader, I don’t know how much, if you have as much control to build a product the way you want, but try basically.
A New Paradigm for Effective Experiments
Eric Ries: I always tell people to just get fired for doing it the way you think is right. There’s nothing that was a faster career accelerator than being known in the industry as the person who stands for that principle so much that you got fired over it. The people that magnetizes to you are going to be way happier. You’re going to be so much happier being a place that appreciates that. That was a big career accelerator for me. Not that I actually got fired, but I was willing to be fired for my beliefs. And I remember sitting there, I was going to a board meeting where for sure I was going to be fired. And I was like, “Well, I just don’t believe in building products the conventional way.” This was again before Lean Startups. So I was like, I didn’t even have a language for it, but I had a strong conviction that the way I was doing it was the right way.
And I was like, well, I’ll be known as the guy that got fired for that and eventually someone will want to work with me because that’s what they want. If nobody wants that, then I’ll go find a job in a different industry. I don’t want to be an industry that doesn’t… That’s how I want it to be. And I think that people don’t appreciate the extent to which… People act like having a startup fail is the worst thing that can happen to you. And man, that’s not even in the top 10. It’s bad. I’ve done it. It’s awful. Okay, it’s really bad. But far worse is to be in a company that won’t die, a zombie undead company that you hate but you can’t leave. Have I met people like that. And we’re having a mental health crisis among founders that’s not talked about enough. People started to talk about the downside, mental health risk. Obviously the stress of being a founder is hard, but look what’s happening to the people that are so-called successes. When you build a company and you sell it out and it becomes something that you find abhorrent, yeah, maybe you get rich, but it’s not good. And I mean, the other day a friend was telling me about a friend of his who sold this company, exited for, I say $300 million or something, made a lot of money and committed suicide not that long later. People follow the Tony Hsieh story. I knew Tony, it was a horrible tragedy. What’s going on? So I just feel like there are worse things than having your startup fail. And so building a company you hate, that becomes a maligned force in the world that you have to go pretend you weren’t involved with or you feel complicit in what… That’s way, way worse. And I wish more founders would take that more seriously early on.
Lenny: That is such an important point and something I see a lot too, just companies, the founders started that are just alive, but they were almost never going to succeed, and they’re kind of stuck. They don’t want to shut it down, they don’t want to give money back. What advice do you give those sorts of founders that you feel like are just stuck in a zombie company like that?
The AI Era and Lean Startup
Eric Ries: It’s really hard. I mean, of course if it’s going to fail, you should shut it down. But everyone’s like, but I thought it’s always darkest before the dawn. And it’s like, sometimes it is. But you know what else that’s really dark? When you’re dead. So everything comes back to, I used to tell people, the most uncomfortable laugh line in my early talks about lean startups, when I talked about, people used to talk a lot about steep jobs in the reality distortion field. And I would be like, okay, you’re in the reality distortion field. You’re on the flat part of the hockey stick. How do you know the business plans? I actually lived through this once. There’s a business plan, the business plan says six months after launch, we’ll have almost no customers and almost no revenue. And that’s when the hockey stick will commence. And I remember being there in month six being like, we are on time, on budget this plan is cheap.
Everything’s happening just like the plan, hockey stick to commence. I’m walking to my watch. And then we just stayed on the flat part into the far horizon and people would laugh and I’d be like, okay, now if you’re following a startup leader, or you yourself are in the reality distortion field and you’re on the flat part of the hockey stick, how do you know if it’s ever going to stick? Or you’re just going to be… And people would be like, “Ha ha ha ha.” And that is a really scary situation. It’s like we all have these moments of doubt and sometimes the right answer is to overcome the doubt and go for it. And it’s actually part of the genius of the startup system. People people criticize venture financing and all the problems with the Silicon Valley and of course, we do it to ourselves so there’s plenty of things to criticize.
One of the genius things about it is at the end of the day, you do run out of money. You’re forced to declare bankruptcy, that’s actually a really positive feature of the system. So a lot of founders just can’t do it. But I will say, the thing I try to ask people to do is really take some time to introspect and just ask yourself, do you actually want to be here? If you could magically wave a magic wand and you were to start a new company right now, would it be this one? If it’s not, there’s no indentured servitude anymore. It’s okay. It’s okay to move on. And then the question is, then everything moves into a tactical conversation about shut it down, give the money back, do a really hard pivot into the new thing.
It depends a lot on the specific factors of the situation, but I think we do founders a disservice if we act like this is easy, or I mean, and it’s gut-wrenching and it’s really difficult and not to be too spiritual about this, but it really goes right to the issue of ego identification. For so many people, they are their company, and so for the company to die, they also must die. It feels like you’re going to die. And I think that’s part of the mental health problem. So people as they get older, as they mature, as they have a spiritual practice, I hope they can learn to dis-identify with the company and not feel like it’s so personal, so that it is possible for the company to die without them dying. And then they can take what they learned into the next company or into the next pivot or the next iteration of what it should be.
Or sometimes the right thing to do is just to step aside and let a different CEO take over. There’s so many ways you can go about it. But as with any pivot situation, the hardest part is always just to admit the facts as they are. And then you do it yourself versus just like, what is the situation? Is it working? I can’t tell you how often people just can’t accept it’s not working. And you get investor updates. How many investor updates have you ever received where it’s like, we’re crushing it, everything’s going great, and the very next investor update is, we’re out of business, giving money back. It’s like, well, your last investor updates said you were crushing it. There was absolutely no indication there was any problem. No, of course it wasn’t an indication, but we’re just lying about it. So don’t be that person.
The Practical Value of AI
Lenny: That’s actually an amazing segue to where I wanted to move to, which is around pivoting. So I’ve done a bunch of research on consumer startups, B2B startups, and the numbers I see, and I’m curious if you see similar numbers is around a fifth of consumer startups, ended up pivoting to a completely different product and almost half, like 40% of B2B companies ended up pivoting to a completely different product. Does that sound right from your experience?
Fairness in AI Driven Markets
Eric Ries: Yeah, that does sound right. So it’s interesting, I used to do this exercise for founders where I would tell them the story of IMVU two different ways. The first way is one in which we got everything completely wrong and just sounds like we were to complete idiots and we had to pivot so many times. And then I would tell the same story again, but where we got everything right and we were geniuses and everything that happened vindicated. And then people wouldn’t believe me. I was like, “Both stories are true.” Choosing which elements of what actually happened to emphasize and where to put your weight, you can tell the story. So yes, I would say there’s the percentage of people that know that they pivoted and then there’s another percentage of people that didn’t admit it, but they actually did. And it’s really hard to assess from the outside.
I mean, I know this from having worked with enough startups where I know the outside story and I know the inside story, and they’re just really different. I mean, Paul Graham used to boast about that as one of the features of Y Combinator, that the founders they would get to come to the dinners would tell you the true story of what really happened, and that was critical information you couldn’t get anywhere else except at Y Combinator. And I was like, wait a minute, is he saying that everyone just lies on stage when they tell the story publicly? I don’t know if that’s something to boast about. You’re calling all your guests liars. And then I’ve hosted a lot of those events. I’m like, oh yeah, people lie on stage. I see. So I think part of the problem is just trying to find out the truth of what happened is really challenging. But there’s another element that I think is interesting, which is, so pivot is defined as a change in strategy without a change in vision. So we have this idea that the founder has the vision, then they try to figure out how to make the vision happen and they find a different way, but the vision stays constant. But the truth is a little bit more complicated than that, because the founder’s vision also evolves. Founders discover what the vision actually is through the process of building the company. So again, not to be spiritual about it, but I actually believe that building a startup is in large part a process of self-discovery. You actually are discovering what is true about the thing that you yourself believe. And partly that’s because when we fantasize about the vision that we never have to make any trade-offs, the product is infinitely popular, infinitely cheap, everything’s great.
It’s just like there no trade-offs to be made, where real life forces us to make trade-offs. And then you find they’re like, “Oh, I have to choose?” Would I rather have a high-end product only for really rich people or would I rather have a product for this industry or this sector or this group or that group? It can be painful, but also founders learn to talk about the things they care about over time.
So I mean, go on YouTube. If you’ve never watched the interview with Zuck on his couch at Stanford talking about Facebook, it’s remarkable. Just some college news person just happened to catch an interview with this guy at a critical moment in his life when he just didn’t know what he was talking about yet. And now he’s really ambitious. Now Facebook’s vision is this really ambitious thing and people are very at pains to say that he always had that vision from the beginning. And he talks about it all, and that may be true, but the documentary evidence exists, and I’ll just give you one of my own stories about that.
In one of my companies, we moved offices relatively early on because we started out in a really, really crappy small office and we moved to a larger space. And I remember sitting in that office when there was just five of us and we sat down and we had a meeting where we specked out what we would now call the MVP as well as the longer term vision for the company. And we had this whiteboard and all of us debated what features should be in and out or what should be on the whiteboard. And I can remember writing down, I have a visceral, I have a physical memory of writing down the things that were going to be necessary for the company to succeed and being right, and my co-founders arguing with me about not having their perspective and their being wrong, and I’m like, good thing.
It’s one of the things I remember telling this story. Anyway, we moved offices shortly after that time and that whiteboard was lost, and then years later we moved offices again and someone was cleaning out an old storage closet and that whiteboard was just happened to have gotten shoved in the closet. Someone pulls it, they’re like, “What’s this?” And I looked at it, I’m like, “Oh, that’s the whiteboard we used to hang in our main conference room, and it still had the notes from that meeting sitting there on the whiteboard.” I was like, this is so cool. Historical moment. I’m taking a picture. And then I’m like, wait a minute, this is weird. It’s in my hand. It’s recognizably in my own handwriting, but many of the things that I’m writing there are horrible ideas. I’m like, that’s not what I thought back then. I was on the other side. That’s my co-founder’s bad idea. That’s not my idea.
And I had this realization that it’s not just that I learned something through the course of building this company about strategy. My own understanding of the vision of what we were doing was different, but the thing that’s so critical to understand is I even to this day have no psychological memory of having changed my mind about it. I still have the original memory of writing those things on the whiteboard, and I really thought someone was playing a trick on me at first because what I could witness with my own eyes on the whiteboard contradicts my actual memory of the thing. And I was like, which of us is right? And of course, I had to admit, the physical evidence, you got to win out.
But I think it’s important also to understand how much that is true. And that’s like, again, go back to why science is so important. Why do we write all the elementary things about science? We write our hypothesis down, we subject our things, we do that because human psychology is so buggy and people who are fighting with their generative AI chatbots and stuff are starting to see, those are nothing but a reflection of human theory of mind. So we’re starting to see it in alien creature that, wow, our intelligence has all these weird behaviors and gaps and things, and of course we want our robots to be perfect, but it’s like, well, first look in the mirror and realize that you yourself carry all those cognitive biases and you’ve got to put in place a structure to get around them or they’re going to destroy you.
Lenny: That is an incredible story. It makes me think about, I think it’s exactly what you said, not only are the stories you hear often intentionally not accurate or true, and they’re kind of a myth that have been created, but also you forget exactly what happened and you tell a story that didn’t really happen. That’s wild. Imagine a place where you can find all your potential customers and get your message in front of them in a cost-efficient way. If you’re a B2B business, that place exists, and it’s called LinkedIn. LinkedIn ads allows you to build the right relationships, drive results, and reach your customers in a respectful environment. Two of my portfolio companies, Webflow and Census, are LinkedIn success stories. Census had a 10x increase in pipeline with a LinkedIn startup team. For Webflow after ramping up on LinkedIn in Q4, they had the highest marketing source revenue quarter to date.
With LinkedIn ads, you’ll have direct access to and can build relationships with decision makers including 950 million members, 180 million senior execs, and over 10 million C-level executives. You’ll be able to drive results with targeting and measurement tools built specifically for B2B. In tech, LinkedIn generated 2x to 5x higher return on ad spend than any other social media platforms. Audiences on LinkedIn have two times the buying power of the average web audience, and you’ll work with a partner.
The Responsibility of Next Generation Founders
Eric Ries: Have two times the buying power of the average web audience and you’ll work with a partner who respects the B2B world you operate in. Make B2B marketing everything it can be and get $100s credit on your next campaign. Just go to linkedin.com/podLenny to claim your credit. That’s linkedin.com/podLenny. Terms and conditions apply.
Lenny: So you talked about pivoting. I never knew this about the term pivot is that you keep the vision the same. And I actually have this list of companies, B2B companies in this case, that pivoted. In their cases it was very different. Let me go through them real quick. And I’m curious, what do you call it if they don’t keep the vision consistent? Segment started as a university classroom lecture tool. Loom, which just sold for a lot of money to Atlassian, started as a marketplace for companies to hire subject matter experts. Slack famously started as a game called Glitch. Box apparently started as a box to put photos and content in on Facebook. And Retool started as Venmo for the UK.
Balancing Power and Trust
Eric Ries: That one I didn’t know.
Lenny: I found that out through Twitter where one of the YC partners replied and shared that story with me because I also didn’t know. And I think that’s a good example where no one ever tells you that fact.
What Organizations Truly Care About
Eric Ries: Right.
Lenny: [inaudible 01:09:16].
Redefining the Meaning of Profitability
Eric Ries: It’s like well, that’s an important part of the story to know. Exactly, exactly.
Lenny: I think it was a quick pivot. I think they got to YC and they’re like “No, let’s do something else.”
Redefining Profitability Part Two
Eric Ries: This is why it’s so hard to evaluate from the outside because oftentimes there is a thread of connectivity that from the inside makes sense why we went from this thing to that thing. And it’s often something unusual. It’ll be something like well, while we were building the game we had to do this thing first because our values were like, you can’t build a game without this thing. And then this thing turned out to be more valuable than the game. And then you’re like “Oh, why are we building this game? This is the thing we actually really care about.” So people are like “Well, I thought your vision was to build a game.” I was like “Well, maybe but why did they build that thing first? Other people build games without building that thing.” So that’s why I say it’s a process of self-discovery. Oftentimes the vision is discovered more than chosen in these stories. And from the outside it can seem really, really weird. And part of it is also that some people really need the failure of the first thing to appreciate what is great about the second thing.
Lenny: I think Segment is a good example where they tried a bunch of stuff and then they posted this random thing on Hacker News and it just blew up.
Building Organizations Like Products
Eric Ries: Sometimes it blows up so much that you’re like wow, that’s amazing. But oftentimes the new thing will get 100 signups. And if you’d started with that as your first thing you would’ve be like “100 signups who cares.” But if you’ve done 20 things in a row with zero signups, man, you’re like “God, somebody likes this thing.” Sometimes it’s your first try but sometimes not.
And again, not to go back to Facebook all the time but that’s such a famous story that people know it. In the movie, there’s that famous scene where they live in the house in Palo Alto and they run into Sean Parker and does the … Apparently that’s real, and they really were living in that house. They really were working on Face … But the thing that gets missed from the story is that Facebook’s already started at that point, it’s already being successful, and they’re on the break between semesters on the summer. And Dustin Moskovitz, and Mark Zuckerberg, and Adam D’Angelo, and I forget who else are living in this house together.
Someone told me this story about what they were actually working on at that time. It wasn’t Facebook. Zuck and D’Angelo had already moved on from Facebook and decided it wasn’t successful enough and they were like … One of them was working on a system where you could use gestural controls on your webcam to program computers because they wouldn’t get carpal tunnel. They were sitting on Facebook and it was … I should ask Dustin if this is true. I believe Dustin Moskovitz taught himself to program so that he could keep up with the demand of people who wanted to open Facebook at new schools because the virality was so insane. And I was like wow, you can be even building Facebook and not know. People are like “Well, if I told you that story without telling you the name of the founder you’d be like that’s no visionary.” I’m like “No, it’s Mark Zuckerberg.” No one has ever accused him of having a lack of vision, but somehow this is true.
So I feel like part of this is about just admitting the complexity of these stories is a little bit more … There’s more to the story than you really realize. These are imperfect people. And the pivot is really about learning, right? The original idea of the pivot was one foot anchored in the thing, one other foot moving from basketball, right? Anchored in what you’ve learned and yet so connected. So if it was truly a complete shutdown, start over again, nobody and nothing in common I would be like, that is not a pivot. But that’s just actually quite rare.
Usually, there’s some critical insight or element of the first thing that is carried over into the second thing. And if you were there, if you know the story, you’ll be like oh, wow, that makes sense in retrospect. And again, I think this cuts against our whole idea. The thing we think it’s like to be a visionary is you just … You know all the answers. You’re like a biblical prophet who sees the future. But that’s not what it’s like. It’s really much more about having the courage to see the world as you think it ought to be and then dance with that wherever it takes you. I think it’s a much more element of courage than of intelligence or anything else.
Lenny: You’re so full of amazing stories, Eric Ries. I never knew this Facebook story. Thinking about a founder who is maybe going through a period of considering a pivot … There’s a lot of startups right now just not struggling. Not struggling but struggling.
Competing With Love in Business
Eric Ries: The 2021 Vintage is having a rough time.
Lenny: Exactly. I’m curious just what advice you have for founders that are trying to decide, is this time to pivot? Should we give up? Should we stay the course? Just any tactical pro tips.
How to Take the First Step
Eric Ries: So first of all, if you’re asking whether you should pivot or not you probably know the answer already. That’s the first thing to just straight up say. People call me sometimes they’re like “I want to know if I have product market fit.” Well, the fact that you have time to call me and ask me this question already tells me the answer. If you have had product-market fit you would not have time for this. When it’s working there’s no time for naval gazing, there’s no time for … If you’ve never lived through a company going through product market fit it’s such a rush, it’s such an insane thing. The whirlwind of it is so amazing and so horrible at the same time. There’s just no time for anything else. If you’re not sure you’re sure. So that’s the first thing is just can you admit it to yourself? Can you do it?
If you admit that it’s not working then what do we do? What I really ask people to do in these situations … It’s like give yourself a fixed period of time to take some decisive action and see if it feels better. If you can’t agree that it’s not working then okay, can we take six weeks to try to make it work? We have this assumption, we’re going to try to test it, and we’re going to … Instead of working on the 29 other things that we have to do in the future, can we just spend all of our time 100% on the one and only one thing that matters right now? And just go heads down on that for a fixed period of time and see can we move … Is it possible to move the needle in that thing?
Because one of the things that is true about startups when it’s not working is that they’re stuck in this law of diminishing returns where you just … You can test yourself. Everything you try doesn’t work, nothing works. It’s like Thomas Kuhn and his theory of scientific paradigms. One of my favorite things that he talks about is that when an old paradigm is replaced with a new paradigm it’s usually only because the old people die out and the new people replace them. A lot of researchers are like why does that happen? It’s not that the new people are magically smarter than the old people, it’s just the old paradigm gets to the point where there’s no more experiments to run that are fruitful. And young, ambitious people want to run productive experiments. And the new paradigm gives them a thing where every experiment reveals something really amazing and they just naturally gravitate. It’s not actually a formal process of abandoning the old and going to the new.
And that’s true here too. Most people are not consciously able to let it go, they can’t do it. So okay, find yourself a paradigm where the experiments are productive. I can’t tell you how many startups I’ve been where they’re like “It’s up to the right.” I’m like “Oh, really?” It’s like “Yeah. At first it was 10%, now it’s 10.5%, then it was 10.6%, then it was 10.65%, then it was 10.657%.” You’re like “Guys, yes, it’s up and to the right but it’s not what you need it to be for this to work.” So give yourself a fixed amount of time on the current thing if you have to. Once you say, “Okay, the thing’s not working.” Do we have enough time for something new? Could you give yourself six weeks to try the new thing? If you’re a bigger company you’d be like okay, we’re in a … Maybe you can’t have a whole company work on it but can we have a team? Somebody, at least one whole human being, not a part-time committee. Somebody is full-time heads down trying to make the new thing work on a deadline.
And if you’re a small team and there’s only three of you, can you just put the whole thing on pause for a minute and give yourself a fix? If it was one weekend, that would be better than nothing. What can you actually afford? It’d be nice if it was six weeks but if it’s only one weekend, no problem. What is the new thing you want to do? Now that you’ve agreed among yourselves that you’re going to do that for a fixed period of time, everyone go around the room and say, “If I could start this company over again the thing I wish we were doing was.”
And no one wants to go first, but since everyone’s going to have to do it you’re like “Okay, I wish we were doing this other thing.” “Why?” And sometimes the first person goes and everyone’s like “I’ve been wanting to say that same thing for months.” And it’s like “Wait, we all actually secretly want to do this other thing but we all are afraid to let the others down.” It’s like “That’s actually great new” … “Well, good.” Don’t mourn that 18 months of your life you just wasted, be excited about the fact that you actually are on the same page about the new thing.
Anyway. Everyone says the new thing. See if one of them stands out to you, do that one. If you all agree do that one. If you have three things that you’re equally excited about do them in order. Everyone gets one month. We’ve got three months of money left, well, the first month we’ll do the first thing, second one we’ll do … And we’ll just see if anything sticks. And then sometimes people do that and they’re like “I’m just exhausted, I just want to die.” If everyone goes around and says, “I’ll just give the money back it’s okay, don’t die. Let the company die it’s all right.” When you raise too much money, and you’ve got your cap tables all messed up, and you … I feel like for our 2021 Vintage brothers and sisters, sometimes you just need to try again with the right capital structure, it’s not possible. Make the structural changes you actually have to make to give yourself a chance to do something you care about. If you don’t have that what are we doing here? Life’s too precious for that.
Lenny: That is awesome advice. Very tactical, very real. On the point you just asked of what would we be doing? What would you want to be working on? I imagine many people today would be answering AI-related products. And this is a good segue to where my next question is just, how are you seeing AI change the lean approach, product development, the way teams are run? Trends and changes you’re seeing in all of the stuff that we’ve been talking about?
Always Too Early Until Too Late
Eric Ries: I mean, we’re entering into a really interesting phase of humanity, and technology, and civilization. To know the right answers to those questions requires empirical knowledge that nobody has. That’s exciting about it. And also, of course, some people find it terrifying. What I’m seeing so far is that, first of all, people are afraid of AI. But not really afraid of AI they’re afraid of corporations with AI. AI doesn’t do anything by itself, it has to be embodied and deployed by humans who have bad alignment. And so everyone’s talking about AI alignment. I’d be a little more sanguine about AI alignment if the company’s doing the aligning were better at aligning their human intelligences. I’m blanking on whose law it is. Always these named laws. So the idea that your software comes to reflect the organization that produces it is somebody’s law. Conway’s law maybe.
Lenny: Let’s call it Ries’s Law.
Always Too Early Until Too Late (Continued)
Eric Ries: No, please don’t. I forget whose law. Anyway. That is going to be true on steroids in AI where we’re training in machine intelligence and we’re transmitting values to it. Those are going to be our organizational values. And tech has severely underinvested and neglected in creating governance and alignment structures to have those values be explicitly embodied in the organization which is why so many of our companies have become disappointments, and so many people feel betrayed by the companies they themselves help build. It’s actually sad. So AI is going to force us to take those issues seriously no matter what.
You asked about how it affects lean startup. At the end of the day, AI is a management technology. The thing it does is manage intelligence and other intelligences. Watch people struggle to make AI agents work. It’s been really interesting to me how often I’ve played with a lot of those tools myself. It’s amazing how often you program the agent and it will be like okay. You give it a task like go take over the world or whatever, go make a really successful ice cream stand. Or you give it any task you want and it’ll be like whatever. These AIs are programmed to never say no so whatever you say it’s like no problem and it’ll be okay.
First step, write a business plan for taking over the world, and that will spawn a new agent with a new AI. Be like okay, your job is to go write this business plan. And sometimes it’ll go try to write the business plan, but oftentimes it’ll spawn … The subtask will be like oh, first thing I need to do is spawn an AI agent that knows how to take over the world and have it write the business plan. Everyone will just pass the buck to somebody else. It’ll go into an infinite regress. It’s like, first thing we need to do is research this. And then the thing that comes back, and now we’ve learned to research that we need to research 10 more things. Each of those 10 things require researching 10 more. So many ways this can go wrong.
And anyone who’s been a middle manager who reads these transcripts will be like “Tell me about it, buddy.” I once was doing an AI agent and I wanted it to write me a 30-day course. A transcript or a 30-day course for teaching people lean startup. Someone had asked me to do that. And I was like great, I’ll have the AI do it for me. And it was amazing. It was like okay, no problem. Day one, what is lean startup? How does it work? Day two topic, what is an MVP, how’s it work? Dot, dot, dot continue like this for 30 more days and you’ll have a 30-day course. I was like yes, because I know that but could you please fill out the other 30 days? It was like no problem and it spawned a new subagent. It was like, given these two as an example please fill out the other 30 days. And it was like got it. And it was like how to do a task in 30 days.
Well, first look at these two tasks, then create similar things for 28 more days, then return the results. It’s like no, no, I’m asking you to do that. If you ever watched the Monty Python no, can’t guard him If he’s a guard skit, it’s like that. We’re talking past each other. It’s really interesting to see how AI, as we learn to understand it we will be able to develop different … It will really change management a lot because it changes the individual span of control quite a lot.
One of the most hard problems in management is that in order to touch a lot of things we need a lot of hierarchy because just the expansion factor of each one has this many reports, has this many reports, this many reports. That’s what’s required in order for us to summarize a large amount of information. If you have 100,000 employees they’re having 100,000 … Doing 100,000 things, the CEO needs to know what’s going on, that’s really difficult. So there’s this massive amount of human labor required just to summarize and understand what’s happening. Well, AI’s really good at summarization, it’s actually one of its main strengths. In the not to distant in future it’ll be very easy for someone to just be like “Summarize for me everything my company is doing.” You won’t need tons of middle managers to do that for you it’ll just know. We don’t even know what we would like to build organizations where those powers are available to everybody.
The last thing is, people are very focused on whether this is going to lead us to utopia or dystopia. Most people have a very firm conviction about what’s going to happen, and therefore what needs to happen now to prevent something bad from happening. So you have the people who are like “We have to stop, pause, blow up the data centers because it’s going to be this bad thing.” People who are like “We have to have government intervention to prevent research on this thing, we have to ban open source. For every one of the things somebody wants to do somebody else is saying, “That is actually the thing that leads to the dystopia,” right? Banning open source is what leads us into a corporately controlled autocracy that’s even worse than the thing that you were afraid of.
A very common question I’ve been getting from people in the industry who are having an existential crisis about what should they do with their life in the face of this AI. And as you say, all the people whose startups are failing are like I want to be working on AI. Okay. For everyone who’s got a grass is greener attitude about that … The people that are actually doing it, and quitting their jobs, and going to work on AI, the first thing they confront is What should I do that will have a positive effect? It’s actually really difficult to find out some … Everything I propose doing someone’s telling me that that’s actually the evil thing to do and vice versa. A big part of lean startup has been … For me, it’s been just about learning how to deal with uncertainty in my life, right? So how do you deal from a business perspective? How do you deal with situations of high uncertainty? And it could be scientific, it could be experimental, it could be … Just all these different values that are about coping with uncertainty.
And so the question people are asking me is basically, how do you take ethical action in the face of high uncertainty? And so the framework that I’ve been using I found very helpful and I hope other people will get something out of it is … The aperture of possible futures that AI makes possible is so incredibly wide. And Predicting it requires a really deep understanding of a bunch of empirical questions, as far as I can tell, nobody knows the answers to right now. So given that it’s not knowable what’s going to happen … You’re not an expert enough. As far as I can tell nobody knows. The people who are most expert are actively doing the research to try to find out the thing that will help us predict. There’s nothing you can do about that, you’ve got to wait for someone to go find that thing out. The thing that makes sense is to pick actions today that make ethical sense in a wide variety of future scenarios and do those. And what’s really interesting is that takes people out of total paralysis about it into there’s actually some very clear things that we should do. Whatever your doomsday scenario is we need more transparency right now, right? That’s critically important. We need more state capacity, we need more competent leaders, we need companies who are committed to certain values. It changes the way that we think about these technologies in relation to ourselves as individuals.
And that’s really my goal, in my work here, is just like how do we get people off the sidelines and into the fight doing things that are actually likely to be worthwhile? As I’m talking to you I’m like oh, it sounds a lot like the idea that most of the work you do in a startup is going to be thrown away after you pivot so don’t do that work. It’s not really a special unique insight it’s just taking the same approach to coping with the uncertainty of entrepreneurship and now applying it much more broadly because everything’s about to change.
The Companies Actually Doing It
Lenny: It sounds like this connects a lot with the work you’re doing now around governance and long-term stock exchange. And I want to get into that. But before I do, I’m curious if personally you’re finding AI to be useful in some part of your life using a tool? Or, if there’s a more interesting example of a company using a lean startup approach with AI integrated, is there anything that comes to mind of just something that you’ve seen to be practically valuable with some of the AI tooling?
Toyota and Unconventional Corporate Governance
Eric Ries: When GPT-4 was first launched, I saw a friend of a friend just was posting a friend of theirs. A story about someone who was using it for cold sales outbound right when it first came out. They had downloaded a list of 10,000 Shopify stores with metadata, which I googled you can buy that for 100 bucks. They just told GPT to the roll of Python script for each store. GPT-4, please pitch this store owner something that my outsourcing company could build for them and be really creative about it. He didn’t even read the emails he just sent them one after the other. And when the person would reply he would look in the email to see what he had pitched them and then talk to get the person on the phone.
The things that it pitched were so wildly creative and cool, it was really interesting. They were in the tone of the store, they were really … They used clever puns. So he was showing me screenshots of the person writing back being like “You had me at blah, blah.” It was very good sales prospecting. So okay. That’s an incredible story about experimentation at a super high rate. We could never have done anything like that even five minutes ago and now it’s going to become completely routine.
But it’s also a story about how all of our communication channels are about to be supersaturated with AI-generated experimentation whether we want it to or not. If this technology holds the way we think it will, it’s pretty clear that every company is going to be sending just huge volumes of AI- generated email such that people are going to need AI-generated email clients to read the email for them and summarize it for them. People make jokes about, I’m going to write three bullet points, my AI will turn that into an email, send it to you, your email will redact it back to the three bullet points. Couldn’t we have just exchanged three bullet points in the first place? We’re wasting a lot of silicon. A lot of GPUs were burned to move these three bullet points from point A to point B.
I don’t think that’s as silly a story as it sounds. I actually am excited about the possibility of AI-powered marketplaces to be far more fair than the way we do it today. Because if you think about that story, the problem fundamentally is that you have a massive number of people who could create things for store owners. And if all of those people emailed every store owner on the planet every day what they could do, it would be an overwhelming amount of information. So neither side is really capable of doing that. But AI is capable of doing that.
So imagine a world in which everybody is required and it’s also … But it’s also really easy to do, to define a procurement policy. What am I interested in buying, right? What problems do I want to solve? What am I interested in buying? What information do I want to have? And my agent reads my inbox for me with this filter and applies the filter completely fairly. It doesn’t matter if I played golf with so-and-so last week or whatever, just if the proposal comes in and it’s credible, and the AI is really diligent about making sure it’s not spam, then it will bring the thing to my attention. But it won’t show me the email, I won’t be tricked by its clever Al Ries-style positioning nonsense.
It will give me the essential information that “Hey, did you know that there’s this thing that could actually save you a bunch of money?” You’re like “Oh, okay, great.” That is so much better than advertising could ever be. And it’s also way more fair. We could then say, “Okay, actually, we would like every conceivable vendor to send every … As many emails as they want at any time, just blow me away with all the things your thing could do. Let your AI go crazy trying to convince my AI that your thing is actually good. That’s not wasted time, that actually create a lot more fairness in the world, and create a really even playing field for procurement for information. Think about journalists and pitches. You just see so many situations where we’ve established hierarchical gatekeeping structures because that was the only way to manage the volume of information that had to be processed for that thing to happen. When those limits are removed it could be terrible but it could be really awesome.
Lenny: I love your optimistic vent on all the things we’re talking about, I really respect that. And that might be a good segue to the final thing I wanted to chat about which is just the stuff you’re working on now. So you talked about you’re working a lot on helping companies with their governance. Also, I think you work on company culture, and helping companies make money, and profit, and build real businesses. And then that, obviously, connects to long-term stock exchange. Can you just talk about what it is you’re spending your time on there and where you see big opportunities?
Competitive Advantage of Foundation Control
Eric Ries: Yeah, sure. I mean, I think it’s one of the big, big, big issues that we’re grappling with. If I compare the questions I get asked by founders today versus 10 years ago, these issues are so much more in people’s minds than they were back then. When I work with a founder for the first time, we’ll often do a thought exercise that’s like “Okay, we’ve talked about culture, we’ve talked about the company, it’s like okay, great, why should anybody trust you?” They’re like, “What do you mean?” We’re talking about companies now building stuff that is so invasive into people’s lives, so much more powerful than it was before. And really honestly, this is just about taking seriously the rhetoric about changing the world.
And so many tech companies, when they’re raising money are going to change the world in fundamental ways. And when they’re being asked to take moral responsibility they’re like “I’m just a little database vendor.” It’s like you can’t have it both ways, buddy. Are you the Superman who’s reordering the whole world to your whim? Well, then you have moral responsibility for the outcomes. That’s a hotly debated point right now but I feel like it’s actually super … People have already made it clear in their rhetoric where they stand there, they’re just trying to evade that responsibility.
So the next gen of founders aren’t … Don’t find this complicated at all. People feel a real sense of responsibility to make the world a better place for the real things. Okay, great. I’m working on so many companies that are straight out of science fiction. I mean, the company that does travel from San Francisco to Tokyo in one hour. And you’re like, it’s unbelievable. Actual weather changing. The real thing where you’re like, we’re going to change the amount of habitable land on the planet and solve famine and drought forever. Obviously AI, there’s these just incredible possibilities and … In nanomaterials and quantum computing. I could go on. There’s just so many cool things going on.
But they all have really dire consequences. I mean, I was talking to a really advanced biological AI company where … Imagine having an LLM. Imagine ChatGPT but for biological compounds. We’d be like “AI, could you just synthesize for me a molecule that has these properties just described in natural language?” And it’s like “Sure.” That could be incredible. Think about the cures and the diseases. That’s going to be an amazing transformation on technology. But it has such obvious risk and downside that we have to take it really seriously. Why should anyone trust you with this power? And people always say the same thing, “Well, I’m a really good guy. I’m your bro, man. I’ve got good intentions.” It’s like “Well, who gives a crap about that, you could die, you could be replaced. Some investor could sweep you.”
How many stories can I tell you about startups where the investors change management involuntarily? The fundamental problem is that no individual person can ever make promises on behalf of an organization, it’s not possible, it’s a category error. You’re replaceable. And the reason is because organizations are actually alive. They are superorganism with their own soul and their own life. We are their selves and organs when we make them. Founders don’t create them we birth. And that’s different. We don’t own them, they’re not slaves. We nurture them. That’s different. If you want your organization to be trustworthy, the promises it makes … If you want to win the public’s trust by making promises they can believe, you have to embody those promises in the structure of the organization itself so that even if you weren’t there the promise would be kept.
And people act like that’s impossible but there are lots of examples. I mean, think about the code of medical ethics and the Hippocratic Oath. In the US if you want to do healthcare you have to have a care delivery organization that is staffed and run by doctors, not by private equity guys. There’s all these things. Where we know how to build structures that are organized around some purpose, so what should that purpose be?
I was talking to a former Googler … I hate to pick on Google because they’ve always done a lot of good things in the world. But they’re, I think, the locus of a lot of tech people’s disappointment in the world. And so this is nothing unique to Google but I just happen to be talking to an ex-Googler. And he was talking about the values, and the company, and his disappointment, and what had happened in the many years that he’d been there, the good and the bad. I said, “Okay, answer me this question. I’m going to ask you two different things, and for each thing I want you to tell me, what is your assessment of the probability that this thing will happen in the future? Thing number one. What do you believe is the probability that Google will file its next quarterly report on time?” He was like “100%.” I was like “How 100%?” I’m like “As certain as the sun will rise, bet your life on it?” He’s like, “As certain as the sun will rise 100%.” I was like “Excellent, I agree.”
A second question. “What is the odds that Google would be complicit in something that would make them money but at the expense of murdering somebody?” He’s like, “They probably wouldn’t do that.” I was like “Probably, 100%. As certain as the sun will rise.” He’s like “Well, no.” I was like “Oh really? What do you have?” I said, “Well, come on, what if they’re running a self-driving car as they kill somebody and they cover it up? What if they will a social media product and it’s conflicted in a genocide? Facebook went through that.” He’s like “I don’t think they would do that on purpose. The difference between the thing that is certain and the thing that we don’t know.” I was like “Okay, why is that?” We treat that normal. Of course, companies file that report sometime. Is it because in the human heart fidelity to finance rules is hard coded, but care for other humans in their life is not? People don’t give a crap about quarterly reports.
We have built a massive apparatus in these organizations to make sure that certain things happen. Companies are really good at things that they care about, and therefore we can learn something interesting about the things that they care about by what they choose to be good at and what they perennially are not good at. My view is that what it means to be a for-profit company, we’ve gotten it wrong. We’re basically incorrect. Making a profit is actually about maximizing human flourishing. That’s what it means to be for profit. So our current category says that Philip Morris is a for-profit company, and the Smithsonian is a not-for-profit but that’s just backwards. Smithsonian does a lot more good in the world than the Merchants of Death at Philip Morris. I think we’ve just gotten confused. All forms of making money are the same but we all know that that’s not true. There’s exploitative extractive ways of making money, and then there’s truly additive ways of making money that actually create net new value.
And so if companies are dedicated to human flourishing and they want to be the company that will create net new value, and they don’t want to become a bureaucratic nightmare, they don’t want to become a rent-seeking blob of discriminate math. We don’t want that to happen. We have to encode those promises into the company in a deep way. And I think that’s really what governance is about. The whole ESG movement, all the discussion about this has gone totally off the rails. Philip Morris has actually great ESG scores it’s absurd. How can that possibly be right? We have to find a better way forward.
Again, going back to lean startup and first principles. Once you go back to that first principle and say, “What would it look like if every person in a company felt a fiduciary duty to promote human flourishing? How would you do it? And what would be different?” I’ve started to meet founders who are building companies like that and it’s almost unrecognizably different. I’ve been framing this, of course, in terms of the big things. Obviously, if you think about the weather control startup. That needs to be right, not just because if a super villain and got their hands on it would be really horrible but also because you can easily imagine … They got bought by some private equity firm who’s like “This is awesome. We can now put every town in a zero-sum competition with their neighbor for the good weather and extort them both from” … That would be just as horrible. We don’t need superviliany for this. None of us should want our technology to be used in that way so we need to prevent that from happening.
So when we put that in that ethos, we make that part of every decision, we finally can create the environment that people really want to work in just to make great products. I feel like most people I meet just want to work on something awesome. And we act like giving, that just happens. But I’ve worked with so many companies now, even the ones that are really truly great … You go meet a team who says, “We had the island of freedom, we had the full” … We were able to do it exactly right. That’s one of the most satisfying experiences in my career.
You ask people, “Why was it” … “How was it made?” And they’re like “How was the product made?” “No, how was the island of freedom that you lived on made?” And they’re like “I don’t know, ask my boss.” I’ve done this exercise. I’ve asked the boss, and the boss’s boss, and the boss’s boss. And sometimes I’ll go all the way up, and it’ll be like “Well, actually the founder of the company, who’s still around, made it so.” But sometimes no one knows. And you’re like “Who’s in charge here? The board is in charge.? They’re like “I don’t know. The CEO asked me to serve on the board.” The company chose its own board, the board chooses the C … It’s all self-referential. And it’s like where did it come from? We view it as an accident, it just happens sometimes. But that’s because we’re not engineering these beings to create the outcomes that we really want. And I just think this is a blind spot.
As company builders, we’re so focused on the product and not nearly focused enough on the organization. We tend to say at the surface level stuff like culture, we don’t get into the deeper stuff. So, yeah, I think as builders, as an industry, we need a new movement to say. “We are going to build companies where human flourishing is the thing at the beating heart. We’re going to wield that as a competitive advantage to not only change the world for the better but also make a lot more money.”
That’s the crazy thing about it is these companies are worth a lot more than their competitors. Without naming the company, I talk about a company that is an incredible ethos of love and just … It actually loves their customers, it’s amazing. Everything cascades down from that central premise, including the instruct … Every CEO I meet is so frustrated all the time. I can’t get people on the same page. I tell them what to do but they don’t listen. It’s like maybe part of the reason you’re having trouble with this is the strategy you’re communicating does not actually make sense to people. Well, this one really makes sense. Treat your customer like you would treat your own parents. People are like “Oh, that’s actually really easy and clear, I understand exactly what that means.” And so their customer loyalty is unbelievable. Can you imagine what their retention is like if you-
The Fallacy of Shareholder Primacy
Eric Ries: And so their customer loyalty is unbelievable. Can you imagine what their retention is like, if you actually felt… You call customer service and it’s like you’re reaching your guardian angel rather than some bored customer service rep? It’s amazing. So their competitors are like, “This is not fair. It’s unfair to compete with this. This thing is a monster.”
But when you’re going to a customer like, “Hi, would you rather work with a company that’s owned by some private equity schmuck, or a company where some of the prosperity you helped me generate gets reinvested in you and your community?” It’s a competitive advantage to do the right thing. So I feel like that we’re going to have a new wave of founders who take that seriously, who are going to just… They’re going to steamroll. It’s going to be really awesome to watch.
Lenny: I can see how you started a movement, maybe accidentally. I am mesmerized, I am sold. If people listening are also sold and want to build a company this way, what can they do? Is it work with the long-term stock exchange? Is there something they could do to help move their company and product team and business in this direction?
The Spiritual Holding Company
Eric Ries: I feel like there will come a time when there’s a lot of resources about how to do this, and there aren’t a lot right now, so it’s a bit thin. I’m working on it, but we’re still trying to figure out the right way. But I am hoping to start doing some events and some things to talk about this stuff, so people should feel free to reach out or follow me on the usual places.
If people want to do it themselves, talk to your lawyer. This is definitely one of these “talk to your lawyer” moments. I’ve worked with enough companies, I’ve actually been around long enough that I’ve worked with multiple companies that have gone from zero to IPO and beyond. So I have personal friends who’ve taken their company public who are now extremely wealthy, and who sometimes even call me and are like, “Hey, I run this massive multinational company. Can we talk about this very specific situation that’s come up at work?” I still am involved in those companies. I still get to hear sometimes what it’s like to actually do it.
And one of my main takeaways of that entire process is that it’s always too early until it’s too late. So what happens is people hear something like what I’m saying, like, “Oh yeah, we need to make sure… Here’s some things you can do.” Obviously you can become a public benefit corp or a social, what’s it called, in California, a social purpose corporation. There’s legal designations you could take. That’s fine. You could talk to a company that, because their startup involves building a technology out of something that comes out of indigenous cultures, they took 10% of their company and they put it in a foundation, the trustees of whom are tribal leaders from the cultures that they interact with, and those people that has an economic role and a role in governance. You can establish a foundation like that.
You can do what’s called the board mission pledge where the people on your board are required as a condition of being on your board to pledge that they’ll use their business judgment under Delaware law to support the mission, not just the narrow interest of shareholders. You can build, we call it an LTSPV, which is basically a financing instrument that operates like a regular SPV, but every LP in the SPV has to sign a contract to say that they will support the mission of the company if you ever want to do something like a B Corp version, or if there’s ever a situation where it’s unclear whether they’re in favor of you doing the thing that’s right by your stakeholders long-term, they’re pre-committed to that, and we assign trustees to oversee those shares rather than just the narrow interest of the shareholders.
I can go on and on. There’s all this stuff you can do. So I get companies, I got founders excited. I’ve been getting founders excited about this for a long time. I’ve now been long enough at this that the founder will go to their lawyer and the lawyer will say, “Listen, sounds great, but you don’t need to do that right now. You can always do it later.” And they’re like, “Okay.” And then the companies going bigger and they’ll do some financing round. Each financing round, it’s like, “Is this the time?” “Ah, that’s fine, we could just do it later.” And then it’s like 18 months from the IPO, they’ve hired a CFO and there’s a bankers and the whole thing’s on rails, and the CEO’s like now, “Hey, did we ever establish that foundation thing for our non-employee stakeholders?” And they’re like, “Huh?”
“The thing, remember we talked about, you said it was too early?” “Oh yeah, I remember that. Have we done it?” “Oh no, it’s too late.” “It’s too late?” “The thing’s on rail, shareholders wouldn’t like it.” The banker’s like, “Oh, you don’t want to put that in your IPO prospectus.” It’s like, “Wait a minute, when was the right time? You’re telling me that 18 months ago, July 3rd through 7th, there was a four-day window and it was the right time? It’s never the right time. Why is it never the right time?” And I’ve seen founder after founder after founder get steamrolled. One of the only exceptions I know is a mutual friend of ours who actually stopped the IPO process and was like, “No, I won’t file the paperwork.” They tried to steamroll him and file the paperwork without him, and he said “No, until I see this thing in there, we’re not doing it.”
And it was a very contentious thing, took a tremendous amount of political capital. I admire him immensely for doing it, but most founders don’t. Most founders just go along, and the thing never gets done. So the most important thing is to sit down with your lawyer and just take them through a few hypothetical situations and ask them, are you properly protected against this specific thing? For example, if tomorrow Philip Morris shows up… Pick your favorite evil. I’ve used Philip Morris because my dad’s a pulmonologist, so I grew up as Philip Morris as the ultimate of evil corporations. But pick your favorite evil corp. If they show up tomorrow and they offer you $1 per share more than your company is currently worth to buy it from you and your investors in order to use it to sell cigarettes to children, do you have a fiduciary duty to say yes or a fiduciary duty to say no?
Believe it or not, if you went to a fancy law firm and they filed for you the standard articles of incorporation that are used in Silicon Valley, most governance experts upon reading those documents would say, “You have a fiduciary duty to say yes.” If that doesn’t horrify you and make you want to jump out a window, I can’t help you at all. And if you don’t care, fine, I got nothing to say to you. If you don’t care, you don’t care. There are really sociopathic people in this world, and they build sociopathic companies, and I can’t help you.
But if you do care, sit down with your lawyer and just be like, “Hey, is this true? Could this happen to me?” And they’ll be like, “Yeah, obviously.” And you’d be like, “If I’d like to prevent that, what could I do?” And they’ll be like, “Well, your investors might not like it. Investors like it this way. It maximizes their returns.” And you’re like, “Yeah, but that’s not what I asked you.” And your lawyer is good enough to know the answer to this question. What can you do? And if they can’t go back and rewind the video that, Lenny, I just had named a bunch of things. You can do founders preferred shares, you can do all this stuff.
Just ask them about some of that stuff. And don’t ask them, “Should I do it?” Ask them, “Tell me for each thing, can you give me the pros and cons of doing it?” And then they’ll be like, “The pros are it might actually allow you to save your company’s soul in a dire situation like this. The cons are investors might not like it too much.” So sit there and be like, “Hmm, I’d like to do it. That sounds good. Thank you for advising me on the pros and cons. You work for me, you’re my lawyer. I’d like to implement this thing.”
And honestly, just pick a few of them to get started with. And over time, what you need to do, last piece of advice, every founder wants the one weird trick, this is how Lean Startup started too. People were like, “Can you just tell me the one weird trick to keep my company innovative forever?” It’s like, there’s no one weird trick. You want to do the right thing, it is a relentless and constant process. We need to build a structure of the company so that at every level, at every moment, every decision the company makes, somebody in the room making that decision feels like they have a fiduciary duty to do the right thing.
If that’s the outcome you want, then this is going to be a perpetual process of constantly trying to find ways to create the alignment that is needed between the governance, the management structure, and the actual operations and culture of the company. Those are not separate things. They’re actually only one thing, because it’s just one being. You have a skeletal system, a muscular system, and a circulatory system, but those things are not separate. They work together to produce you, the being. And the same thing is true for these governance structures. So I think most founders are under-indexed, under-invested in that, and that’s why we get to these horrible outcomes.
Lenny: To give founders a little ammo in doing this, because as you said, investors are not going to be excited to sacrifice profits, employees may not be. Other companies that you can share that have done this and have implemented some of these things that are public? And if not, what confidence can you give founders to be like, “Many, many amazing companies have done this and it’s actually not that hard”?
The Courage of Spiritual Holding Companies
Eric Ries: I don’t like to name specific company names in venues like this, because then people always are like, “Well, but what about…” It’s just, people like to do the whataboutism thing too much. And also I like to let companies speak for themselves. I have way too much confidential information, and I just have to be really careful about it. But nothing I’m describing here is hypothetical or unreal. I have witnessed this many, many, many times myself, and every story I tell, it’s a real company that you can meet. They’re real people that are out there.
And they’re not always the loudest. The loud voices in our industry tend to be the ones that have something to sell, and these companies are just quietly doing the right thing. They don’t really need to tell you about it because they’re doing it. So there’s an inverse proportion to the loudness and controversy and meanness of the thing, and how likely it is to be accurate, unfortunately.
What I will say is this is kind of an open secret in the world. So here’s the thing that really surprised me. I thought the way that we build companies, like the standard corporate governance practices that we all do, must be well established in research. They must be well backed up. It must be the only and best way. But when you start studying these things, you find the exceptions really fast. I got into this because of Toyota, only because I called the thing Lean Startups, I wanted to borrow stuff from lean manufacturing, I studied Toyota.
Well, Toyota has a really weird corporate governance system, which is part of the old Japanese keiretsu system where the Toyota family is intertwined in its ownership with their supplier… By modern standards, it’s a terrible rat’s nest of disastrous corporate governance practices. But if you look at the literature, everybody who studies Toyota is like, “Well, the reason they’re able to do this awesome stuff is because they have a philosophy of long-term thinking.” I used to be like, “Well, how can you have a philosophy of long-term thinking if you’re subject to quarterly returns?” And they’d be like, “Oh, they don’t pay attention to that stuff because they have this other structure.”
I’m like, “Oh, that’s weird.” So for a while I was like, “Oh, I guess it’s just Toyota. We should build more Toyotas.” But then there’s all these family run companies out there where the family has produced this outcome. And then I started to meet foundation controlled companies. It’s funny, OpenAI is the hottest new thing. OpenAI is a nonprofit foundation with a wholly owned for-profit subsidiary. I thought that was like, “Wow, OpenAI.” That’s actually very common. There are a lot of companies like that in the world. In certain European countries, it’s actually 20 or 25% of the companies, public companies, are all structured that way. So it’s actually totally normal.
And I did a talk for one of those companies a few years ago, and it blew my mind, because when I would go there and talk to them, I was talking to their CEO and their leadership and I was like, “Oh, what’s it like to deal with the public markets and quarterly?” And they were like, “Well, we don’t do that.” I was like, “Oh, is that a major source of competitive disadvantage? You guys are a nonprofit.” And they’re like, “Are you kidding me? We crush our competitors, because they have to waste time on all that stuff. We don’t waste time on it.” I was like, “Oh.”
Anyone ever heard of Warren Buffett? Why do his companies have a competitive… People are always like, “Oh, if you’re giving up profit, your investors might not like it.” I’m like, “These companies are world destroyers.” What are we giving up? That’s why I go start with the wrong definition of profit. We’ll actually make a ton more money if we do it this way. And Amazon, for all of its flaws, the long-term thinking and the customer centricity, that’s a major source of competitive advantage.
And I’ve worked with Amazon, I’ve worked with many of their competitors, and the funny part with competitors sitting there, they’re like, “Oh man, Amazon’s entering our market, and we need to do X, Y, Z.” And I’m always like, “Do you think Amazon’s doing X, Y, Z?” And they’re like, “Oh no, they don’t have to do that stuff. We have to do all this bullshit that our investors require and the activists and whatever.” It’s like, “Well, if you’re competing against them and they don’t have to do that crap, maybe you should try doing it their way.” And they’d be like, “Oh, our investors wouldn’t like that.”
I was like, “Your investors want you to get your butt kicked by Amazon? Are you sure?” So I actually think… One of my realizations here is that the investor centric view of corporations, what’s called the shareholder primacy theory of corporations, it’s just wrong. It’s wrong even if all you care about is maximizing returns for shareholders. Shareholders have this propensity to be the person who killed the goose that laid the golden egg, and it’s been going on for 150 years. These case studies go back to the 19th century. This is an old, old, old idea.
And the solutions are actually all hidden in plain sight. So I was reading a research paper the other day that studied some of these, I think it was in Denmark, foundation controlled companies versus level matched public companies that are not foundation controlled. And they found the foundation controlled companies outperformed from a financial performance perspective.
And I was like, “How is this not the most incendiary result ever?” If this is true, so much of the indoctrination we’ve all received about what it means to be a for-profit company is just wrong, because those companies should be riddled with slowness and agency problems. I know the theory so well, I can tell you all the things that ought to be true. If those things are not true, what’s up? And then I was like, “Wait a minute. I’ve been to these foundation controlled companies.” I’ve been to Hershey, it’s just controlled by, the founder donated the shares to a foundation that runs a school.
I’ve seen it with my own eyes, and even having seen it with my own eyes, it took me a while to realize, “Wait a minute, this is an open secret.” So I was kind of horrified to find out that this is well-known, and yet was never offered to me. How come no one ever offered me the chance to build a company this way? And if you study even OpenAI or any of these companies that are now structured this way, it’s always very idiosyncratic. They were on the standard path and then something happened that allowed them to do something unusual. And I was like, “Well, I don’t understand why we have to wait for it to be idiosyncratic. Why don’t we do it on purpose?”
And so I started to conceive of this idea, I call it the spiritual holding company, basically the thing at the center that is responsible for the soul of the overall ecosystem of companies, that is the enterprise. And you can see that obviously in the OpenAI example, if anyone has studied the anthropic long-term benefit trust, it is not coincidentally structured that way, but you see lots of examples where that’s what’s done.
And I started just pitching it to founders and said, “Look, let me try to show you what the benefit of this would be.” And I’ve been amazed how many founders break down. I’ve had founders start crying with me where they’re like, “This is what I always really wanted to do, but I never felt like… No one would talk to me about it. I didn’t think it was possible. I want to put… There’s a critical stakeholder I want to put at the center of everything I’m doing, or there’s some reason why we’re doing this, and I want that to be the thing.” I was talking about a startup who’s trying to get… And again, it’s not like a Kumbaya thing, do-gooder whatever, virtue signaling bullshit. It’s a competitive issue.
There’s a company that is working to recruit academic researchers to leave their lab and come work for the startup full-time. It is one of the most difficult recruiting challenges there is, because it’s not talking about kids out of some school. It’s tenured professors who are leaving behind a very good life to go do this thing. They would only do it if they profoundly believed in your mission, but they’re not naive.
And the first question every one of those people ask is, “How do I know this technology’s not going to be used for evil? I have complete academic freedom. I’m going to give that up for you. Make me believe that this is the real thing.” And if you’re like, “Well, we believed in some manifesto and we’re good people and we got money from some investor who’s promised you he’s going to be a good guy,” that is not going to cut it.
And so they’re having to do the whole thing, sit down with their lawyers. “What do I have to do to make this promise real for these people to make them want to join?” And I’ve just seen that over and over again. If we have the courage to do it, it can solve what are really today considered insoluble business problems. So it’s cool, but I think it’s scary for people to feel like they’re doing something different. And so I think over time we’ll get more of these stories on the record. There’ll be more precedent for it. But I got to say, not to hearken back to Lean Startup too much, but I’ll never forget sitting with early people who were like, “What’s the proof? What big company is doing this?” And I’ll be like, “Just wait. If you don’t want the competitive advantage now, you can get it later.”
And many of the people… Look back on who was profiled in those books. Many of the early Lean Startup adopters are very successful people today because they didn’t feel like they had to wait. They were able to think for themselves and apply the principles in their own life. And I don’t ask anyone to take this on faith. This is not like, “You have to do what I say.” It’s more just like, think for yourself. If this makes sense to you and no one can explain to you why it’s a bad idea in ways that you find compelling, give yourself permission to at least try it and see what happens.
The Lightning Question Round
Lenny: What a beautiful example of first principles thinking, looping back to our very first question. I think this is going to be the longest episode we’ve done yet, which is amazing. Also, I think it might be my new favorite episode, so that’s a good conversation.
Toyota Chief Engineer System
Eric Ries: Thank you for saying so.
Motto for Life
Lenny: With that, we’ve reached our very exciting lightning round. We’ll see what sort of lightning we find. Are you ready?
Eric Ries: I’m ready.
Terminology That Was Not Widely Accepted
Lenny: What are two or three books you’ve recommended most to other people?
Eric Ries: There’s a book called “The Enlightened Capitalists” that is just a hundred years of people trying to fix this problem and failing, and the detailed case studies, and no founder should ever read it end-to-end. It’s too depressing. But it’s important to know some of these stories of founders that they’re willing to talk about their regrets on their deathbed that they wish they had done differently. And for those who think there’s one weird trick, this will help disabuse you of that notion, but it’s helpful to understand that mistakes have been made before so that we can do better going forward.
On a happier note, I really want more people to read Ken Liu’s “The Dandelion Dynasty”. It’s a silk punk fantasy series, but it’s written by… He’s a wonderful writer and it’s like, if you want to see what real optimism about engineering and the ability of engineers to change the world, but transformed into a totally different environment, it’s just such a great read. The audiobook is really good also, and I’ve been kind of plugging that lately.
And then if you want just something that’s pure fun, it’s not really the season for a summer read, but also a really good audiobook is the Murderbot Diaries by Martha Wells. She’s just such a great writer, and I know it’s a very frivolous thing on the surface, but there’s actually a lot in there, and I won’t ruin it for… Nothing I would say would be a spoiler, but there’s actually a surprising amount of depth for what is really just a fun action adventure kind of read. I can’t wait. I hope it is someday made into a movie. If they do it justice, it’ll be really terrific.
The Final Closing Remarks
Lenny: What’s a favorite recent movie or TV show?
The Collective Power of Entrepreneurs
Eric Ries: I really have been try… I don’t have time anymore. I have young kids, so my media consumption days are extremely, extremely limited. But I’ve been kind forcing myself to slog through a bunch of the Disney+ Star Wars series, which have been almost universally disappointing with the exception of Andor. For those that have not seen Andor, it really is very good. Not good for a Disney+ Star Wars show, it’s actually good for television, which I couldn’t believe when I saw the premise of it.
Lenny: Also a huge fan of Andor, and I’ve recommended it many times. And I feel like now every new Star Wars series, I’m just comparing it to that bar, and it’s not…
Average Company Lifespans Are Shrinking
Eric Ries: That’s the problem. It’s like, if they would just make that creative team or they would just let them make shows, it would be okay. It would redeem all the rest of their failures.
Lenny: Yeah, or just keep making sequels. I’ll watch them all. What is a favorite interview question you like to ask people when you’re interviewing them?
Eric Ries: My favorite interview question of all time is to ask people to describe a best practice that they learned in their career and to really get into it. Like, “Tell me the story of how you learned the best practice,” you just let them go on, and then you’d be like, “Great. Tell me a situation where that best practice would not be applicable.” Most people can’t do it.
Lenny: So interesting. What is a favorite product you’ve recently discovered that you really like? Whether it’s an app or something physical.
Eric Ries: I was just singing the praises of a product to a friend. It’s the most boring and stupid product ever. It’s just a humidifier. And you’d be like, “Okay, how can you mess up a humidifier?” But actually, it’s very difficult to get it exactly right. There’s all these little attention to detail craft. We were talking about craft, it was on my mind. It’s a very humble appliance, and it’s actually, I’ve gone through 10 different models trying to find something that doesn’t drive me…
Because you want to run it at night, but if it has a light and it’s on all night, that’s really annoying. If it makes a noise, that’s no good. There’s all these things where you’re like, “Boy, if anyone had used the jobs to be done framework for one minute on what this physical device is supposed to be, this is the product they would’ve built.” And I can’t believe… It’s so interesting in modern capitalism, there’s so many categories where there isn’t a product like that, or there’s just one. You’re like, “How come every product is not like that?” And the forces we’ve been talking about are the reason why it’s funny to see it represented in such a humble object.
Lenny: What is the actual humidifier that you found?
Eric Ries: It’s the Levoit 300S ultrasonic humidifier. Lavoit 300S. It’s just a really well-made $60 appliance. It’s nothing special.
Lenny: They’re about to see a spike in sales. Go Lavoit.
Eric Ries: Somewhere, somehow, there was a product management meeting where they talked about how to make this thing, and I can picture it right now. Somebody was like, “Look, the way we make these things sucks. Can you please let me just put the thing that’s normally on the bottom, let me put it on the top and let me put this in there and put that in there?” And someone was like, “Who cares? That’s going to add 3 cents to the cost and reduce our margins by whatever.” And someone was like, “No, really think we should do it right.” So we like to reward people like that.
Lenny: Yeah, and this is going to take us off-topic potentially, but I’ve been thinking more recently about how most often the best stuff comes from just one person’s singular control over how to do that thing over and over and over.
Eric Ries: Yeah. There’s actually, if you study Toyota production system, they have this job that’s called the chief engineer. I wish more founders knew about this, because when you become large, it’s important to be able to empower the founders that work for you and not just only see yourself as the founder. So the chief engineer is the person with the moral authority over an entire line, like a car from start to finish.
And I don’t know if this is still true, but back in the days when people were writing a lot about Toyota, inside the company, the car is not called a Toyota Camry. It is called “so-and-so san’s car”. And the thing about the chief engineer is they have no reports. Nobody works for them, they just have what’s called the big room, if you’ve studied that whole thing. There’s a place where they are, and they have a small staff, but they have the design and the car, and as the number of people who work on a car, it swells up and then it comes down. Because you have the design phase and then you go into production, and eventually the number of people goes back down.
The most important thing you have to do in a product is make trade-off decisions. And if you just let people make their own trade-off decisions, it’s totally different. It’s really difficult to have any kind of coherent design like thrust. So the idea is like, “This is the person who makes those.” If there’s a fight between two departments about how much the thing should cost or if someone asks a question, like, “Listen, I can improve the aerodynamics of the thing by this much, but it will cost this much, and it will mean we have to take this button and turn it from metal to plastic,” you have someone who can actually be like, “No, that button has to be metal. We made a promise that this is going to be.” Or “Button doesn’t matter.” You just need someone who knows that thing. It has to be that you can keep it in somebody’s head. So yeah, called the chief engineer.
Lenny: Just when I thought we were done with golden nuggets in this episode. That is so interesting. I think that makes so much sense, and I also agree that should happen more often at companies. Wow.
Eric Ries: Yeah, it’s very much missing structures.
Lenny: [inaudible 02:05:12] All right, two more questions. Do you have a favorite life motto that you like to repeat yourself, share with folks, something that you come back to often that helps you do the right sorts of things?
Eric Ries: Well, I won’t get into the whole explanation. The motto is that nothing real can be threatened, and nothing unreal exists.
Lenny: We’ll have a whole other episode just unpacking that. Amazing. Final question, you invented a lot of terms, a lot of concepts. Is there one that you don’t think caught fire and that kind of spread as much as you thought it should?
Eric Ries: Oh, yeah. Yeah. I got tons. I got tons, man. The cutting room floor is full of so many things. I went through a really extended period of trying to use… But it was before Lean Startup, I tried to use cell biology as the metaphor for a startup. It should have been… I wanted something more biological and less mechanical. I never got one person to understand what I was talking about. It did not work.
But I’ll tell you the one that surprised me the most. This is such a small one, but I really thought that this would take off and people would be way into it. I did a lot of writing about it and then I couldn’t get it to stick. So everyone knows viral loops. There’s also engagement loops, and I was like, “Oh, just like we have viral loops, which is about how products grow, we should also have engagement loops, which is about how people stick.”
And there’s a science to stickiness, and people have written about that from the point of view. “How do you make your product more addictive?” But that’s not really what this is about. This is about being able to model the dynamics of how customers flow through. And a version of that is in the Lean Startup. We have what I call the law of sustainable growth, which is a situation where new customers come as a natural side effect of the actions of past customers.
And so that is obviously true of the viral loop. Everyone knows the paid engine of acquisition where you have, whether your marginal profit per customer drives your ability to invest in customer acquisition. But there’s also what I call the sticky engine of growth, which is kind of like the vestige of this idea that made it into the Lean Startup. But I actually think that that’s a whole really interesting thing where, if you think about, the way to think about it is if someone is living their life, having a happy time, then for some reason they use your product, why did they use your product? What made them come back to your product?
And it’s actually, there’s not that many things it can be. It might be because you gave them a synthetic notification like, “Hey, please come back to my product,” and they did. It could be because some kind of organic notification took place, that is, like, they got an email from a friend on your platform that was like, they got a notification that was driven by… One of the most successful customer emails of all time is PayPal’s, “You’ve got money, click here to take the money.” It has a very high click-through rate, maybe second only to Facebook, “Someone has tagged you in a photo, but we’re not going to tell you which photo it was. So you better click, right?”
So something happened. If it’s not one of those two things, then it’s got to be because of positioning. Something in your brain on its own told you to use that product. Why did you go start playing World of Warcraft? Did you have to get a notification to remind you? You like playing it, no one had to tell you. You like you doing it, or something happened in your life that made you need, you’re hungry so you went to the thing to get the food.
Because those are the only ways by which people come back to a product, those all can be measured, studied, and understood to the same level of precision as we have with viral loops. I can’t for the life of me understand why viral loops gets all the attention and no one is remotely interested in the science of engagement. I was a time when I thought engagement loops would make me way more famous than Lean Startup ever would. I was very, very wrong.
Lenny: This also reminds me that you also, I think, were the first person to talk about growth as loops and these specific loops and a growth engine, which now a lot of people talk about.
Eric Ries: Yeah, [inaudible 02:08:57] growth is mine, yeah.
Lenny: Yeah. So wow, we discovered another one. I imagine there’s many more. Eric, this was incredible. I definitely think this is my new favorite episode. Thank you so much for being here.
Two final questions. Where can folks find you a line if they want to reach out potentially? And then how can listeners be useful to you?
Eric Ries: You still find me at theleanstartup.com. I’m @ericries, E-R-I-C-R-I-E-S, on all the usual places, although I really don’t do a lot of social anymore, so joining my mailing list at theleanstartup.com is probably the best way to keep track of the major things that I do in my life. And then, yeah, I do try to get on some social from sometimes to make some announcements. And if people want to reach out, I’m just eric@theleanstartup.com.
Lenny: How can listeners be useful to you?
Eric Ries: Oh, that’s really sweet. Honestly, people need to just take these ideas and do something with them, and then you learn something from it, and then tell me what happens. That’s all I want in exchange. Just tell me what happened. Even if you… The people I’ve learned the most from are the people that tried to use my ideas, thought they were the stupidest thing they ever heard of and doesn’t work and they’re mad. Tell me, I want to know all about it, because really my job, the way I see myself is, my job is just to try to understand these things. I just want to know the truth. If I’m wrong… I don’t like being wrong, I’m as attached to my ideas as anybody else, but I try really hard to find out what is true. So if you discover something new, please tell me. That’s all I ask, is let me know.
And if you don’t try, then how are you ever going to discover anything new? Every day I meet people who’ve just heard about Lean Startup for the first time. For me, it’s so old… Every day people read the book for the first time and they have a question about… So if that’s where you’re at and you just discovered it, excellent. I wrote a couple books. You can go read them, please [inaudible 02:10:38] for those ideas. If you’re someone who’s jaded and you’re like, “Oh, I’m so tired of hearing about Lean Startup,” I mentioned a few new ideas in this conversation. Go try one of those. And even if it doesn’t work, or if it does work, or even the people I’ve learned a lot from are people who, they tried it, and in the process of it not working, they had another even better idea, and then they went and did that.
Well, go do that. All I ask is you tell me what happened. I want to learn from people who are making this happen. And I hope as we come together as a, not just as a Lean Startup movement, but as a startup movement, as a group of people who fundamentally believe that we can make the world a better place through the building of organizations from scratch. We have that as our belief, that the future is not scary. Newness is not something to be feared or opposed, but there is a responsible, methodical, value-creating way to make positive change in the world. That’s an ethos that is not universally shared, and that if someone’s listening to your podcast, my assumption is you’re in our tribe. So we have a collective responsibility to think about, “What do we stand for beyond just my company, my founder, my story, but what do we as a group? What are our shared and collective values, and how do we make those manifest in the world?”
I’ll leave you with one last idea. We collectively have so much power it’s crazy. Because the system the way it works now, what I’ve learned about our current, what they call late-stage capitalism, financialized engine of capitalism, is that it is value-destroying in so many ways. And although markets discipline companies and they require them to create more value than is destroyed to stay alive, new companies, the ones that we’ve built, are so valuable.
The core insight that makes the thing is so valuable that the being that you’ve built can withstand so much damage before it collapses. And yet, if you look at the average tenure of companies in public companies, you just look at average CEO tenure, all the metrics you look at are really grim here. We’re destroying companies faster than we create them, such as that the total net number of public companies in America is down more than half from a peak that was in our lifetime. It’s not that long ago. What’s going on? Well, the system requires a steady supply of fresh meat to replenish that which it destroys.
And if you’re listening to this podcast, you probably are one of the peddlers. So you’re the manufacturer of the thing that the system requires for its basic survival. And so as a matter of fact, we as a group, we get to set the terms by which this is done. And so if we want to stop selling our beings, our babies, our children for money, we don’t have to do that. If we want them to have a certain ethos themselves in the world, we want them to be trustworthy counterparties and do something for the people we care about, we can insist that that is the precondition upon which we will give our everything to.
And so I don’t want to say that I know all the answers or that my values are the ones that should win out, but I do think that we do have a certain shared set of values that we ought to be more ardent in defending. And I hope your listeners will take that to heart.
Lenny: A beautiful and empowering way to end this. Eric Ries, thank you so much for being here.
Eric Ries: My pleasure.
Lenny: Bye, everyone.
Thank you so much for listening. If you found this valuable, you can subscribe to the show on Apple Podcasts, Spotify, or your favorite podcast app. Also, please consider giving us a rating or leaving a review, as that really helps other listeners find the podcast. You can find all past episodes or learn more about the show at lennyspodcast.com. See you in the next episode.
Glossary
| English | 中文 |
|---|---|
| 2021 Vintage | 2021 年份(指 2021 年融资的那批创业公司) |
| agency problems | 代理问题 |
| agent | agent(智能体) |
| Agile | Agile(敏捷开发) |
| Al Ries | Al Ries(定位理论创始人) |
| alignment | alignment(对齐) |
| Andor | 《Andor》(安多) |
| Ben Silbermann | Ben Silbermann |
| best practice | 最佳实践 |
| big room | 大房间 |
| board mission pledge | 董事会使命承诺 |
| cap table | 股权结构表 |
| chief engineer | 总工程师 |
| Clay Christensen | Clay Christensen(克莱顿·克里斯坦森) |
| cold sales outbound | 冷启动销售外联 |
| Conway’s Law | Conway’s Law(康威定律) |
| deductive logic | 演绎逻辑 |
| Delaware law | 特拉华州法律 |
| descriptive | 描述性 |
| DevOps | DevOps |
| engagement loops | engagement loops(参与循环) |
| ESG | ESG(环境、社会和公司治理) |
| fiduciary duty | 受托责任 |
| gatekeeping | 把关 |
| Hershey | Hershey(好时) |
| hockey stick | 曲棍球棍曲线 |
| IMVU | IMVU |
| innovation accounting | 创新会计 |
| inverse kinematics | 反向运动学 |
| jobs to be done framework | jobs to be done 框架(待办任务框架) |
| just so stories | ”就是这样”的故事 |
| Karri Saarinen | Karri Saarinen |
| keiretsu | keiretsu(企业集团) |
| Ken Liu | Ken Liu(刘宇昆) |
| law of diminishing returns | 收益递减定律 |
| law of sustainable growth | 可持续增长定律 |
| Lean Startup | 精益创业 |
| leap of faith assumption | 关键信念假设 |
| Lenny | Lenny |
| Levoit | Levoit |
| Linear | Linear |
| local maxima | 局部最优 |
| long-term benefit trust | 长期利益信托 |
| LP | LP(有限合伙人) |
| LTSE | LTSE |
| magic pixie dust | 魔法仙尘 |
| Martha Wells | Martha Wells |
| Merchants of Death | 死亡商人 |
| Murderbot Diaries | 《Murderbot Diaries》(杀手机器人日记) |
| MVP | MVP(最小可行产品)(已在术语表中) |
| naval gazing | 自省/内省 |
| net positive | 净正面 |
| outtakes | 废弃版本 |
| paid engine of acquisition | 付费获客引擎 |
| Paul Graham | Paul Graham |
| Philip Morris | Philip Morris(菲利普·莫里斯,烟草公司) |
| pivot | pivot(转向) |
| prescriptive | 规范性 |
| procurement policy | 采购策略 |
| product-market fit | product-market fit(产品市场契合) |
| public benefit corp | 公益型企业 |
| reality distortion field | 现实扭曲力场 |
| referrer logs | 反向链接日志 |
| SEC | SEC(美国证券交易委员会) |
| shareholder primacy theory | 股东至上理论 |
| silk punk | 丝绸朋克 |
| singular control | 单方面控制权 |
| Six Sigma | Six Sigma(六西格玛) |
| Smithsonian | Smithsonian(史密森学会) |
| social purpose corporation | 社会目的公司 |
| span of control | 管理幅度 |
| spiritual holding company | 精神控股公司 |
| SPV | SPV(特殊目的实体) |
| Stage-Gate | Stage-Gate(门径管理) |
| Steve Jobs | Steve Jobs |
| sticky engine of growth | 粘性引擎(增长引擎之一) |
| The Dandelion Dynasty | 《The Dandelion Dynasty》(蒲公英王朝) |
| The Sims | 《模拟人生》 |
| Thomas Kuhn | Thomas Kuhn(托马斯·库恩) |
| turn the crank | 拧扳手式(机械式) |
| uncanny valley | 恐怖谷 |
| viral loops | viral loops(病毒式循环) |
| Warren Buffett | Warren Buffett(沃伦·巴菲特) |
| willfully ignorant | 故意的无知/故意装无知 |
| Y Combinator | Y Combinator |
| Zuck | Zuck |
Reformatted by reformat_english.py
对一个运动的反思 | Eric Ries(精益创业方法论的创立者)
访谈实录
Eric Ries: 人们总觉得创业失败是能发生在你身上的最糟糕的事。天哪,这连前十都排不上。它确实不好,我经历过,确实很痛苦。真的很难受。但更糟糕的是身处一家死不了的公司——一家僵尸公司,半死不活,你恨它,却又离不开。唉,我见过太多这样的人了。创业者群体中正在经历一场心理健康危机,而这个问题还没有被充分讨论。人们开始谈论心理健康方面的下行风险,当然,做创始人的压力本身就很大。但看看那些所谓”成功”的人又怎样呢?当你建了一家公司,然后把它卖掉,它变成了让你厌恶的东西——也许你赚了钱,但这并不好。所以,建一家你讨厌的公司,一家在世界上变成了恶意力量的公司,一家你不得不假装自己与它无关、或者让你觉得自己是同谋的公司……那要糟糕得多,糟糕得多。我真希望更多创始人在早期就能更认真地对待这个问题。
嘉宾介绍
Lenny: 今天我的嘉宾是 Eric Ries。如果你对他的工作不熟悉,我会很震惊。他最著名的成就是创立了精益创业方法论和运动,以及他那本极具影响力的著作《精益创业》。他还创造了比我想到的任何人都多的、已成为科技文化一部分的术语和概念。他目前把时间花在为创始人和创业公司提供咨询上。他曾是创始人和 CTO,目前是长期证券交易所的创始人兼执行董事长。
这是我目前最喜欢的一期播客,我迫不及待想让你们听到。在我们的对话中,我们涉及了很多话题,包括精益创业运动的现状、他看到的未来趋势、如果能重来一次 Eric 会怎么做、最令他困扰的关于该方法的误解、AI 可能如何影响产品开发和创业公司、如何正确理解 MVP、何时 pivoting 何时坚持、以及如何建立一个与人类繁荣相一致的价值观的企业。此外还有大量精彩的故事和教训,这次对话里有太多有价值的内容了。
Lenny: Eric Ries,非常感谢你来做客。欢迎来到播客。
Eric Ries: 谢谢你的邀请。
Lenny: 能邀请你上播客感觉很不真实。你在产品和创业领域是如此传奇的人物,我真的非常激动能向你学习,也让听众能向你学习。我的第一个问题是,你最近在忙什么?我知道你在做长期证券交易所。Eric Ries 的生活中还有什么?
精益创业之后的历程
Eric Ries: 你信不信,《精益 Startup》是 2011 年出版的,到现在已经有一段时间了,这一路走来真是疯狂。基本上自从它出版以来,每年我都在想,“好了,结束了。应该差不多了吧。“然而每一年都变得越来越疯狂,新的事物不断涌来。所以,我在建设长期证券交易所(LTSE),我们后面肯定会聊到,同时我仍然花大量时间和创始人以及大公司高管在一起,帮他们解答关于如何以正确方式建设公司的问题。
显然,最近我也被卷入了很多 AI 相关的事情,和所有人一样。有一个让我意外的有趣之处——过去十年我花了很多时间研究治理和公司治理的问题,研究公司应该如何架构,本来并不是特别针对技术的,而是以长期主义和人本主义为核心。结果发现,这些问题在 AI 公司身上变得尤为紧迫。正因为我在治理方面的工作,我得以接触并和一些顶尖的 AI 公司合作,这非常有趣。但我觉得自己是跟着人们关心的问题走的,这些问题把我引向新的、真正有趣的事物,一直在牵引着我。
对创业文化与语言的深远影响
Lenny: 太好了,这些问题我们在后面都会聊到。有件事我想提一下,我觉得人们对你的贡献给予了不够的认可——你发明或普及了多少创业和产品构建的核心概念。据我所知,显然有精益创业方法论,然后有 MVP 这个术语,还有 pivoting 这个术语,还有 A/B 测试、客户开发、持续部署、虚指标这个说法,我猜还有很多,然后还有长期证券交易所。我从中感受到的主题是,你做了大量的第一性原理思考。所以这里有两个问题:第一,你有没有想过自己会对创业文化和语言产生这么大的影响?第二,你觉得这种第一性原理的思维能力从何而来?你似乎特别擅长这个。
Eric Ries: 首先,谢谢你这么说。你给我太多功劳了,不过我挺喜欢这个想法的——可以抵消掉我据说在别处得到的太少的认可,挺好的。回答你的问题:不,我完全没想到会发生这样的事。现在说出来真的很尴尬,但让你感受一下那个时代和金融危机时有多么不同——我当时写博客觉得特别不好意思,连名字都没署。我是匿名发表的,因为那时候人们不做这件事,别人都跟我说这是职业自杀。当时写创业博客的人太少了,排名前三的博主全都亲自联系了我,因为他们看到自己网站的反向链接日志里出现了我的链接。他们的流量就是那么少,以至于当有新人写相关话题并链接到他们时,他们就觉得:“这人是谁?怎么回事?“那个时代真的很不一样。
我当时刚结束一段创业经历。我离开了一家正在走向成功的创业公司,在风险投资人中间有个名声,说我能让工程团队的生产力超乎寻常。说我有魔法仙尘。所以他们请我去跟他们的团队合作,我就说:“不,说真的,我没那么特别。跟我本人没关系,就是这套框架。我有这些想法,团队采用之后确实有帮助。“他们就会拍拍我的头说:“好的好的,孩子,我相信那很好,但你能不能还是把魔法仙尘带到我公司来?”
我去开那些会,我没有夸张,真的有人在会上冲我大喊大叫。我被人吼。实际上,人家不太客气地请我离开,我就说:“我不明白。你们冲我喊什么……你们为什么生我的气?这会是你们约我来的啊。“他们就……我只是在说我亲眼所见的事情。我们在 IMVU,平均每天发布产品五十次,而那时候能做到每月发布一次的就算非常先进了。要知道,不太久以前我们还会把产品发布的年份放到产品名称里——这就能告诉你当时的发布周期有多长。那被认为是正常的,而人们就说:“不可能。那绝不可能奏效。”
那个时候,我甚至不是在提出什么理论,我只是在解释我亲眼看到的事情,但人们就是不敢相信。于是我就想了个好主意:把这些故事写下来,这样当有人要求我……我一直在做这种会面。我当时不理解硅谷是怎么运作的,不理解为什么这些会面会发生在我身上。我就说:“我把故事写下来,把文章发给他们,如果他们觉得这太离谱,那我们就不用见面了,我也不会被人吼。“这就是我当时想到的最远的地方了。然后这件事基本上占据了我的全部生活,有人请我演讲,人们真的很想揭开匿名者的身份——这个人到底是谁?于是我就公开了身份,开始用自己的名字写作、演讲、跟公司合作,一切变得像旋风一样。
关于第一性原理的部分,我不知道,我一直就是这样的思维方式。我觉得自己可能是一个失败的科学家。我真希望自己擅长化学之类的,那样我也许能成为一个真正的科学家。但我一直对真理、对科学思维怀有一种由衷的敬畏。这是我从小被灌输的重要文化价值观。我的父母都是医生,这确实是我成长环境的一部分。我也很喜欢读那种以科学方式运作的东西。所以这对我来说非常自然。当我作为创业者取得成功的时候,我特别想知道:这到底是为什么有效?别人给我的建议全是这种模式:“Steve Jobs 曾经做过这件事,所以如果你也这么做,你就会像 Steve Jobs 一样。“我就想,他还穿黑色高领毛衣呢,那也是其中的一部分吗?帮帮我理解一下,它到底为什么有效?在当下还适用吗?对那个行业来说合理吗?我不喜欢这种”就是这样”的故事。
回头看,我当时并没有这种认知,但回过头来我觉得,第一性原理思维其实有两个组成部分,人们往往把它们当作一回事,但其实其中一个是更根本的。我的很多工作其实只是描述性的。并不是在告诉人们该做什么,而是给一个已经存在的事物赋予名称和概念。这是最有趣的部分。在早期反对精益创业的那波批评中,人们对我说创业就是一个实验这个说法非常愤怒——“你不能把创业当成实验。“我就说:“你可以用任何你想要的方式对待它,但它确实就是一个实验。“人们会说:“人们不喜欢 pivot 这个概念。“这些反 pivot 的人、反 MVP 的人,偶尔还会冒出来抱怨一番。
我就说:“我没有叫你一定要 pivot。我只是在说,当你改变了策略但仍忠于愿景的时候,这个行为需要一个名字。“它需要一个名字,因为我们一直在做这件事,即使没做的时候我们也在谈论做这件事。我们无法对它进行推理,我们无法……所以这是描述性的部分,而说实话这才是真正的工作——就是说出它是什么……对我来说,从一开始最大的问题就是:创业公司到底是什么?它为什么跟大公司如此不同?为什么我职业生涯中学到的那些最佳实践一个都不管用?这就像一种生存本能,我想要知道答案。一旦你界定了它是什么,厘清了你所做的事情的基本原则,那么规范性的部分就只是演绎逻辑的事了。
就像说:“既然我们做了这个假设,那它的推论是什么?“当然,规范性部分也很重要,因为那是你检验假设是否正确的方式。所以这是另一件我觉得人们没有意识到的事情。人们叫我第一性原理思考者,我很喜欢这个说法,这是一个很大的赞美,但我觉得它给人造成一种印象,好像我一个人在山洞里待了好多年然后想出了一个绝妙的点子。不是的。我真希望人们能看到我尝试过的所有那些失败的概念。精益创业不是我的第一次尝试,这套概念体系不是我第一次就弄对的。它经历了大量的试错,大量的摸索什么有效、什么无效——先是在我自己的职业生涯、自己的工作中尝试,然后是试图给别人提供建议,把对我有效的东西转化过去看看对他们是否也有效,然后回答我认为自然而然会产生的问题。
在什么条件下它有效,什么时候不适用?这个理论做出了哪些可以被验证为真或为假的预测?再往大了说,我们到底在这里做什么?成功到底是什么样子的?我觉得这是我们所有人真正在挣扎的问题——想出一个理论、方法或流程来最大化某个东西,这很容易,但问题在于:“那个东西是好的吗?它真的在帮我们达成某个目标吗?“所以我认为第一性原理思维在这方面也很有帮助,但说到底,我真不知道该怎么回答你的问题。真实的答案是:我不知道。我就是一直觉得自己是那样的人。我热爱思想,想知道事物为什么有效。
第一性原理思维的两种成分
Lenny: 对我来说,你刚才分享的让我有两点收获。第一是你天生就是用这种思维方式思考的——非常合乎逻辑、科学地思考。我有类似的感受,当别人给我讲一个故事的时候……我当初开始写作,部分原因也是人们不断问我 Airbnb 是怎么做事的,因为他们正在搭建自己的交易平台。我的反应就是:“他们确实做了这些,但谁知道那是不是正确的做法呢?”
Eric Ries: 没错。他们是因为这些做法而成功的,还是尽管有这些做法依然成功了?
精益创业方法的现状
Lenny: 让我们深入聊聊精益创业方法和方法论吧。感觉我们已经很久没有听到关于它的最新进展了——它走向了哪里,它持续有多受欢迎。我很好奇,你对精益方法论和精益创业运动的现状有什么感觉?
Eric Ries: 哦,这是个很有意思的问题。我们以前每年都会举办精益创业大会,后来因为 COVID 停了,一直没恢复。所以我们过去每年有一次机会把大家聚在一起,评估现状,听取最新动态和故事。现在我真的很缺这个,很怀念。我们试过做线上版本,但感觉不太对。这有一部分是 COVID 的原因,也是我们这个世界过去几年所经历的变化,但还有一部分是——在真正做之前,我根本不知道发起一场运动会是什么样。
我还记得我们最初办的几次大会、最初的活动,有一种宗教复兴的感觉。人们会说:“我们拿到了新东西。我们要跟那些老家伙对着干。“我真的以为,“我的工作就是带领这群人冲锋陷阵。“我脑子里真的是一个军事比喻。我刚才也用了这个说法——思想的战斗。我已经准备好了,旧思想会跟我们拼到死,然后我们会赢。但完全不是那样。我们冲上战场,根本没有另一支军队来反击我们。我们是不战而胜,因为那些用老办法做事的人其实也不喜欢那些办法,也不想……旧方法的捍卫者非常少。几乎没有人站出来说:“不,Stage-Gate 才是正确的产品方法论。你们搞错了。“这种事从未发生过。事实上,那些光鲜的创业者们,从嘲笑精益创业毫无意义,到抱怨它被过度炒作、已经过时了,中间根本没经过一个”先了解它到底是什么”的阶段。这真的很有意思。我永远不会忘记,TechCrunch 很多年前发过一篇文章,说精益创业被过度炒作了。我当时就想,“你们甚至还没……”那是他们第一次提起这个话题。我说,“你们能不能至少先学会正确使用 pivot 这个词,然后再批评它被过度炒作?这有可能吗?”
所以,它从一股叛逆的复兴运动,变成了人们默认的做事方式。即便是反对它的人,为了批评它也不得不先接受这套概念框架。发现自己站在了这件事情的另一边,这种感觉很奇妙。我会遇到一些刚进入这个行业年轻创始人,他们会说:“这不是显而易见的吗?“我就想,“才五年前人们还觉得这东西疯狂到对我大吼大叫,现在你觉得显而易见了?”
看到一样东西从充满争议变成理所当然,这真的很有意思。当然,那些一开始强烈反对它的人后来都回来了。现在他们说他们一直都知道,说自己是我的大……你能看到人的观念是怎么变的。所以,当每个人都觉得某件事理所当然的时候,你很难再围绕它制造出那种宗教般的热情和兴奋。对我的……就说自尊心吧,我想要那种胜利的认可和种种肯定。但在某个时刻我意识到,胜利不在于某个大牌公司使用了它,真正的胜利在于人们告诉我这本书对他们有帮助的那些故事。
这才是全部的意义。我们赢了,不需要任何人输。现在人们不太这么做了,但偶尔还是有人想把精益创业和 Agile、或者 Agile 和设计思维、或者这个和那个对立起来打架……我真的尽力了。如果你看精益创业那本书,里面充满了对 Six Sigma、精益制造、设计思维、客户开发、DevOps、软件工匠精神以及我能找到的一切的引用和关联。我真的想展示这些东西是如何协同工作的。然后我们经历了一个阶段,人们想让我把它真正变成一个宗教。有人写信给我说:“某某人在网上说精益创业怎样怎样,他们说的不对。你得让他们闭嘴。“我就想,“你觉得我有什么权力让人不写东西?我觉得他们想写什么就写什么。”
还有另一件事人们会说:“精益创业就是这一套特定的实践。如果有人发现了新的实践,他们就是异端,应该被逐出教门。“我就想说,“各位,不不不。我在书里写得很清楚……你们读了吗?这是一个科学理论,这意味着凡是有效的就是精益创业。所以如果有人提出了新的东西并且它有效,我们不应该感到威胁和不安,我们应该采纳它。“当你抱着这种态度——凡是有效的就是对的,真相才是重要的,人们私下里觉得有帮助——这就是我们衡量成功的标准——那它就把那些驱动媒体报道的炒作、戏剧性冲突之类的东西全部抽走了,而我一点也不怀念那些。我现在这样比以前……我以前经常出差,做更多公开活动,感觉好多了。我觉得部分原因还是 COVID 的变化,会议活动消停了一段时间,所以是一个安静的时期,但我感觉这段时间其实非常棒。那就是胜利的样子。他们不会给你举办一场游行,说这个东西……而且尽管那些批评你的人仍然会抱怨,但到头来,新进来的创业者理所当然地认为这就是创业的概念词汇体系,然后他们用它做自己想做的事,我学会了真正欣赏这一点。尽管我的自尊心当然也喜欢跟人争论。但就像——不,这样其实更好,这才是赢的样子。
Lenny: 很有意思的一点是,你在传播这些理念方面做得太好了,以至于它已经变成了文化的一部分,人们甚至不会想到:“哦,这些全是 Eric Ries 提出来的。“我也觉得人们会对你不同意的部分提出质疑,但对那些运作得很好的部分却从不给你功劳。他们只会说:“哦不,MVP 已经不管用了。”
Eric Ries: 每年至少还有一次,会有人写信给我说我们应该用一个不同的词来替代 MVP,说围绕 MVP 的误解都是因为这个词的选择造成的。我每次都回复他们说:“请告诉我更好的词是什么,把它推广开来,我就会用。这是个好主意。“比任何人都更清楚 MVP 的局限性的,是我自己。我再告诉你一件我觉得人们不够重视的事——现在人人都想当思想领袖,这简直像一种流行病。Clay Christensen 去世的时候,我认识他,也非常敬佩他,他对我的工作的认可,是最重要的……
思想的责任
Eric Ries: 对我来说,来自老一代管理思想家阵营的他能说这项工作是值得的,对我来说真的是一件大事。然而当时有一场很大的争论——现在稍微久远一些了——关于他是否应该为人们以”颠覆”之名做下的所有坏事负责。我第一反应是,“这是对 Clay 的恶意中伤。太不公平了,一堆科技圈的人把’颠覆’当成各种蠢事的代号,跟他本人的理论毫无关系,跟他的工作毫无关系。“就像,“他们到底读过那些著作吗?“我是他的捍卫者之一,尤其在他去世之后,我对此很激动,这太不公平了。
但后来我越想越觉得,“天哪,也许我们对放进这个世界的东西确实负有一定责任。“这很可怕,因为当然,你无法控制人们会做什么,你无法控制……很多人是故意的无知。每当有人说什么话让我觉得,“天哪,上 Wikipedia 查五秒钟就能搞清楚他们在说的事情的真相。“所以我们生活在一个无知是可选项的时代,这挺酷的,但也有点可怕。所以我认为承担起这种责任——这对任何做产品的人都适用。我认为这其实完全不限于作者。当你做出一个东西,把它放到这个世界上,一方面,你对接下来发生的事没有任何控制权,但我确实认为我们对它是否真的是一个净正面、是否对人们有帮助,负有某种责任。
这也是激励我努力把它做对的原因之一——不仅仅是为了自己的满足感而做对,而是通过人们使用后的实际效果来验证它是否做对了。这就是为什么测试和实验是我生活和我理论中如此重要的一部分。对我来说,这就是我们在这个高度不确定的世界中承担自身责任的方式——去搞清楚人们使用这些理念后到底发生了什么,如果不对,即使排除了他们的误解之后还是不对,那你就得改进它,找到更清晰的传达方式。所以这可以说是一个永恒的挑战,但我认为是一个重要的挑战。
关于精益创业的误解
Lenny: 顺着这个思路,关于精益创业或者你提出的其他东西,有没有什么误解真的让你很沮丧,是那种人们反复搞错的持续性的误解?
Eric Ries: 我写过一篇文章,天哪,五还是十年前。天哪,我现在都记不清是多少年前了。标题大概是《关于精益创业的最大误解》,里面说,“第一,精益就是省钱,所以如果你在做精益创业,就意味着你不融资。“然后就是,“这不是事实。“然后我大概还记得后面几条……它们现在依然很普遍,什么都没变。我今天完全可以把那篇文章再写一遍。比如,“精益与拥有愿景是对立的。如果你真有远见,就不会去做实验。实验会导致局部最优”,等等。
我在某人写的书里读到精益就是关于这些东西的。我当时就想,“天哪,我是不是忘了在书里写……”我被搞得都自我怀疑了,甚至回去翻开一本《精益创业》,想看看”我是在哪一页讲到这个的?“结果就在引言部分的第九页,全书最早的内容之一,我当时就松了口气,“好吧,我确实写了。“所以那种把精益等同于拧扳手式优化的看法非常普遍。然后就是 pivot,人们 perennially 误解 pivot 是什么,而且,它到底是一个描述性还是规范性概念,这确实很微妙。
我现在觉得人们只是在故意装无知。批评精益创业完全可以,没问题。我不介意。我觉得批评是好事,是健康的,但你只是把那些老掉牙的东西再搬出来,这对推动最前沿有什么帮助?对创业者有任何帮助吗?我甚至不太确定他们这样做到底想达到什么目的。这就是我越来越困惑的地方——这种话语的目的到底是什么,除了条件反射式地唱反调?我不知道。
Lenny: 还有在 Twitter/X 上赚点点赞。
Eric Ries: 也许你说得对。
如果可以修改原书
Lenny: 顺着同样的思路,你写那本书已经是 12 年前了。如果你能回去修改书里的某些内容,有没有什么你想改的?
Eric Ries: 天哪,我什么都想改。这其实正是我不让自己去改的原因。当我还是个读者的时候,我最讨厌的就是作者推出第二版,把第一版的精华都毁掉——包括那些错误,因为他们在试图给所有内容加上限定条件、让一切变得更细致的过程中,完全丧失了论证的力度和本来要做的事情的力量,而且他们自己的成名在心理上也会搞乱你。所以,别让你的读者承受这些。去看心理治疗,别毁掉当初让你的书了不起的东西。所以作为作者,当然,我觉得书里那么多地方都可以改进。最让我觉得刺眼的……嗯,大概有两处。我说两处吧,因为它们比较相关。
一处是我有点不好意思,书的后半部分有一章讲的是精益创业的规模化。大概就是,“好吧,你已经取得了成功。那么,当你扩大一个现在已经包含许多精益创业团队的大型企业时,遵循的原则是什么?“之类的。我觉得那一节我并没有写错什么,但我现在读起来真的觉得很难为情,因为太轻描淡写了,就像,“第一步,做这个。第二步,做那个。“听起来好像特别容易、特别简单。
当然,我后来专门写了一整本书——比《精益创业》还长——就是为了阐述那些原则,解释它们在现实中如何运作,因为我确实有幸能在真正非同寻常的规模上与公司一起去做这些事情。所以我现在有点不好意思,觉得”我当时把它说得太容易了”。我不想误导任何创业者让他们以为创业的任何部分会是容易的,因为那只是在让人失望。另一处好笑的地方是,你知道 Reddit 上那个梗吗,星球大战那个梗,Anakin Skywalker 说”我要改变世界,让它变得更好”,然后停顿一下,然后她说的是,“变得更好,对吧?“我觉得这就是我们这个时代科技行业的缩影。每家公司都在说要改变世界,而我们忘了问,变得更好?我一直理所当然地认为我们的目标是让世界变得更好。这是我从小被教育的价值观。我以为我们做的就是这件事,我也为此抱怨过一段时间。我会变得沮丧。还记得”不作恶”吗?科技行业有过那么多这样的时刻,看起来我们真的要做一些非凡的事情了,然后这些公司变成的样子又那么令人沮丧。我当时就想,“好吧,至少我说清楚了。“然后我又想,“我说清楚了吗?"
"变得更好”的遗漏
Eric Ries: 后来我真的回去重新翻开了那本书,心想:“我有告诉过任何人他们应该改变世界吗?“如果我确实说过,我假定我加了”变得更好”,结果确实白纸黑字写在那里。就在《精益创业》的引言里,有一句——几乎是随口一句——“这项技术将为未来的创业者提供改变世界所需的工具。“我当时就想,“不,我应该加上’变得更好’的。我不知道还需要特别说出来。我以为这是不言自明的,以为我们对此有共识。“所以我觉得这也是一个重大疏漏。我现在说得轻松,但它的后果是灾难性的。其实也不是真的灾难性,我不认为《精益创业》在那个意义上误导了人们。我觉得更多是其他社会力量的作用,但我很遗憾自己错过了一个机会,在一个后来被证明相当重要的问题上表明立场。
观众提问:MVP 还适用吗?
Lenny: 你看到我在推特上征集问题了。
Eric Ries: 看到了,我非常期待。
Lenny: 我这里有几个观众的问题。第一个是关于 MVP 这个概念的。Karri Saarinen——他最近刚好上过我的播客——
Eric Ries: 哦,太好了。
Lenny: Linear 的创始人,他有一个关于 MVP 的问题。他的问题核心是:考虑到如今各个市场的用户期望都大幅提高了——比如 iOS 应用、Web 生产力软件等等——你是否仍然认为 MVP,也就是发布最基础的版本然后不断迭代,仍然是正确的做法?
Eric Ries: 我很喜欢这个问题,首先因为我经常被问到,其次这确实需要我们真正理解清楚。人们以为 MVP 是关于某种具体战术。所以 MVP 就是一个简陋的、功能被砍得七零八落、会把你的电脑搞崩溃的东西——但这跟 MVP 完全无关。MVP 的本质很简单:对于我们试图验证的假设,什么是最有效的方式来获得我们需要的验证——这个假设到底成立还是不成立?而一次成功测试的一部分,就是要理解我们市场中客户的需求是什么。
有趣的是,人们在产品上对失败有一种真正的恐惧——我们所有人内心深处都有这种深深的恐惧。所以我想说的第一点是,你觉得 iOS App Store 里用户期望很高?你应该去试试数据中心备用电池系统的采购者,或者深海石油钻井设备的客户,又或者是喷气发动机的买家。我合作过的一些公司,标准才是真正的高,相比之下 iOS 应用简直是小菜一碟。“你们的应用又不会爆炸”,但即便在这些真正高端的行业里、在真正高风险的场景下,向别人提供一个他们不喜欢的产品,成本其实没那么高。他们只会说不要。而 MVP 很大程度上就是控制犯错带来的损失。
所以我确实鼓励大家去做的一件事是:去找 10 个客户——不是一千万个,就是 10 个。让他们使用你的产品,然后让他们告诉你哪里糟糕透顶。这完全可以。我以前特别受不了……以前我发布产品的时候,如果客户跟我争论,人们会说:“你这个白痴,我简直不敢相信你居然没有 X、Y、Z 功能就敢发布这个产品,这些功能明明那么重要。”
我当时的反应是:“你错了……”我会冲他们生气,但后来我聪明了一点。我会说:“X、Y、Z,你说?“然后开始记笔记,“哦,有意思。“有时候 X、Y、Z 刚好就是我路线图上接下来的三个功能,我又会想跟他们争辩:“我早就知道了……”最终我学会了说:“哦,感谢您的宝贵反馈。“等我们发布 X、Y、Z 的时候,他们会产生一种主人翁感。“哇,是我告诉他要这么做的,他本来都不会做。“当然了,作为创业者,我们真正害怕的——如果我们足够诚实的话——不是别人说”你个白痴,你没有 X、Y、Z 就发布了”,而是别人说”你个白痴,你居然没有 A、B、C 就发布了”。
你心想:“A、B、C 是什么东西?从来没听说过。“这才是我得到的更常见的回应。
所以,让我们接受这个问题的前提:期望确实很高,如果我们发布一个应用,达不到人们的期望,他们就不会喜欢。首先,这到底有什么后果?在《精益创业》的早期阶段,我们花了很多功夫向人们解释产品发布和营销发布之间的区别。你可以在不告诉任何人的情况下把产品发布给客户。这完全可以。没人会发现,也不会成为什么大新闻。当然,一旦你是一家成功的大公司,情况就完全不同了——这时候你把企业品牌贴在了产品上——但如果你是一个真正的创业公司,没人在乎。所以去试吧。
但另一件人们没有充分意识到的事情是,MVP 本身是非常灵活的。所以如果你跟我说……当我跟创业者合作时,我会说:“听着,抱歉,我们的客户……门槛在那里,产品必须达到那个水平。“我会说:“太好了,那它能有很多功能吗?恐怕不行吧?“他们会说:“你什么意思?""你刚才亲口说门槛在那里那么高,所以我们应该只把那一个关键功能做到那个质量水平,如果验证通过了,再去做下一个。“当然他们接着会说:“哦,不不不,我需要把所有功能都做到那个质量。“我就说:“那我们到底在验证什么?等你发现那东西到底好不好用的时候……”如果你去看那些 pivot 的故事,我相信你也一定发现了——第一个版本中 90% 的工作最终都会被扔掉,不管当初质量有多高。
事实上,作为一个工程师,这也是我最初对 Agile 软件运动感到幻灭的原因,后来对精益创业也是如此——人们向我承诺,只要我遵循这些最佳实践,就能消除工作中的浪费。但如果我做出了客户根本不想要的东西,那才是终极的浪费,而没有人保护我免受这种浪费。所以我并不在乎初始 MVP 的质量有多高,因为质量是由客户定义的。如果我们连自己的客户是谁都不知道,我们甚至根本不知道”质量”这个词是什么意思。
我认为创始人真正陷入困境的地方就在这里。每个人都觉得:“我是 Steve Jobs,我有那种神奇的品味。“好吧,首先,你有没有经历过那些塑造出那种品味的长达数年甚至数十年的失败和经验积累?如果你有,那我愿意相信你。如果你没有,“那绝对不要依赖这一点,一定要做一些实验,去积累那些伤疤。“但即便你真的有那种品味,即便你完全正确,再验证一下又能有什么坏处呢?我的想法是:“让 10 个客户看看这个东西,让他们对它糟糕到什么程度大吐苦水,这至少能确认我们的方向是对的。“
建立基线
Eric Ries: 然后在精益创业的创新会计和度量指标这一侧,我们把这叫做”建立基线”——先去看看现在的情况有多糟,这样我们才能知道自己是否有所改进。因为我从那些以质量为名迟迟不肯发布产品的人身上真正看到的是,他们以很高的质量发布了产品,但没人用,而我们不知道为什么。是因为我们对质量的理解是错的吗?是因为价值主张从根本上就有问题?我觉得,把一个错误的价值主张做得高质量,没什么值得骄傲的,这里面没有任何匠人精神。我曾经遇到一家公司,做了一个非常漂亮的产品,精美的落地页,一切都棒极了,但有一个 JavaScript 错误——当你点击下载按钮、注册按钮,什么按钮都点不了。他们上线了这个东西,注册率是 0%,然后他们说:“这是灾难,我们做错了价值主张。“我就说,你怎么知道?从一次实验中你永远无法知道。你必须愿意做一系列实验。再说一遍,MVP 的逻辑其实很简单:产品需要时间来构建。世界上最优秀的产品,所有东西都需要时间。一个月,一年。AI 会意味着我可以瞬间构建任何我想要的产品——但并不是瞬间的,还是需要时间的。
在产品处于构建期间,总会存在某个时刻你只有半个产品。从逻辑上讲,你先有 10%,再 20%,以此类推。当你拥有半个产品的时候,就让客户来用这半个产品。这不会额外花费你任何成本。你本来就必须经过这个阶段,自然而然地就在做一个实验。让他们用它,让它告诉你它只有一半好,或者糟糕到根本没法用。太好了,很好。
少即是多
创业文献中充满了”少即是多”的案例——那些做出来的粗糙小东西最终比竞争对手的产品更有价值。我只举一个例子,也是我最喜欢的一个。我的背景是 3D 图形、虚拟形象和元宇宙相关的东西,当然现在这东西名声不太好。这些技术会经历各种炒作周期,我就在想:“我们到底是在炒作周期里,还是我们是泡沫破裂后落下的水滴?”
Lenny: 我觉得现在又在上升了。
Eric Ries: 又在上升了?我真的很搞不清楚。即便我还在做这一行的时候,每隔几年,它要么是未来,要么没人愿意碰。我猜那些相关产品的销量也是如此,会有那种周期性的波动。热的时候人们看,不热的时候就不看。不管怎样,我们当时做了一个版本——IMVU 的第一个版本,在咖啡厅或场景中内置了 3D 虚拟形象。这些虚拟形象全都固定在原地,因为我们没有一种叫做反向运动学(inverse kinematics)的技术。我们没有那种能让虚拟形象站起来走动的技术。做 3D 动画的时候,它必须完美运作——脚必须贴着地面滑动,否则看起来就不对劲。你会掉进恐怖谷,人们会觉得它很糟糕。所以我们说”那不行”,就让虚拟形象坐在那里。结果人们觉得这个产品很压抑,他们总是抱怨:“《模拟人生》都能让我到处走动,为什么我在这里被困在椅子上?”
这是客户反馈中非常普遍的意见。我们说:“我们做不起那个昂贵的东西。做不到,做不到。“后来有一天,作为一个 hack,一个小 demo,我说:“那我就改一下——当你在场景中点击另一个位置时,我们直接停止渲染原来位置的虚拟形象,然后把它放在新位置上。“我们甚至没有做正确的切换动画,镜头会嗖地移过去,真的很丑。我对此非常尴尬,我说,让我先玩玩看会发生什么。
我们发布了它,我内心觉得这低于质量标准。作为工程师我很过意不去,所有参与的人都觉得很糟糕。但我们已经承诺过要尝试做实验,要看看会发生什么。这是在精益创业存在之前,我们甚至没有 MVP 这个概念。我们就打算试试看会发生什么。发布之后,什么事也没有,没人在意。我松了一口气。那时候我们会做每日跟踪调查,随机抽样用户问他们对产品的看法。在跟踪调查中,人们开始说 IMVU 是他们用过的最先进的虚拟形象产品。
以前人们总是拿我们和《模拟人生》比,而且比得很不利。现在人们会说比《模拟人生》还先进。我就想,这是什么意思?我们怎么变得更先进了?我们改进了 3D 图形设计吗?没人理解他们为什么这么说。我们以前会做客户访谈,终于有个客户说:“太棒了老兄,在 IMVU 里你想去哪儿就传送过去。《模拟人生》里你得等那个人站起来走过去,太烦了。“我们当时就想,天哪,太丢人了。我们在很长一段时间里都没有做反向运动学。
Lenny: 我太喜欢这个故事了。听你讲 MVP,我学到的一点是,“最小”这个概念是非常灵活的,非常因情境而异的。本质上,这个信息不是说”做一个低质量、简单的东西”,而是要搞清楚你的市场和产品的最低门槛是什么。核心信息就是尽可能少地构建——无论那个”少”对你来说意味着什么。
最小不等于低质量
Eric Ries: 对,尽可能少但必要。这很有意思。我没开玩笑,我做过喷气发动机、机器人、癌症疗法,各种各样的事情。LTSE 那个项目,我们花了 10 年才做出 MVP。所以它不一定意味着廉价低质量。它只是说——无论你的假设是什么,无论你真正相信什么会奏效,让我们把那个东西拿去验证。这就是我真正想要的。当我和公司合作时,尤其是初次创业者,他们通常对此非常固执,不愿意这样做。我就说:“听着,你告诉我你相信什么。我不需要是对的,我只是想用某种方式验证一下你是对的。“然后当人们真正开始翻来覆去想这个问题的时候,他们往往会说:“嗯,你知道吗……”一个经典的例子是:“在造出实体设备之前,我们没法卖这个东西。”
我就说,你的产品是在展厅里人们看到实体设备后购买的吗?不是,他们是看宣传册购买的。那好,“那我们给他们发一份宣传册,让他们预购,怎么样?“然后他们说,什么?我不能那样做,我不能拿一份粗制滥造的宣传册出去。我说,谁说是粗制滥造的宣传册了?我们可以在极短的时间内做出一份质量很高的宣传册,比造出那台机器所需时间的一小部分就够了。然后他们说,那你什么意思,要用好看的字体和漂亮的图片?也许吧。你的理论是什么?你告诉我你相信什么……然后他们会说:“实际上,真正重要的是宣传册里设备规格说明的质量。“这就对了——现在我们找到了关键信念假设(leap of faith assumption)。让我们去验证它。
数据中心的教训
Eric Ries: 然后他们说,等等,但我们还不知道那些规格参数会是什么。我就说,好吧,我看到那边那份 500 页的计划了。你希望它是什么?他们说,如果客户能接受这组性能指标那就太好了。那我们就去承诺这个,也许他们会说:“太好了。“我举一个例子。我之前提到过数据中心。有一个团队卖数据中心设备。他们有一支由博士组成的高端团队,专门做研究来提高设备效率,为了把这个产品做出来,他们甚至要去拿材料科学的诺贝尔奖。我说服了他们,与其花三年时间和 1800 万美元还不一定能造出这个东西,不如拿一份宣传册,去找一些客户。
我永远忘不了,他们把那份宣传册拿给客户看,客户把他们笑话出了门,而且跟他们关心的任何东西都毫不相干。跟效率完全无关,客户说,我们绝对不会买这个。团队说:“但你连第二页都没看,看看效率有多高。“客户说:“我们不为效率买单,我们建数据中心。运营成本是别人的问题。“你只能感叹,“天哪。”
这就是人们没有意识到的地方——这些洞察在事后看来如此显而易见,以至于每个人都坚信这种事不会发生在自己身上。当我在教精益创业课程时,学生们会说:“这太蠢了。“他们在商学院做一个接一个的案例研究,每次都说”蠢、蠢、蠢、蠢、蠢,对吧?“他们觉得:“我绝对不会犯那种错误。“我总是说:“听着,你不用跟我争。等你找到工作再说。等你犯了其中一个错误的时候,记住我们有过这些对话,然后去读那本书。“当然,碰巧世界上也有些人就是从来没犯过错。
掷骰子嘛,你掷足够多次,总有人每次都是双六。好吧,确实有这样的人,而且世界上很多建议正是他们给出的。但我不觉得我们这些没有双六运气的人就非得去追随那些建议。
关于 pivot
Lenny: 这是一个非常好的过渡,可以聊聊 pivot 了,因为你的观点中很重要的一部分就是——你所做的大量工作本来就是会浪费掉的。既然如此,为什么要花那么多时间去构建它?不过,关于 MVP 这个概念,我想最后再说一点,因为我们收到很多问题都在问:到底什么是 MVP?Pinterest 的联合创始人 Ben Silbermann 有句话,他没有在我们的帖子里提问,但他有一句话——
Eric Ries: [听不清]
Lenny: 对。他的话是,MVP 的难点在于,你不知道”最小”是什么,你也不知道”可行”是什么。所以对于那些正在试图弄清这条线应该画在哪里的创始人来说,如果给一个非常简洁的建议,你会怎么帮助他们理解——你的 MVP 应该做到什么程度。
Eric Ries: 他没说,你也不知道”产品”是什么。这也是对的。我不知道有多少次,一个服务最终被证明应该是一个产品,或者反过来。人们用错误的方式将他们对客户的洞察具象化,这太常见了。这很有意思。
MVP 的建议
对 MVP 的所有批评,其实都是人们在表达对创业不确定性的沮丧。没错,这确实很糟糕。这不是一个好的谋生方式。如果你想保持心智健全,如果你觉得认知失调很痛苦,这份工作不适合你。因为你脑子里要同时装着两件事——我有这个宏大的愿景,我知道什么会奏效——你需要有信心,把所有人团结起来。在这个世界上,没有任何东西能替代一个人站出来说:“我要把这件事做成。“而当你这样做的时候,你就把自己推到了悬崖边上。这真的很难。很难承认这一点。同时我确实也不知道怎么才能把它做成。这是事实。不管你承不承认,这就是事实。
所以我说,让我们验证一下就好。简单、简洁的建议。第一个技巧是:列出你的 MVP 所需的功能清单,砍掉一半,再砍掉一半,然后构建那个。说实话,如果你只是这样做,就已经相当不错了。MVP 的好处在于它是一条 U 形曲线——只要你大致在最优附近,就没问题。而大多数人对”什么是必要的”的自然判断,离谱到可笑。人们通常会偏离一到两个数量级,这是很常见的。你经常遇到一个团队,他们试图做的工作量是验证其假设所需工作量的 100 倍。
所以第一件事就是,在你实际构建的东西上,试着做到让你自己感到不舒服的小。如果你在这方面遇到困难,在《创业之道》和精益创业的框架中,我详细讲解了——那本书更多是面向大型企业读者的——我真正系统地讲解了,如果你想有一个流程来决定在这里该做什么,你应该先头脑风暴你的关键信念假设,按这种方式分类,按这种方式筛选。好,现在对每个假设,头脑风暴你可能使用的指标,选择你想要的指标,你的学习指标是什么,然后从那里出发,头脑风暴 MVP,按这个公式给它们打分,然后用这个方法来帮助大家。
不管怎样,愿意跟的人可以跟着做。我学到的一点是,如果人们在构思 MVP 时遇到困难,问题通常出在从愿景到 MVP 的演绎推理链条的上游。通常人们只是不太清楚”我到底想学到什么”,而最常见的原因是人们不愿意承认自己不知道。作为创业者,说”我们想通过测试来发现客户想要什么”,这真的很难。
我举一个自己的例子。LTSE,我说过,我们花了 10 年才拿到批准并把事情推进起来。这是一个非常难做的产品。其中一个早期迭代版本没有成功,当时我们是在合作伙伴的基础设施上构建的,产品形态跟现在完全不同。我不想深入所有法律细节,那些东西比较枯燥。总之,我们花了整整六个月。我们需要花时间——筹集资金、组建团队、做各种事情,把所有游说者和律师都安排到位,然后我们还得去签这个合作协议。合作协议本身就花了六个月才搞定,然后我们还得去 SEC(美国证券交易委员会)获得批准。所以这是一个非常复杂的多阶段流程。大概花了——我想我刚才描述的就是我生命中整整两年的事情。
而整个过程当中,我一直在跟团队说,我们做这一切的目标是学习——到底是什么才能让客户愿意在我们的证券交易所上市。我们的目标是学习。人们会说,但我以为我们的目标是把合同签完。我以为我们的目标是——我就说,是的,但我们做那些事的原因是为了学习这个东西到底能不能行得通。我记得——我永远忘不了——在合同谈判期间,有人说,我们得多花一周来处理这个条款,我就说:“我们没有多余的一周。“他们说,但这个东西已经花了两年了,一周算什么?我说:“不行,每一天都太珍贵了,我们必须尽快获取学习。“我一直在催促所有相关人员。
从失败中学习
Eric Ries: 我们必须更快、更快、更快。你在这个点上必须让步,人家把我们吃得死死的。在谈判中我们简直被碾压。我是最弱最窝囊的那个人,满脑子只想赶紧融到资,为了速度什么条件都能接受。总之,长话短说,我们一路走到了最后。我们拿到了 SEC 工作人员的批准,然后我们遭到了伏击。总之,SEC 撤回了对我们的批准,整个事情就死了。合作伙伴那边,我们不得不解除合作协议,一切都崩了。我当时以为公司要完蛋了,那真是一段极其痛苦的日子。那是我人生中经历过的最糟糕的事情之一。我整个人都崩溃了。现在回想起来,我可以笑谈此事,因为结果当然它是发生在我身上最糟糕的事,但实际上也是发生在我身上最好的事——很多创业的事情都是这样。
因为正是从那之后,我们才得以 pivot 到最终获得批准的那个方向。我讲这个故事的原因是,我们为那份合作协议所做的一切工作,将近两年的时间,最后都白费了,因为合作协议最终破裂了。所以如果我们当时多争取到几个有利条款,利润率会更高,如果我们从合作伙伴那里多榨取一些费用,或者我们承担更少的责任——人们总是说,我们要为这个承担责任、为那个承担责任。我回想所有那些东西,那份文件有几百页几百页长。但这一切都无关紧要。最终合作根本没有完成,没有客户,什么都没有发生。
失败经历的价值
而这其中有意思的、也是我讲这个故事的原因是,跟我一起经历过那件事的人,在做下一件事的时候速度快了不知道多少。真的非常惊人,因为他们会说:“现在我明白你说的意思了。“大多数人过着非常被呵护的生活,在商业生涯中并没有真正经历过这种失败。显然在个人生活中,他们会有各种各样的创伤。你在一家烂公司工作,有一段糟糕的经历,但他们没有经历过这种特定的事情——在这件事里你是英雄之旅中的英雄,你是故事的主角。而这个事情不只是一个挫折,而是你直接就死了。
它真的失败了。这在心理上造成的伤害如此之深,以至于即使是经历过这件事的人,心理防御机制也会告诉自己那没有发生过。你比任何人都清楚,有多少创业创始人坚称自己从未失败过,从未 pivot 过。一切都和他们最初设想的一模一样。但说真的,我见过足够多的创业公司,我知道那种说法有多少时候根本就不是真的,但我们需要给自己讲那个故事。但不管怎样,科学方法的第一课是:如果你不能失败,你就不能学习。
所以如果你把自己置于一个在心理上无法承认某件事没成功的位置,你就无法学习。而这才是 MVP 的本质。就是——不管那个会摧毁你公司的灾难性事件是什么,我们只是想尽快发现它,这样还有时间做点什么来应对。我们不是直接死掉然后继续往前走。
工艺与 MVP 的平衡
Lenny: 我现在实际上看到了很多这样的情况。我投资了一批创业公司,其中很多现在正在失败。他们做出了如此精美的产品,如此出色的产品,花了几年时间精雕细琢,做到极致。然后这一切都白费了。最近播客里其实有一个反复出现的主题,就是工艺和设计的重要性。我想听听你的看法——什么时候应该深度投入产品的工艺和体验,什么时候应该采用最小可行产品的路径?你是从极简开始,然后再投入,还是其他方式?
Eric Ries: 对,嗯,我觉得这是一个很好的问题。人们用”工艺”这个词来指代太多不同的东西了。我个人在简洁中发现巨大的美感。所以我宁愿人们把一件事做到极致,也不愿他们把二十件事做得半吊子。你用过多少产品,你会想,天哪,这里有 92 个按钮,我其实只用过其中一个。你能不能把那一个按钮的体验做得特别好?我宁愿牺牲其他所有东西。我实际上参与过微波炉和冰箱这类带物理按钮的产品。即使是你花 60 美元在沃尔玛买的那种破厨房微波炉,如果你没见过那些东西的制造过程——把每一个按钮接线接好的背后,投入的工艺量,实际上如果你开始想,哇,那真是大量的人类精力被浪费了。
所以工艺的那些要素我认为是至关重要的。首先,工艺的一部分是完美。完美是质量的一个属性。而我们必须回到最初的问题:为了知道什么是完美,你必须了解客户的信念。你必须知道——你必须为某个目的、某个特定的人做到完美。有些人会说:“好吧,我就为自己而建,这样就解决了问题。“这完全没问题,如果那是真的,如果那个人是你,放手去做吧。你可能只有一个客户的市场,但你会很开心。那就这样吧。确实有这样的创业公司——我不在乎我的市场有多大,我就为所有像我一样的人建造产品,如果我有一个庞大的社群,那我就会非常成功。如果不是,那我就做一个做自己的事情的艺术家。我完全尊重这种选择。
这就像目的和方法完全一致。我认为人们陷入麻烦的地方是,当他们想把工艺当作逃避反馈的借口。所以我告诉人们的是,首先,工艺有一个层面就是——不管你做什么,你怎么去做?不管你在构建一个 MVP 还是一个功能完备的产品,我是受过训练的软件工程师。我的代码有特定的风格,我写软件有一种特定的工艺,因为我认为那是最好的方式,能产生最好的结果。我永远不会在这方面妥协。以追求速度为名在这方面妥协是愚蠢的。写烂代码,你以后花在救火上的时间比写代码的时间还多。
所以一定程度的工艺就是为了保持你的选择空间。实际上这非常有利于速度——使用正确的基础构件,有正确的抽象层,用正确的方式使用正确的工具,团队里有高素质的人——所有这些让你更快而不是更慢的东西。而精益创业的一个很大部分就是帮助团队进入那个最优配置,通过充当一个速度调节器来实现。因为存在最优速度这样的概念,如果大家感兴趣的话我们可以展开讲。
但工艺的另一个层面就是纯粹作为逃避反馈的借口——按照我想要的方式把东西做得漂亮,因为我就想要那样。即便在那里,当我遇到那样的人,有人就说:“抱歉,我的艺术理念就是这样,它必须是这样。“那我就说:“太好了。“那这就不是你愿意测试的关键信念假设之一了。那就这样吧。不要假装去测试它,直接说——我们不测试,我们就是要这样做,因为这就是我们的做事方式,这没问题。而且我认为当一家公司有一个核心价值观——我宁愿倒闭也不愿违背我的核心价值观——我也同样尊重这一点。
工艺与实验并不矛盾
Eric Ries: 但对我来说,这意味着对工艺、对设计、对完美的尊重,反而增加了对实验和测试以及其他所有方面的需求,因为我们已经在自己真正在乎的事情上投入了巨大的精力和心血。那其他那些东西呢?我们还必须搞那个花哨的媒体宣传吗?我们还必须把创始人的脸贴在广告牌上竖满全城吗?那些也都是你工艺的一部分吗?我只是觉得人们把这当作一个万能借口,用来做各种各样的事情。
但我认为,工艺向你的客户传递了一个信号——关于你真正在乎什么、你会用心对待什么。它是向客户做出承诺的一部分,而且是客户会相信的那种承诺。如果你的论题就是这个,那么做一个缺乏工艺的 MVP 根本无法验证这个论题。所以我觉得人们在这里想制造一个并不真正存在的二元对立。当人们有那种真实的追求时,我完全支持。但我也遇到很多人,他们只是在装。他们没有品味,没有设计理念,他们没有……那么如果你没有这些东西,你从哪里去获得呢?好消息是,实验会帮你到达那里——如果你让它发挥作用的话——因为你最终会学到那些思维习惯,学到那些有效的、合理的模式。
Lenny: 我很喜欢这个建议:作为创始人,去建立你想要的公司。既然要创业,为什么要创建一个你自己都不兴奋的公司呢?如果最终失败了,那是你的责任,那确实是个大错,但至少你试过了,你不是在建立一个自己不想待的公司。然后作为产品负责人,我不知道你有多大的控制权去按自己的方式打造产品,但至少要尽力而为。
宁可被解雇也要坚持做对的事
Eric Ries: 我总是告诉人们,宁可被解雇也要按你认为正确的方式去做。没有什么比”因为坚守某个原则而被解雇”更能加速你的职业发展了。那些被你吸引过来的人会比你开心得多。你会开心得多,因为你身处一个欣赏这一点的环境。这对我来说就是一个巨大的职业加速器。并不是说我真的被解雇了,而是我愿意为我的信念被解雇。我记得有一次去参加董事会,我确信自己会被解雇。我就想,“好吧,我就是不相信用传统方式做产品。“那时还是在精益创业之前,所以我甚至没有一套语言来描述它,但我有一种强烈的信念——我的做法是正确的方式。
我就想,好吧,我会成为因为这件事被解雇的那个人,最终会有人想和我合作,因为那正是他们想要的。如果没人想要,那我就去别的行业找工作。我不想待在一个不……的行业里。这就是我想要的状态。我觉得人们没有充分意识到……人们把创业失败当作最糟糕的事情。天哪,那连前十都排不上。它确实糟糕。我经历过,很痛苦。好吧,确实很糟糕。但更糟的是待在一个不会死的公司里——一个你讨厌但离不开的僵尸企业。我见过太多这样的人了。创业者中正在经历一场心理健康危机,而这还没有被足够多地讨论。人们开始谈论心理健康风险的负面影响。显然,作为创始人的压力是巨大的,但看看那些所谓”成功”的人身上发生了什么。当你建立了一家公司,把它卖掉,它变成了你觉得令人厌恶的东西——也许你变富了,但这并不好。前几天一个朋友告诉我,他的一个朋友卖掉了公司,退出价大概是三亿美元什么的,赚了很多钱,但没过多久就自杀了。人们都关注 Tony Hsieh 的故事。我认识 Tony,那是一场可怕的悲剧。到底是怎么回事?所以我觉得,创业失败并不是最糟糕的事情。建立一个你讨厌的公司,它变成了世界上的一股扭曲的力量,你不得不假装自己跟它无关,或者你对它所做的事情感到共谋——那要糟糕得多。我真心希望更多的创始人能在早期就把这件事看得更严肃一些。
僵尸企业的困境
Lenny: 这一点太重要了,也是我经常看到的——创始人创办的公司还在喘气,但几乎不可能成功,他们被卡住了。他们不想关掉它,不想把钱退回去。对于那些感觉被困在僵尸企业里的创始人,你会给出什么建议?
Eric Ries: 这真的很难。当然,如果它注定失败,你应该关掉它。但每个人都说,“可是黎明前不是最黑暗的吗?“是啊,有时候确实如此。但你知道还有什么也很黑暗吗?死的时候。所以一切归根结底——我以前在精益创业的早期演讲中有一个最让人不安的笑点。那时候人们经常谈论 Steve Jobs 和他的现实扭曲力场。我就会说,好吧,你身处现实扭曲力场中,你正处在曲棍球棍曲线的平直部分。你怎么知道商业计划是对的?我真的亲身经历过一次。有一份商业计划书,上面写着:发布六个月后,我们几乎没有客户,几乎没有收入。然后曲棍球棍式增长将由此开始。我记得在第六个月时心想,我们按时、按预算、按计划推进。
一切都在按计划进行,曲棍球棍式增长即将开始。我盯着手表看。然后我们就一直待在那个平直部分,延伸到无尽的天际。人们会笑,我就说,好吧,现在如果你跟随一个创业领袖,或者你自己身处现实扭曲力场中,你正处在曲棍球棍的平直部分,你怎么知道它最终会扬起来,还是你永远就那样……然后人们就会”哈哈哈哈”地笑。但那真的是一个非常可怕的处境。我们每个人都会有这样的自我怀疑时刻,有时候正确的答案确实是克服怀疑,全力以赴。这其实是创业体系的精妙之处。人们批评风险投资融资,批评硅谷的各种问题——当然,我们自己制造了很多问题,所以值得批评的东西确实很多。
但这个体系的一个精妙之处在于:最终,你的钱会用完。你被迫宣布破产,这其实是这个系统一个非常积极的特征。很多创始人就是做不到这一点。但我想说,我试着让人们做的一件事是——真正花一些时间反省,问自己:你真的还想待在这里吗?如果你能挥一下魔杖,现在立刻开始一家新公司,你会选择这家吗?如果不是,现在已经没有契约奴役了。没关系的。可以继续前行。然后问题就变成了一个战术性的对话——关掉它,把钱退回去,或者做一个真正彻底的 pivot 转向新方向。
关于关闭公司的心理挣扎
这很大程度上取决于具体情况,但我觉得如果我们假装这很容易,那是对创始人的不负责任。我的意思是,这个过程令人痛苦,非常艰难。不想把话说得太玄乎,但这确实直指自我认同这个核心问题。对很多人来说,他们就是他们的公司,所以公司要死了,他们也得死。感觉就像自己要死了一样。我认为这也是心理健康问题的一部分。所以我希望随着人们的成长、成熟,以及灵性修习的深入,他们能够学会与公司解绑,不再觉得一切都不容侵犯地指向自身,这样公司可以死,但人不会。然后他们可以把学到的经验带到下一家公司,或者下一个 pivot,或者下一个迭代中去。
或者有时候正确的做法就是退位让贤,让别的 CEO 来接手。有很多种处理方式。但就像任何 pivot 的处境一样,最难的部分始终是直面事实本身。然后你自己做出决定——现在的情况到底是什么?它奏效了吗?我无法告诉你有多少人就是无法接受事情没有奏效这个事实。你会收到投资者的更新报告。你收到过多少这样的投资者更新——上面写着我们势如破竹,一切都很顺利,然后紧接着下一篇更新就是,我们关门了,把钱退回去。你会想,你上一篇更新明明说势如破竹啊,完全没有任何出问题的迹象。当然了,没有迹象,因为我们就是在撒谎。所以别做那种人。
Pivot 的频率与数据
Lenny: 这其实是一个非常好的过渡,正好引向我想聊的下一个话题——pivot。我对消费端创业公司和 B2B 创业公司做了不少研究,我看到的数字是——我也很好奇你看到的数字是否相似——大约五分之一的消费端创业公司最终 pivot 到了完全不同的产品,而 B2B 公司中几乎有一半,大概 40%,最终 pivot 到了完全不同的产品。从你的经验来看,这个数字听起来对吗?
Eric Ries: 对,听起来是对的。很有意思的是,我以前会给创始人做一个练习,用两种不同的方式给他们讲 IMVU 的故事。第一种方式是我们把一切都搞错了,听起来就像我们是彻头彻尾的傻瓜,不得不 pivot 了那么多次。然后我再把同一个故事讲一遍,但这次是我们把一切都做对了,我们是天才,发生的一切都证明我们是对的。然后人们不相信我。我说:“这两个故事都是真的。“选择强调实际发生的事情中的哪些元素,把重心放在哪里——你可以讲出不同的故事。所以没错,我认为有一部分人知道自己 pivot 了,还有一部分人不承认,但他们实际上也 pivot 了。而且从外部真的很难判断。
我的意思是,这一点我在与足够多的创业公司合作后深有体会——我知道外部版本的故事,也知道内部版本的故事,它们真的非常不同。Paul Graham 曾经把这当作 Y Combinator 的一大特色来夸耀:他们邀请创始人来参加晚宴,那些创始人会告诉你真实发生的故事,那是你在其他任何地方都得不到的关键信息,只有在 Y Combinator 才能获得。我就想,等等,他是在说所有人上台讲的故事都是在撒谎吗?我不确定这有什么好夸耀的。你这是在说你的嘉宾都是骗子。后来我自己也主持了很多这类活动,我就明白了——哦,确实是,人们上台就是在撒谎。所以我认为部分问题在于,要搞清楚真正发生了什么,真的非常困难。但还有一个我觉得很有意思的层面——pivot 的定义是在不改变愿景的前提下改变战略。所以我们有这样一个观念:创始人有愿景,然后他们想办法让愿景实现,过程中找到了不同的路径,但愿景始终不变。但事实比这要复杂一些,因为创始人的愿景本身也在演变。创始人在创建公司的过程中才发现自己的愿景到底是什么。所以,再次声明不想说得太玄乎,但我确实相信,创建一家创业公司在很大程度上是一个自我发现的过程。你实际上在发现你自己所相信的东西中什么是真实的。部分原因在于,当我们幻想愿景时,我们永远不需要做任何取舍——产品无限受欢迎,无限便宜,一切都很美好。
愿景在取舍中成型
就好像没有任何取舍需要做,而现实生活却逼迫我们做出取舍。然后你发现,“哦,我必须做选择?“我是要做一个只面向富裕人群的高端产品,还是要做一个面向这个行业、这个领域、这个群体或那个群体的产品?这可能很痛苦,但创始人也在这个过程中逐渐学会谈论他们真正关心的事情。
去 YouTube 上看看吧。如果你从来没看过 Zuck 在斯坦福的沙发上接受采访、谈论 Facebook 的那段视频,那真的值得一看。不过就是一个校园新闻记者恰好在这个人生命中的关键时刻采访了他,那时候他还根本不知道自己在说什么。而现在的他雄心勃勃。现在 Facebook 的愿景是非常宏大的,人们也极力强调他从一开始就有这样的愿景。他自己也到处这么讲,这也许是事实,但纪录片证据是存在的。关于这一点,我再讲一个我自己的故事。
在我的一家公司里,我们很早就搬了办公室,因为我们起步的那个办公室又小又破,我们搬到了一个更大的空间。我记得当时办公室里只有我们五个人,我们坐下来开了一个会,在那个会上我们制定了现在所说的 MVP,以及公司更长远的愿景。我们有一块白板,所有人都在争论哪些功能应该纳入、哪些应该排除,或者白板上应该写什么。我记得我写下的那些东西——我对那个场景有一种身体的记忆——我写下那些公司成功所必需的东西,我是对的,我的联合创始人们在跟我争辩,因为他们没有正确的视角,而他们是错的,我当时想,还好我在。
这是我讲这个故事时常提到的一点。总之,在那之后不久我们就搬了办公室,那块白板就丢了。多年以后我们又搬了一次办公室,有人在清理一个旧储物间时,那块白板恰好被塞在了那个储物间里。有人把它拽出来,说:“这是什么?“我一看,“哦,那是我们以前挂在主会议室里的白板,上面还留着那次会议的笔记。“我想,这太酷了,历史性时刻,我要拍张照。然后我就想,等等,这不对劲。那是我的笔迹,分明是我自己的笔迹,但我写的很多东西简直是糟糕透顶的主意。我心想,那不是我当时的想法,我当时站在另一边的,那是我联合创始人的馊主意,那不是我的主意。
记忆的不可靠性
Eric Ries: 我于是意识到,不仅仅是在创建这家公司的过程中我学到了关于战略的东西,我对自己所做之事愿景的理解本身就已经不同了。但真正关键的是,直到今天,我在心理上完全没有”自己改变过想法”的记忆。我仍然保留着当初在白板上写下那些东西的原始记忆。一开始我真以为是有人在跟我恶作剧,因为我亲眼看到的白板上的内容,跟我实际记忆中的完全矛盾。我心想,到底哪个是对的?当然,我不得不承认,物理证据必须胜出。
但我觉得同样重要的是要理解这种情况有多普遍。再说回来,这就是为什么科学如此重要。为什么我们要把科学中那些最基础的东西都写下来?我们把假设写下来,把我们的东西提交检验,我们之所以这样做,是因为人类的心理充满了缺陷。那些正在跟生成式 AI 聊天机器人较劲的人已经开始看到了——那些东西不过是人类心理理论的一面镜子。我们开始在一个异质生物身上看到,哇,原来我们的智能有这么多奇怪的行为、缺陷和盲区。我们当然希望自己的机器人完美无缺,但首先照照镜子吧,认识到你自己就带着所有这些认知偏差,你必须建立一套机制来克服它们,否则它们会毁了你。
Lenny: 这个故事太不可思议了。这让我想到,正如你所说的,你听到的那些故事不仅常常不是准确的或真实的,它们是被创造出来的某种神话,而且你自己也会忘记到底发生了什么,转而讲述一个实际上并没有发生过的故事。太疯狂了。
关于 pivot 的真正含义
Lenny: 你谈到过 pivot。我之前一直不知道 pivot 这个词的含义是你保持愿景不变。我这里恰好有一份公司名单,都是 B2B 公司的案例,它们都做过 pivot。但在这些案例中情况非常不同。让我快速过一遍,我很好奇,如果它们没有保持愿景一致,你会怎么称呼那种情况?Segment 最初是一个大学课堂授课工具。Loom,刚以高价卖给了 Atlassian 的那个,最初是一个帮助企业招聘领域专家的市场。Slack 著名地源于一款叫 Glitch 的游戏。Box 据说最初是 Facebook 上用来存放照片和内容的一个盒子。Retool 最初是英国版的 Venmo。
Eric Ries: Retool 那个我不知道。
Lenny: 我是从 Twitter 上知道的,一位 YC 的合伙人回复了我,跟我分享了这个故事,因为我之前也不知道。我觉得这是一个很好的例子——从来没有人告诉过你这个事实。
Eric Ries: 对,这正是故事中很重要的一部分。
Lenny: 好像是一个很快的 pivot。我觉得他们到了 YC 之后就说”不,我们做点别的吧。”
Eric Ries: 这就是为什么从外部评估一家公司如此困难,因为往往存在一条内在的连贯线索,从内部完全可以理解为什么从这件事转向了那件事。而且往往是某种不寻常的东西。比如,会是这样的情形——我们做游戏的时候,必须先做这个东西,因为我们的价值观告诉我们,没有这个东西就没法做游戏。结果这个东西比游戏本身更有价值。然后你就想,“那我们为什么还在做游戏?这才是我们真正在乎的东西。“所以人们会说,“我以为你的愿景是做游戏。“而答案是,“也许是吧,但他们为什么先做了那个东西?其他人做游戏不需要先做那个东西。“这就是为什么我说这是一个自我发现的过程。在很多这样的故事中,愿景更多是被发现的,而不是被选择的。从外部来看,这一切可能显得非常非常奇怪。还有一部分原因是,有些人确实需要第一个项目的失败,才能真正认识到第二个项目的价值所在。
Lenny: 我觉得 Segment 是一个很好的例子。他们尝试了很多东西,然后在 Hacker News 上发了个看似随意的东西,结果一下就爆了。
Eric Ries: 有时候爆的程度让你觉得,哇,太惊人了。但更多时候,新东西可能只获得一百个注册用户。如果你一开始就做这个东西,你会觉得”一百个注册,谁在乎”。但如果你连续做了二十个东西都是零注册,天哪,你就会觉得”老天,居然有人喜欢这个东西”。有时候第一次就中,但有时候不是。
愿景是发现的,不是选择的
再说一次,不是要一直拿 Facebook 举例子,但那个故事太有名了,大家都知道。在电影里有那个著名的场景——他们住在 Palo Alto 的房子里,遇到了 Sean Parker,然后发生了那些事。据说那是真的,他们确实住在那栋房子里,也确实在做 Facebook……但故事中被遗漏的一点是,那时候 Facebook 已经启动了,已经开始取得成功,他们是在学期之间的暑假。Dustin Moskovitz、Mark Zuckerberg、Adam D’Angelo,还有谁我忘了,一起住在这栋房子里。
有人跟我讲过那时他们实际上在做什么。不是 Facebook。Zuck 和 D’Angelo 已经从 Facebook 那个项目上移开了注意力,觉得它不够成功,他们觉得……其中一个在做的是一个系统,可以用摄像头的手势控制来操作电脑,这样就不会得腕管综合征。他们明明坐在 Facebook 上,而它却……我应该问问 Dustin 这是不是真的。我相信 Dustin Moskovitz 是自学的编程,目的是跟上那些要求 Facebook 开放到新学校的用户需求,因为病毒式传播太疯狂了。我当时就想,哇,你甚至可以在正在构建 Facebook的时候都不自知。人们会说,“如果我告诉你这个故事但不告诉你创始人的名字,你会觉得这人毫无远见。“我会说,“不,那是 Mark Zuckerberg。“从来没有人指责过他缺乏远见,但这确实是事实。
所以我觉得,关于这些故事的复杂性,我们得承认,实际情况比你意识到的要复杂得多。这些人并不完美。而 pivot(转向)本质上关乎的是学习,对吧?pivot 最初的概念来自篮球——一只脚固定不动,另一只脚移动。固定在你已经学到的东西上,同时保持关联。所以如果真的完全关停、从头再来,跟之前毫无交集,我会说那不叫 pivot。但这种情况其实相当罕见。
通常,第一个项目中有某个关键的洞察或元素会被带进第二个项目中去。如果你当时在场,如果你了解整个故事,你会说”哦,哇,事后看来这完全说得通。“再说一次,我觉得这和我们整个关于远见者的认知是相悖的。我们以为所谓有远见就是……你知道所有答案,像一个能预见未来的先知。但实际情况并非如此。它更关乎你有勇气按照你认为世界应该有的样子去看待世界,然后与这种愿景共舞,无论它将你带向何方。我认为这其中勇气的成分远远大于智力或其他任何东西。
Lenny: Eric Ries,你肚子里真是装满了精彩的故事。我从来不知道 Facebook 还有这段故事。想到那些可能正在考虑 pivot 的创始人……现在有很多创业公司——不是在挣扎,但确实在挣扎。
Eric Ries: 2021 年份的这批确实日子不好过。
Lenny: 没错。我很好奇,对于那些正在犹豫是否该 pivot 的创始人,你有什么建议?该不该放弃?该不该坚持?有什么战术层面的实操建议?
Eric Ries: 首先第一条,如果你在问自己要不要 pivot,那你大概已经知道答案了。这是首先要直说的。有时候人们给我打电话说”我想知道我有没有达到 product-market fit。“你有时间给我打电话问这个问题,这个事实本身就已经告诉我答案了。如果你真的达到了 product-market fit,你根本没时间做这些。当一切奏效的时候,没有时间自省,没有时间……如果你从来没有经历过一家公司达到 product-market fit 的过程——那是一种狂飙,一种极其疯狂的感觉。那种旋风既令人惊叹,同时又令人恐惧,根本没有时间顾及其他任何事情。如果你不确定,那你就是确定了。所以第一件事就是——你能不能对自己承认?你能不能做到?
给自己一个固定的时间窗口
如果你承认情况不妙,那我们接下来怎么做?在这种情况下我真正要求人们做的是……给自己一个固定的时间段,采取一些果断的行动,看看感觉是否会好一些。如果你还无法承认它不奏效,那好,我们能不能拿出六周时间试试让它奏效?我们有这样一个假设,我们打算测试它,我们要……与其去做未来需要做的其他 29 件事,我们能不能把所有时间百分之百地投入到当前唯一重要的那一件事上?在一段固定的时间里全力以赴,看看能不能……能不能推动那个指标?
因为当创业公司不奏效的时候,一个真实的情况是它们陷入了收益递减的困境,你就是……你可以自己验证。你尝试的一切都不管用,什么都不管用。这就像 Thomas Kuhn 和他的科学范式理论。他讲过的一个我最喜欢的观点是:当旧范式被新范式取代时,通常只是因为老一代研究者退出了,新一代研究者取代了他们。很多研究者会问,为什么会这样?并不是新一代人神奇地比老一代更聪明,而是旧范式已经到了一个阶段——再也没有能产出成果的实验可做了。而年轻、有抱负的人想要做有成效的实验。新范式给了他们一个方向——每个实验都能揭示令人惊叹的东西——于是他们自然地被吸引过去。这其实不是一个正式的抛弃旧范式、转向新范式的过程。
找到一个实验有成效的新范式
这里的情况也是如此。大多数人无法有意识地放手,他们做不到。那好,给自己找一个实验能出成果的新范式。我见过太多创业公司了,他们跟我说”数据在往右上方走。“我说”哦,是吗?“他们说”对啊,一开始是 10%,然后 10.5%,然后 10.6%,然后 10.65%,然后 10.657%。“你心想”各位,是的,是在往右上方走,但那不是你需要达到的水平。“所以如果必要的话,给当前的方向一个固定的时间。一旦你说”好吧,这个方向不奏效了。“我们有没有时间尝试新的东西?你能不能给自己六周时间去试新方向?如果你是一家大公司,你会说好吧,我们处于……也许不能让整个公司来做这件事,但能不能派一个团队?某个人,至少一个全职的人,不是兼职的委员会。有人全职投入,在截止日期前拼了命地让新方向跑通。
如果你是一个只有三个人的小团队,能不能把整个项目暂停一下,给自己一个窗口?哪怕是一个周末,也比什么都不做强。你实际能承受多少?六周当然最好,但如果只有一个周末,也没问题。你想做的那个新方向是什么?既然大家已经同意在一段固定时间内做这件事,那就让每个人轮流说:“如果我能重新创办这家公司,我希望我们做的是——”
没有人愿意先开口,但因为每个人都要说,所以你只好说”好吧,我希望我们做的是另一件事。""为什么?“有时候第一个人说完后,所有人都说”我这几个月一直想说同样的话。“然后发现”等等,我们所有人心里其实都想做另一件事,只是都害怕让其他人失望。“这就是”这其实是好消息”……”嗯,太好了。“不要为你浪费的那十八个月感到惋惜,要为你们终于在新方向上达成共识而感到兴奋。
总之,每个人说出自己想做的事。看看有没有哪一个特别突出,就做那个。如果大家一致同意某个方向,就做那个。如果有三件事你们同样兴奋,那就按顺序来。每人一个月。我们还剩三个月的钱,好,第一个月做第一件事,第二个月做第二件……看看有没有能立住脚的。有时候人们做完这些之后会说”我只是筋疲力尽了,我只想死。“如果每个人都表示”我宁愿把钱退回去,没关系的,别撑了。让公司死掉也没关系。“当你融了太多钱,股权结构表一团糟,而且你……我觉得对于我们 2021 年份的兄弟姐妹们来说,有时候你只是需要以一个正确的资本结构重新开始,但这不现实。那就做出你真正需要做的结构性调整,给自己一个机会去做你在乎的事情。如果你连这个都没有,我们在这里干什么?生命太宝贵了。
Lenny: 这条建议太棒了。非常实操,非常真实。回到你刚才问的那个问题——我们真正想做什么?我想今天很多人会回答跟 AI 相关的产品。这也正好引出我的下一个问题——你怎么看 AI 正在改变精益创业方法论、产品开发、团队运作的方式?在我们讨论的所有这些话题中,你观察到了哪些趋势和变化?
AI 时代与精益创业
Eric Ries: 我的意思是,我们正在进入人类、技术和文明的一个非常有趣的阶段。要回答这些问题的正确答案,需要的是没有人拥有的经验性知识。这正是它令人兴奋的地方。当然,也有人觉得这很可怕。到目前为止我看到的是,首先,人们害怕 AI。但他们并不是真的害怕 AI,他们害怕的是拥有 AI 的公司。AI 本身不会做任何事情,它必须由人类来体现和部署,而人类的 alignment(对齐)做得一塌糊涂。所以大家都在谈论 AI alignment。如果那些做对齐的公司在对其人类智能进行对齐方面做得更好的话,我对 AI alignment 会更乐观一些。我一时想不起来这是谁提出的定律。总是有这些以人名命名的定律。那个”你的软件会反映出生产它的组织的结构”的观点,是某个人的定律。也许是 Conway’s Law(康威定律)。
Lenny: 那我们就叫它 Ries 定律吧。
Eric Ries: 别,求你了。我忘了是谁的定律了。不管怎样。这一点在 AI 领域将被放到极致——我们正在训练机器智能,正在向它传递价值观,而那些价值观将是我们组织的价值观。科技行业在创建治理和对齐结构方面严重投入不足和被忽视,没有让这些价值观在组织中被明确体现出来,这就是为什么我们这么多公司变成了令人失望的存在,为什么那么多人觉得自己亲手参与建造的公司背叛了自己。这其实挺可悲的。所以 AI 会迫使我们认真对待这些问题,无论如何都会。
你问它对精益创业有什么影响。归根结底,AI 是一种管理技术。它做的事情就是管理智能和其他智能。看看人们在努力让 AI agent(智能体)工作时的挣扎就知道了。我自己也玩过很多这类工具,有一个现象一直让我觉得很有意思——你给 agent 编程,它会说什么好的。你给它一个任务,比如去统治世界之类的,或者开一个非常成功的冰淇淋摊。你给它任何你想要的任务,它都会说没问题。这些 AI 被编程为永远不说”不”,所以不管你说什么,它都说没问题,好的。
第一步,写一个统治世界的商业计划,然后它会生成一个新的 agent 和一个新的 AI。它会说,好的,你的任务是去写这个商业计划。有时候它确实会去试着写商业计划,但更多时候它会生成……子任务会说,哦,我需要做的第一件事是生成一个知道怎么统治世界的 AI agent,让它来写商业计划。每个 agent 都会把皮球踢给下一个 agent。它会进入无限倒退。比如,我们需要先研究这个,然后回来的结果是,我们研究发现还需要再研究十件事,而那十件事每件又需要再研究十件事。太多出问题的方式了。
任何当过中层管理者的人看到这些对话记录都会说”可不是嘛,老兄。“有一次我在做一个 AI agent,想让它帮我写一个 30 天的课程——精益创业的教学课程转录稿。有人请我做这件事,我就想,太好了,让 AI 替我做吧。结果特别有意思。它说好的,没问题。第一天,什么是精益创业?怎么运作的?第二天,什么是 MVP(最小可行产品),怎么运作的?然后省略号,以此类推继续 30 天,你就有了一个 30 天的课程。我心想,是啊,因为我知道这些,但你能不能把剩下 30 天的内容填出来?它说没问题,然后生成了一个新的子 agent。它说,以这两天的内容为例,请填出其余 30 天。那个子 agent 说明白了。然后它说,如何在 30 天内完成一个任务。嗯,首先看看这两个任务,然后为剩下 28 天创建类似的内容,然后返回结果。我心想,不不不,我就是要你来做这件事啊。如果你看过 Monty Python 那个”不,不能看守他,如果他是个守卫”的小品,就跟那个一样。我们完全在鸡同鸭讲。这真的很有意思——随着我们对 AI 的理解加深,我们将能够发展出不同的……它真的会极大地改变管理,因为它相当大程度上改变了个人的管理幅度。
管理中最困难的问题之一是,为了触及很多事物,我们需要大量的层级结构,因为每个人有这么多下属,每个下属又有这么多下属,一直扩展下去。这就是为了让大量信息被汇总所必需的结构。如果你有 10 万名员工,他们在做 10 万件事,CEO 需要知道发生了什么,这真的很难。所以需要大量的人力劳动仅仅用来汇总和理解正在发生的事情。而 AI 非常擅长总结,这实际上是它的主要优势之一。在不远的将来,一个人直接说”给我总结一下我的公司正在做的所有事情”将变得非常容易。你不再需要大量的中层管理者来做这件事,它直接就能知道。我们甚至都不知道我们想建立什么样的组织——一个让这些能力对每个人都可用的组织。
最后一点是,人们非常关注这将把我们引向乌托邦还是反乌托邦。大多数人对将要发生的事情有一个非常坚定的信念,因此也对现在需要做什么来阻止坏事发生有一个坚定信念。所以有人说”我们必须停下来,暂停,炸掉数据中心,因为会发生这种坏事。“有人说”我们必须让政府介入来阻止对这件事的研究,我们必须禁止开源。“对于每一件有人想做的事,都有另一个人在说”那件事恰恰是通向反乌托邦的道路,“对吧?禁止开源才是把我们推向一个由公司控制的独裁体制,比你所害怕的那个东西还要糟糕。
我最近经常收到行业里一些人的提问,他们正在经历存在主义危机,不知道在 AI 面前自己的人生该怎么办。正如你所说,所有创业失败的人都说我想去做 AI。好吧。对于每个觉得”这边的草更绿”的人……那些真正去做的人,辞掉工作去搞 AI,他们首先面对的问题是:我应该做什么才能产生积极的影响?实际上真的很难找到一些……我提出的每一个想做的事,都有人告诉我那其实是在做恶,反过来也是一样。精益创业很大一部分……对我来说,就是关于学习如何应对生活中的不确定性,对吧?所以从商业角度你怎么应对?你怎么应对高不确定性的情况?它可以是科学的,可以是实验性的,可以是……就是所有这些关于应对不确定性的不同价值观。
在高不确定性下如何行动
所以人们问我的问题基本上是:面对高度不确定性,你如何采取合乎伦理的行动?我一直在使用的一个框架,我发现非常有帮助,也希望其他人能从中获益——AI 使可能的未来范围极其宽广。而预测它需要对大量实证问题有非常深刻的理解,据我所知,目前没有人知道这些问题的答案。既然未来不可知……你的专业知识不够。据我所知,没有人真正知道。最顶尖的专家正在积极做研究,试图发现能帮助我们做出预测的东西。对此你无能为力,只能等别人去找到那个答案。有意义的做法是,今天就选择在广泛多样的未来情境中都合乎伦理的行动,然后去做。真正有趣的是,这把人们从完全的瘫痪状态中解救出来——实际上有一些非常明确的事情是我们应该做的。不管你设想的末日场景是什么,我们现在都需要更多透明度,对吧?这至关重要。我们需要更强的国家治理能力,我们需要更有能力的领导者,我们需要致力于特定价值观的公司。这改变了我们思考这些技术与我们作为个体之间关系的方式。
这也确实是我在这里工作的目标——就是如何让人们不再袖手旁观,投入战斗,去做那些真正可能有价值的事情。跟你聊着聊着我突然意识到,哦,这听起来很像精益创业里的那个观点——你在创业中所做的大部分工作在 pivot 之后都会被扔掉,所以别做那些工作。这不是什么特别独特的洞见,只不过是把应对创业不确定性的同一套方法,现在更广泛地应用开来,因为一切即将改变。
AI 在实践中的价值
Lenny: 听起来这和你现在在治理和长期股票交易所方面的工作有很多联系。我想聊聊这个。但在此之前,我很好奇,就个人而言,你是否在生活中某个方面发现 AI 工具确实有用?或者,有没有更有趣的例子,一家公司将精益创业方法与 AI 结合使用?你有没有见过什么实际有价值的 AI 工具应用案例?
Eric Ries: GPT-4 刚发布的时候,我看到一个朋友的朋友发了另一个朋友的故事。是关于一个人用它做冷启动销售外联的,就在它刚出来的时候。他下载了一个包含一万家 Shopify 店铺及其元数据的列表,我搜了一下,一百美元就能买到。他直接让 GPT 写一个 Python 脚本,为每家店生成内容——“GPT-4,请向这家店老板推销一些我的外包公司能为他们构建的东西,发挥创意。“他甚至没读那些邮件,就一封接一封地发出去了。当对方回复时,他才去看邮件里自己推销了什么,然后想办法把对方约到电话上。
它推销的内容极其有创意、很酷,真的很有意思。邮件风格契合店铺的调性,用了巧妙的双关语。他给我看截图,对方回复说”你说的某某点打动了我。“销售前景开发做得非常好。好吧。这是一个关于超高频率实验的惊人故事。五年前我们根本做不到这样的事情,而现在它将变得完全日常化。
但这也是一个关于我们所有的沟通渠道即将被 AI 生成的实验内容过度饱和的故事,不管我们愿不愿意。如果这项技术按我们预期的方式发展,很显然每家公司都会发送海量的 AI 生成邮件,以至于人们将需要 AI 驱动的邮件客户端来替他们阅读和总结邮件。人们开玩笑说,我要写三个要点,我的 AI 会把它变成一封邮件,发给你的 AI,你的 AI 再把它还原成三个要点。我们难道不能一开始就直接交换三个要点吗?我们浪费了大量算力。大量 GPU 被烧掉,只为了把这三个要点从 A 点搬到 B 点。
AI 驱动的市场与公平
我不认为这个故事像听起来那么荒唐。我实际上对 AI 驱动的市场能够远比我们现在的做法更加公平感到兴奋。因为如果你想想那个故事,问题的根本在于,有大量的人可以为店铺老板创造东西。如果所有这些人每天给地球上每个店铺老板发邮件,告诉他们自己能做什么,那将是压倒性的信息量。所以双方实际上都做不到这一点。但 AI 可以做到。
所以想象一个世界,每个人都被要求——而且也很容易做到——定义一个采购策略。我感兴趣买什么?我想解决什么问题?我想要什么信息?我的 agent 用这个过滤器替我阅读收件箱,并且完全公平地应用这个过滤器。不在乎我上周是否和某某打了高尔夫球之类的,只要提案进来是可信的,并且 AI 非常尽职地确保它不是垃圾邮件,那它就会把这个信息推送给我。但它不会给我看原始邮件,我不会被它巧妙的 Al Ries 式定位花招所迷惑。
它会给我关键信息——“嘿,你知道吗,有这个东西可以帮你省一大笔钱?“你说”哦,好的,太好了。“这比广告能做的要好得多。而且也公平得多。我们可以说,“好吧,实际上,我们希望每个可能的供应商随时发尽可能多的邮件,用你产品能做的一切来轰炸我。让你的 AI 放手去试图说服我的 AI,你的东西确实好。“那不是浪费时间,那实际上会在世界上创造更多公平,为采购和信息创造一个真正公平的竞争环境。想想记者和公关提案。你会看到很多这样的情况——我们建立了层级化的把关结构,因为那是处理必须处理的信息量以完成那件事的唯一方式。当这些限制被移除时,可能是灾难,但也可能真的很棒。
Lenny: 我喜欢你对我们讨论的每件事的乐观态度,我真的很钦佩这一点。这可能是一个很好的过渡,引向我想聊的最后一件事——就是你现在正在做的事情。你谈到你在治理方面做了很多工作,帮助公司。另外,我觉得你也在做公司文化方面的工作,帮助公司赚钱、盈利、建立真正的商业。而这显然与长期股票交易所有关。你能否谈谈你在这方面花时间做什么,以及你看到的大机会在哪里?
Eric Ries: 当然可以。我觉得这是我们正在应对的最重大、最重大、最重大的问题之一。如果拿今天创始人问我的问题和十年前相比,这些问题在人们心目中的分量比那时候重得多了。当我第一次和一位创始人合作时,我们经常做一个思维练习——“好吧,我们谈了文化,谈了公司,那好,为什么别人应该信任你?“他们会说,“什么意思?“我们讨论的是,现在公司正在构建的东西如此深入地渗透到人们的生活中,比以前强大得多。说实话,这不过是认真对待”改变世界”这套说辞而已。
那么多科技公司在融资的时候说要从根本上改变世界。而当被要求承担道德责任时,他们又变成”我不过是个小小的数据库供应商”了。你不能两头都占,老兄。你是那个随心所欲重新排列整个世界的超人吗?那你就得对结果承担道德责任。这个观点目前争论很激烈,但我觉得实际上超级……人们在自己的 rhetoric 里已经清楚地表明了自己的立场,他们只是在试图逃避那份责任。
下一代创始人的责任感
所以下一代的创始人不会……他们完全不觉得这很复杂。人们对让世界在真实的层面上变得更好有一种真正的责任感。太好了。我在做的那些公司简直就是科幻小说里走出来的。我是说,有的公司能做到从旧金山到东京一小时直达。你会觉得,这太不可思议了。真正的改变天气。真正意义上的——我们要改变地球上可居住土地的面积,永远解决饥荒和干旱。当然还有 AI,还有纳米材料和量子计算,各种不可思议的可能性。我可以继续说下去,正在发生的事情实在太酷了。
权力与信任
但它们都伴随着极其严重的后果。我之前在跟一家非常前沿的生物 AI 公司交流——想象一下有一个 LLM,想象一下 ChatGPT,但是针对生物化合物的。我们会说”AI,你能帮我合成一个具有这些属性的分子吗?只用自然语言描述就行。“然后它说”没问题。“这可能会非常了不起。想想那些治愈疾病的方案,那将是技术的惊人变革。但它也有非常明显的风险和负面后果,我们必须极其认真地对待。为什么任何人应该把这种权力交给你并信任你?人们总是说同样的话,“嗯,我是个好人。我是你的兄弟,哥们。我的出发点是好的。“这就像——“谁在乎这个啊,你可能会死,可能会被替换掉,某个投资者可能会把你扫地出门。”
我能跟你讲多少创业公司的故事,投资者非自愿地更换了管理层?根本问题在于,没有任何个人能够代表一个组织做出承诺,这是不可能的,这是一个范畴错误。你是可以被替换的。原因在于,组织实际上是活着的。它们是拥有自己灵魂和自己生命的超个体。我们制造它们的时候,是它们的自我和器官。创始人并不是创造它们,我们是孕育它们。这不一样。我们不拥有它们,它们不是奴隶。我们培育它们。这不一样。如果你希望你的组织值得信任,它所做出的承诺——如果你想通过做出公众可以相信的承诺来赢得公众的信任,你就必须把这些承诺嵌入到组织结构本身之中,这样即使你不在了,承诺也会被遵守。
人们觉得这是不可能的,但其实有很多例子。想想医学伦理准则和希波克拉底誓言。在美国,如果你想从事医疗行业,你必须有一个由医生——而不是私募基金的人——运营和管理的医疗提供机构。有诸如此类的种种安排。我们知道如何围绕某种目标来构建组织结构,那么这个目标应该是什么?
什么才是组织真正在乎的事
我跟一位前 Google 员工聊过——我不想专门针对 Google,因为他们确实在世界上做了很多好事。但他们可以说是很多科技从业者对世界感到失望的一个焦点。所以这不是 Google 独有的问题,只是我恰好跟一位前 Google 员工在聊。他在谈价值观、谈公司、谈他的失望,以及他在那里待了这么多年所经历的好的和坏的事情。我说,“好,回答我这个问题。我要问你两件不同的事,对于每件事,我想让你告诉我,你认为这件事在未来发生的概率是多少?第一件事。你认为 Google 按时提交下一份季度报告的概率是多少?“他说”百分之百。“我说”多确定?“我说”像太阳会升起一样确定,拿命来赌?“他说”像太阳会升起一样确定,百分之百。“我说”太好了,我同意。”
第二个问题。“Google 会参与某件能赚钱但代价是谋杀某人的事情,这个概率有多大?“他说”他们大概不会那么做。“我说”大概,百分之百。像太阳会升起一样确定。“他说”嗯,不是。“我说”哦,真的吗?那你认为多少?“我说,“来吧,如果他们在运营自动驾驶汽车时撞死了人然后掩盖真相呢?如果他们做了一个社交媒体产品而卷入了种族灭绝呢?Facebook 就经历过这种事。“他说”我觉得他们不会故意那么做。确定的事和不确定的事之间的区别。“我说”好,为什么会这样?“我们把这当作正常的。当然,公司会按时提交报告。难道是因为在人类的心中,对金融规则的忠诚是硬编码的,而对他人生命的关怀则不是?人们根本不在乎季度报告。
重新定义盈利的意义
我们在这些组织中建立了一个庞大的机制来确保某些事情发生。公司对自己在乎的事情非常擅长,因此我们可以从它们选择擅长什么、以及它们长期不擅长什么中,了解到一些有趣的关于它们真正在乎什么的信息。我的观点是,作为一家营利性公司意味着什么,我们搞错了。我们基本上是错的。赚取利润实际上应该是关于最大化人类的繁荣。这才是营利的真正含义。而我们目前的分类说 Philip Morris 是一家营利性公司,Smithsonian 是一家非营利机构,但这完全是颠倒的。Smithsonian 在世界上做的好事比 Philip Morris 那些死亡商人多得多。我觉得我们就是搞混了。所有赚钱的方式都是一样的——我们都知道这不是真的。有剥削性的、榨取式的赚钱方式,也有真正增值的、实际创造净新价值的赚钱方式。
所以,如果公司致力于人类繁荣,想成为创造净新价值的公司,不想变成官僚噩梦,不想变成一个寻租的、无差别数学的庞然大物——我们不希望那种情况发生。我们就必须把这些承诺深度编码到公司之中。我觉得这才是治理的真正含义。整个 ESG 运动,所有关于这个的讨论都完全跑偏了。Philip Morris 的 ESG 评分其实非常高,这太荒谬了。这怎么可能是对的?我们必须找到一条更好的前进道路。
重新定义盈利的意义(续)
Eric Ries: 再回到精益创业和第一性原理。一旦你回到那个第一性原理,问自己:“如果公司里的每一个人都感受到一种受托责任去促进人类繁荣,那会是什么样子?你会怎么做?会有什么不同?“我开始遇到一些以这种方式构建公司的创始人,那种差异几乎令人难以辨认。当然,我一直在从大处着眼来阐述这件事。显然,想想那个天气控制创业公司。它必须走正道,不仅仅是因为如果某个超级反派拿到了它会非常可怕,而且因为你可以很容易想象……它被某个私募基金收购了,那家基金会说:“太棒了。我们现在可以让每个城镇和它的邻居进行零和博弈,竞争好天气,然后两边都敲诈……”这同样可怕。我们不需要超级反派。我们谁也不希望自己的技术被那样使用,所以我们需要防止那种情况发生。
所以,当我们把这种精神融入其中,让它成为每一个决策的一部分,我们终于能创造出人们真正想要的工作环境——只是为了做出伟大的产品。我觉得我遇到的大多数人只是想做一些令人惊叹的东西。而我们表现得好像这一切自然而然就会发生。但我现在已经和这么多公司合作过了,即使是那些真正非常优秀的公司……你去和一个团队交谈,他们会说:“我们有一座自由之岛,我们拥有完全的……”我们做到了完全正确。那是我职业生涯中最令人满足的体验之一。
你问人们:“为什么是那样……”——“它是怎么建立起来的?“他们说——“产品是怎么做出来的?""不,你生活的那座自由之岛是怎么建立起来的?“他们会说:“我不知道,问我老板吧。“我做过这个练习。我问了老板,又问了老板的老板,再往上问。有时候我会一直追到最上面,答案是:“嗯,其实是公司的创始人,他还在,是他定的规矩。“但有时候没人知道。你就会想:“这里到底谁说了算?董事会说了算吗?“他们说:“我不知道。CEO 请我来当董事的。“公司选了自己的董事会,董事会选了 C……全是自我指涉。这一切到底从何而来?我们把它当作一次意外,有时候碰巧就发生了。但那是因为我们没有像工程师一样去设计这些存在,让它们创造出我们真正想要的结果。我认为这是一个盲区。
把组织当产品来构建
作为公司建设者,我们如此专注于产品,对组织的关注远远不够。我们倾向于停留在文化这种表层的东西上,不会深入到更深的层面。所以,是的,我认为作为建设者,作为一个行业,我们需要一场新的运动来宣告:“我们要建设这样的公司——人类繁荣是它跳动的核心。我们将把它作为竞争优势来运用,不仅为了让世界变得更好,也为了赚更多的钱。“
以爱为本的商业竞争
疯狂之处在于,这些公司比它们的竞争对手值钱得多。不点名具体公司,我谈到过一家公司,它有一种令人难以置信的、以爱为核心的精神,真的……它实际上是爱自己的客户的,这很了不起。一切都从那个核心前提向下延伸,包括指导原则……我遇到的每一位 CEO 都一直在沮丧。“我没法让所有人步调一致。我告诉他们该做什么,但他们不听。“也许你在这个问题上遇到困难的部分原因是,你传达的战略对人们来说其实说不通。而这家的战略真的说得通。像对待自己的父母一样对待你的客户。人们会说:“哦,这实际上真的很简单、很清晰,我完全明白那是什么意思。“所以他们的客户忠诚度令人难以置信。你能想象他们的留存率是什么样的吗——
所以他们的客户忠诚度令人难以置信。你能想象他们的留存率是什么样的吗,如果你真正感受到……你打客服电话,感觉像是联系到了你的守护天使,而不是某个无聊的客服代表?太棒了。所以他们的竞争对手会说:“这不公平。跟它竞争是不公平的。这东西就是个怪物。”
但当你走向客户,说:“你好,你更愿意和一家由某个私募基金小白领拥有的公司合作,还是和一家把一部分你帮我创造的繁荣重新投入到你和你的社区中的公司合作?“做正确的事本身就是一种竞争优势。所以我觉得我们将迎来新一波认真对待这件事的创始人,他们会直接……他们会碾压一切。那将会非常值得期待。
Lenny: 我能看出你是如何——也许是无意间——发起了一场运动。我被迷住了,我被说服了。如果听众们也被说服了,想以这种方式建立公司,他们能做什么?是与 LTSE 合作吗?有什么他们可以做的,来帮助推动他们的公司、产品团队和业务朝这个方向发展吗?
如何迈出第一步
Eric Ries: 我觉得会有一天,关于如何做这件事会有很多资源,但现在还不多,所以还比较薄弱。我正在做这件事,但我们还在摸索正确的方式。不过我希望能开始举办一些活动、做一些事情来讨论这些内容,所以大家随时可以联系我,或者在通常的地方关注我。
如果你想自己动手,去找你的律师谈谈。这绝对是一个”去找你的律师谈谈”的时刻。我和足够多的公司合作过,我在这个行业待得足够久,已经见证了多家公司从零走到 IPO 及更远。所以我有个人朋友已经把自己的公司上市了,现在非常富有,有时候甚至会打电话给我说:“嘿,我经营着这家大型跨国公司。我们能聊聊工作中出现的这个非常具体的情况吗?“我仍然参与着这些公司。我仍然偶尔能听到实际操作是什么感受。
永远太早,直到为时已晚
我从整个过程中得出的一个主要心得是:总是太早,直到为时已晚。 所以发生的事情是,人们听到我说的这些话,比如:“哦对,我们需要确保……这里有一些你可以做的事。“显然,你可以成为一家公益型企业(public benefit corp),或者在加州,叫做社会目的公司(social purpose corporation)。有一些法律上的身份可以选择。那没问题。你可以和一家公司谈谈——因为他们的创业公司涉及从原住民文化中开发某项技术,他们拿出公司 10% 的股份放入一个基金会,基金会的受托人是与他们有互动的那些文化的部落领袖,这些人在治理中既拥有经济角色,也拥有治理角色。你可以建立一个这样的基金会。
你可以做所谓的”董事会使命承诺”(board mission pledge),即你的董事会成员作为担任董事的条件,必须承诺他们在特拉华州法律下行使其商业判断时会支持使命,而不仅仅是股东的狭隘利益。你还可以构建我们所谓的 LTSPV,它基本上是一种融资工具,运作方式类似普通的 SPV,但 SPV 中的每个 LP 都必须签署一份合同,声明如果你想做类似 B Corp 版本的事情,或者如果出现不清楚他们是否支持你从利益相关者长期利益出发做正确之事的情况,他们会支持公司的使命——他们事先就对此做出了承诺,而且我们会指定受托人来监督这些股份,而不仅仅是代表股东的狭隘利益。
永远太早,直到为时已晚(续)
Eric Ries: 我可以一直说下去。你能做的事情有很多很多。所以我找到很多公司,让创始人很兴奋。我让创始人对这些事情兴奋了很久。现在我在这件事上已经做了足够久,以至于创始人去找他们的律师,律师会说:“听着,听起来很好,但你不需要现在就做。你随时可以以后再做。“然后创始人就说:“好吧。“公司越来越大,到了某一轮融资,“是时候了吗?""啊,没事,我们可以以后再做。“然后离 IPO 就剩 18 个月了,他们已经雇了 CFO,银行家也到位了,所有事情都已经上了轨道,CEO 这时候才说:“嘿,我们当初有没有给非员工利益相关者设立那个基金会的东西?“他们说:“啊?”
“那个东西,记得我们讨论过的吗,你说还太早的?""哦对,我记得。我们做了吗?""哦不,现在太晚了。""太晚了?""事情已经上了轨道,股东不会同意的。“银行家会说:“哦,你不想把那个写进 IPO 招股书里吧。“就像,“等等,什么时候才是合适的时间?你告诉我 18 个月前的 7 月 3 号到 7 号,有一个四天的窗口期,那才是合适的时间?从来就没有合适的时间。为什么永远没有合适的时间?“我看到一个又一个一个又一个创始人被碾压过去。我认识的极少数例外之一是我们共同的一位朋友,他真的叫停了 IPO 流程,说:“不,我不会签署文件。“他们试图绕过他、不经他签字就提交文件,他说:“不行,除非我看到这个东西在里面,否则我们不做了。”
那是一件非常有争议的事情,耗费了巨大的政治资本。我无比钦佩他这么做了,但大多数创始人不会这么做。大多数创始人就随波逐流了,然后这件事就永远做不成了。所以最重要的事情就是坐下来和你的律师一起,带他们走几个假设场景,问他们,你对这个具体情况是否有充分的保护?比如,如果明天 Philip Morris 出现了……选一个你心目中的邪恶代表。我一直用 Philip Morris 举例,因为我爸爸是肺科医生,所以我从小就把 Philip Morris 视为邪恶企业的终极代表。但你可以选你最喜欢的邪恶公司。如果他们明天出现,提出以比你公司目前价值每股多 1 美元的价格,从你和你投资者手中收购公司,目的是用它来向儿童销售香烟,你有受托责任(fiduciary duty)说”是”,还是有受托责任说”否”?
信不信由你,如果你找了一家顶级律师事务所,他们为你提交了硅谷通用的标准公司注册文件,大多数治理专家读到这些文件后会说:“你有受托责任说’是’。“如果这不能让你感到恐惧、不能让你想从窗户跳出去,那我完全帮不了你。如果你不在乎,好吧,我对你无话可说。你不在乎就是不在乎。这个世界上确实有反社会型的人,他们建立反社会型的公司,我帮不了你。
但如果你在乎,就坐下来跟你的律师说:“嘿,这是真的吗?这种事会发生在我身上吗?“他们会说:“当然会。“然后你说:“如果我想防止这种事情发生,我能做什么?“他们会说:“嗯,你的投资者可能不太喜欢。投资者就喜欢现在这样。这样能最大化他们的回报。“你说:“对,但这不是我问你的问题。“而你的律师足够优秀,知道这个问题的答案。你能做什么?如果他们答不上来,就让他们回去倒带重看——Lenny,我刚才已经列举了一大堆东西了。你可以做创始人优先股(founders preferred shares),你可以做所有这些事情。
就问他们其中一些东西。不要问他们”我该不该做”,而是问他们”告诉我每一项,做了的利弊各是什么?“然后他们会说:“利是它可能真的能让你在极端情况下拯救公司的灵魂。弊是投资者可能不太喜欢。“所以你就坐在那里说:“嗯,我想做。听起来不错。谢谢你就利弊给我建议。你为我工作,你是我的律师。我想实施这个东西。”
说实话,先挑几个做起来就好。随着时间推移,你需要做的——最后一条建议——每个创始人都想要一个”奇怪的小妙招”,精益创业起步的时候也是这样。人们会说:“你能不能就告诉我那一个让我的公司永远保持创新的奇怪小妙招?“答案是,没有什么奇怪的小妙招。你想做正确的事,这是一个永无止境、持续不断的过程。我们需要建立公司的结构,使得在每一个层面、每一个时刻、公司做出的每一个决策中,会议室里做决策的人里,有人觉得自己有受托责任去做正确的事。
如果你想要的是这个结果,那么这将是一个永续的过程——不断寻找方法,在公司治理、管理结构以及实际运营和文化之间建立所需的 alignment(对齐)。这些东西不是割裂的。它们实际上只是一件事,因为公司就是一个整体。你有骨骼系统、肌肉系统和循环系统,但这些系统并不是分开的。它们协同运作,构成了你这个生命体。公司的治理结构也是一样的道理。所以我认为大多数创始人在这个方面投入不足、重视不够,这就是为什么我们最终会看到那些糟糕的结局。
真正在做的公司
Lenny: 为了让创始人在做这件事时能有点底气——因为正如你所说,投资者不会很乐意牺牲利润,员工可能也不会。你能分享一些已经实施了这些措施的公司吗,公开做过的?如果没有,你能给创始人什么样的信心,让他们觉得”很多非常了不起的公司都做了这件事,其实并没有那么难”?
Eric Ries: 我不太喜欢在这种场合点名具体的公司,因为人们总是会来一句:“但是那个……”大家太喜欢搞”那又怎么说”(whataboutism)了。而且我也喜欢让公司自己为自己发声。我掌握了太多机密信息,必须非常谨慎。但我所描述的一切都不是假设性的、不是虚构的。我自己亲眼见证了这一切很多很多很多次,我讲的每一个故事背后都是一家真实的公司、你可以去认识的真实的人。
他们不一定是最响亮的声音。我们这个行业里最响亮的声音往往是那些有东西要卖的人,而这些公司只是在默默做正确的事。他们不需要到处宣扬,因为他们正在做。所以不幸的是,声音越大、争议越多、越刻薄的内容,反而越不可能是准确的。
我想说的是,这在业界几乎是一个公开的秘密。让我真正感到惊讶的是:我一直以为我们建立公司的方式——那些我们所有人都在遵循的标准公司治理实践——一定在研究中被充分论证过,一定有充分的证据支持,一定是唯一且最好的方式。但当你开始研究这些东西的时候,你很快就会找到那些例外。我之所以进入这个领域,是因为丰田——仅仅因为我把我的东西叫做精益创业,我想从精益制造中借鉴一些东西,所以我研究了丰田。
丰田与非传统公司治理
Eric Ries: 丰田有一种非常奇怪的公司治理体系,属于旧式日本 keiretsu(企业集团)体系的一部分,丰田家族在所有权上与其供应商交织在一起……以现代标准来看,这是一团糟糕透顶的公司治理乱象。但如果你去看文献,每个研究丰田的人都会说:“嗯,他们之所以能做出这些了不起的事情,是因为他们有一种长期主义的哲学。“我以前会想:“如果你要受制于季度回报,怎么可能坚持长期主义的哲学?“他们就说:“哦,他们不在乎那些东西,因为他们有另外一套结构。”
我就想:“哦,那倒挺奇怪的。“有一段时间我琢磨着:“嗯,大概只是丰田而已。我们应该建造更多丰田那样的公司。“但后来我发现还有很多家族经营的公司,家族也创造出了这样的成果。再后来我开始接触由基金会控制的公司。有意思的是,OpenAI 现在是最炙手可热的新事物。OpenAI 是一个非营利基金会,旗下有一个全资拥有的营利性子公司。我最初觉得:“哇,OpenAI 真独特。“但实际上这非常普遍。世界上有很多这样的公司。在某些欧洲国家,实际上 20% 到 25% 的上市公司都是这样架构的。所以这其实完全正常。
基金会控制公司的竞争优势
几年前我去其中一家公司做演讲,那次经历让我大开眼界。我去那里跟他们的 CEO 和管理层交流时,我问:“在公开市场和季度压力下运作是什么感觉?“他们说:“嗯,我们不搞那些。“我又问:“哦,那这会不会是一个重大的竞争劣势?你们是非营利组织啊。“他们说:“你开玩笑吧?我们把竞争对手打得落花流水,因为他们不得不在那些破事上浪费时间。我们不用浪费那个时间。“我就想:“哦。”
有人听说过 Warren Buffett(沃伦·巴菲特)吗?为什么他的公司具有竞争优势……人们总是说:“哦,如果你放弃利润,你的投资者不会喜欢的。“我就想:“这些公司是碾压级别的存在。我们放弃了什么?“这就是为什么我说要从利润的错误定义开始纠正。如果我们用这种方式来做,实际上会赚更多的钱。以 Amazon 为例,尽管它有很多问题,但它的长期思维和以客户为中心,这是一个重大的竞争优势。
我跟 Amazon 合作过,也跟他们的很多竞争对手合作过。竞争对手坐在那里抱怨的时候特别有意思:“哎呀,Amazon 要进入我们的市场了,我们必须做 X、Y、Z。“我总是反问:“你觉得 Amazon 会做 X、Y、Z 吗?“他们说:“哦不,他们不用做那些。我们才不得不做这些投资者要求的破事,还有那些激进分子什么的。“我就说:“嗯,既然你在跟他们竞争,而他们不用做那些破事,也许你应该试试他们的做法。“然后他们说:“哦,我们的投资者不会喜欢的。”
我就说:“你的投资者希望你被 Amazon 揍得满地找牙?你确定吗?“
股东至上理论的谬误
所以我确实认为……我的一个认识是,以投资者为中心的公司观,即所谓的股东至上理论,它就是错的。即使你只关心为股东最大化回报,它也是错的。股东有一种倾向,就是杀掉下金蛋的鹅,而且这种情况已经持续了 150 年。这些案例研究可以追溯到 19 世纪。这是一个非常、非常古老的想法。
而解决方案其实都藏在明面上。前几天我读到一篇研究论文,研究的是——我记得是在丹麦——基金会控制的公司与匹配了条件的非基金会控制上市公司的对比。他们发现基金会控制的公司从财务表现来看跑赢了对照组。
我就想:“这个结果怎么会没有引起轩然大波?“如果这是真的,那么我们所有人受到的关于什么是一家营利性公司的教条灌输,很大一部分就是错的,因为那些公司按理说应该充满了低效和代理问题。我对那些理论太了解了,我可以告诉你所有应该成立的东西。如果这些东西不成立,那到底是怎么回事?然后我恍然大悟:“等等,我去过那些基金会控制的公司。“我去过 Hershey(好时),它就是由基金会控制的——创始人把股份捐给了一个运营学校的基金会。
我亲眼见过,但即使亲眼所见,我也花了一段时间才意识到:“等等,这是一个公开的秘密。“所以当我发现这是众所周知的事、却从来没有人给我提供过这个选择时,我感到有些震惊。怎么从来没有人给我机会用这种方式建一家公司?而且如果你去研究即使是 OpenAI 或任何现在采用这种架构的公司,情况总是非常特殊——他们本来走在标准路径上,然后发生了一些事情让他们可以做一些不寻常的事。我就想:“我不明白为什么非要等到特殊情况才这样做。我们为什么不能有意为之?“
精神控股公司
于是我开始构思这个想法,我称之为精神控股公司——基本上就是处于中心位置、负责守护整个公司生态系统灵魂的那个实体,也就是企业的核心。你在 OpenAI 的例子中可以明显看到这一点。如果有人研究过 Anthropic 的长期利益信托,它的结构并非巧合地与此相似。但你可以在很多例子中看到这种做法。
然后我开始向创始人推介,说:“看,让我试着向你展示这样做的好处。“令我惊讶的是,有多少创始人会情绪崩溃。有创始人在我面前哭了出来,他们说:“这才是我一直以来真正想做的事情,但我从来没觉得……从来没有人跟我谈过这件事。我不觉得这是可能的。我想把某个关键的利益相关者放在我所做一切的中心位置,或者我们做这件事有某个原因,我想让那个原因成为核心。“我之前提到过一家创业公司,他们试图招聘学术研究人员离开实验室、全职加入创业公司。这是最困难的招聘挑战之一,因为你要招的不是刚从学校毕业的年轻人。而是那些终身教授,他们要放弃非常好的生活来做这件事。他们只有在你深深相信你的使命时才会这么做,但他们并不天真。
每一个这样的人问的第一个问题就是:“我怎么知道这项技术不会被用于邪恶的目的?我在学术界有完全的学术自由。我要为你放弃这些。让我相信这是真的。“如果你的回答是:“嗯,我们相信某个宣言,我们是好人,我们还从一个承诺会做好人的投资者那里拿了钱,“那是远远不够的。
精神控股公司的实践与勇气
Eric Ries: 所以他们不得不从头做起,坐下来和律师一起研究:“我到底要怎么做,才能让这个承诺对这些人是真实的,才能让他们愿意加入?” 这样的情况我一次又一次地看到。如果我们有勇气去做这件事,它可以解决今天那些被认为是无解的商业问题。这很酷,但我认为人们要去做一件与众不同的事情时,会感到害怕。所以我认为随着时间的推移,我们会有更多这样的案例被记录下来,会有更多的先例。但我必须说,不是为了过度回溯精益创业,但我永远不会忘记那些早期的人坐在我面前说:“证据在哪里?有什么大公司在做这个?” 我会说:“等着吧。如果你现在不想要这个竞争优势,你以后也可以拿到。”
回过头来看那些书中被报道过的人。许多早期的精益创业实践者如今都是极其成功的人,因为他们没有觉得自己必须等待。他们能够独立思考,把这些原则应用到自己的生活中。我不要求任何人凭信仰接受这件事。这不是”你得照我说的做”。更像是,你自己思考。如果这对你来说是有道理的,而且没有人能用让你信服的方式向你解释为什么这是个坏主意,那就允许自己至少试一试,看看会发生什么。
闪电问答环节
Lenny: 多漂亮的第一性原理思维的例子,兜兜转转又回到了我们最初的问题。我想这会是我们做过的最长的一期节目,太棒了。而且我觉得这可能是我新的最喜欢的一期,所以这是一次很好的对话。
Eric Ries: 谢谢你这么说。
Lenny: 那么,我们进入了非常令人兴奋的闪电问答环节。看看我们会撞出什么样的闪电。准备好了吗?
Eric Ries: 准备好了。
Lenny: 你最常向别人推荐的两三本书是什么?
Eric Ries: 有一本书叫《The Enlightened Capitalists》,讲的是一百年来人们试图解决这个问题却失败了的故事,里面是详细的案例研究,没有哪个创始人应该从头到尾读完它。太令人沮丧了。但了解其中一些故事很重要——那些创始人愿意在临终前谈起自己的遗憾,谈起他们本希望能做出不同选择的事。对于那些认为存在某种”一招制胜”技巧的人来说,这本书会帮你打消那个念头,但了解前人犯过的错误是有帮助的,这样我们今后才能做得更好。
说点轻松的,我真的希望更多人读一读 Ken Liu 的《The Dandelion Dynasty》(蒲公英王朝)。那是一个丝绸朋克奇幻系列,作者是……他是一位出色的作家,如果你想知道什么是真正的对工程学的乐观主义、对工程师改变世界的能力的信念,但被转换到一个完全不同的环境中,读起来真的太棒了。有声书也非常好,我最近一直在推荐它。
然后如果你只想要纯粹好玩的东西,虽然现在不是夏季读物的季节,但也同样是一本非常棒的有声书——Martha Wells 的《Murderbot Diaries》(杀手机器人日记)。她真的是一位非常出色的作家,我知道表面上看起来很轻浮,但其实里面有很多东西。我不会剧透……我说的任何话都不会是剧透,但在这样一部看起来只是有趣的动作冒险故事里,实际上有着惊人的深度。我等不及了,我希望它有一天能被拍成电影。如果他们拍得好的话,一定会非常精彩。
Lenny: 最近最喜欢的电影或电视剧是什么?
Eric Ries: 我确实一直试着……我已经没有时间了。我有年幼的孩子,所以我现在消费媒体的时间极其极其有限。但我一直在强迫自己硬着头皮刷一堆 Disney+ 的星战系列,结果几乎普遍令人失望,除了《Andor》(安多)。没看过《Andor》的人,它真的非常好。不是”作为一部 Disney+ 星战剧来说不错”,而是作为电视剧来说真的好,当我看到它的设定时我自己都不敢相信。
Lenny: 我也是《Andor》的超级粉丝,推荐过很多次了。而且我觉得现在每一部新的星战剧,我都在拿它和那个标准比,然后就……
Eric Ries: 这就是问题所在。如果他们就让那个创作团队来、让他们去拍剧,就没问题了。那就能弥补他们所有其他的失败。
Lenny: 对,或者就继续拍续集。每一部我都会看。你面试别人的时候最喜欢问的一个面试问题是什么?
Eric Ries: 我有史以来最喜欢的面试问题,是请人们描述他们在职业生涯中学到的一个最佳实践,然后让他们真正深入进去。比如,“给我讲讲你学到这个最佳实践的故事”,你就让他们讲,然后你说,“很好。给我讲一个这个最佳实践不适用的场景。” 大多数人都答不上来。
Lenny: 真有趣。最近发现并且真正喜欢的产品是什么?可以是 app,也可以是实体产品。
Eric Ries: 我刚跟一个朋友在夸一个产品。这是有史以来最无聊、最傻的产品。就是一个加湿器。你会说,“好吧,加湿器还能搞砸?” 但实际上,要做到完全正确是非常困难的。里面有各种各样需要细心打磨的细节。我们之前谈到过匠心,那正好在我脑子里。这是一个非常朴素的家电,实际上我试了十个不同的型号,就为了找到一个不会让我抓狂的……
因为你想在晚上开着它,但如果它有个灯整晚亮着,那真的很烦。如果它有噪音,那也不行。有各种各样的问题让你觉得,“天哪,如果有人哪怕花一分钟用 jobs to be done 框架想想这个实体设备到底应该做什么,他们会造出来的就是这个产品。” 而且我不敢相信……现代资本主义中,有那么多品类连一个这样的产品都没有,或者只有一个。你会想,“怎么不是每个产品都这样?” 而我们之前讨论的那些力量,就是原因所在。在这么一个朴素的物件上看到这一点,还挺有趣的。
Lenny: 你找到的那个加湿器具体是哪个?
Eric Ries: 是 Levoit 300S 超声波加湿器。Levoit 300S。就是一个做得很好的 60 美元的家电,没什么特别的。
Lenny: 他们马上要迎来一波销量飙升了。加油 Levoit。
Eric Ries: 在某个地方、某个时候,一定开过一场产品管理会议讨论怎么做这个东西,我现在都能想象得出来。有人说,“听我说,我们现在做这些东西的方式太烂了。能不能让我把本来在底部的东西放到顶部,让我把这个放进去,把那个放进去?” 然后有人说,“谁在乎?那会增加 3 美分的成本,把我们的利润率降低这个那个。” 然后有人说,“不,我们真的应该把它做对。” 我们应该奖励这样的人。
Lenny: 对,这可能会把我们带偏题,但我最近一直在想,大多数时候最好的东西,往往来自于一个人对做那件事拥有单方面的控制权,一次又一次地去做。
丰田的总工程师制度
Eric Ries: 对,如果你研究丰田生产系统,他们有一个职位叫总工程师(chief engineer)。我希望更多创始人了解这个概念,因为当公司变大时,重要的是能够赋能为你工作的那些创始人,而不仅仅是把自己视为唯一的创始人。总工程师是对整条产品线——比如一辆汽车从始至终——拥有道德权威的人。
我不知道现在是否还是这样,但在人们大量写作丰田的那个年代,公司内部并不把这辆车叫丰田凯美瑞,而是叫”某某桑的车”。总工程师的特点是,他们没有下属。没有人向他们汇报,他们只有所谓的”大房间”(big room),如果你研究过那套体系的话。他们有一个自己的工作地点,有一个小型团队,他们手里掌握着设计和整辆车。参与一辆车开发的人数会膨胀然后再缩减——因为先是设计阶段,然后进入生产阶段,最终人数又会降下来。
做产品最重要的事情就是做权衡决策。如果你让每个人自己做自己的权衡决策,结果会完全不同。你很难形成任何一种连贯的设计推进力。所以这个理念就是:“这个人就是做那些决策的人。“如果两个部门之间为成本应该多少发生争执,或者有人提出这样的问题——“听我说,我可以把空气动力学性能提高这么多,但要花这么多钱,而且意味着我们必须把这个按钮从金属换成塑料”——你需要一个能拍板的人说:“不行,那个按钮必须是金属的。我们承诺过要做到这个。“或者:“按钮无所谓。“你需要一个对这些东西了然于胸的人。关键在于,这些决策必须能装在一个人的脑子里。所以这个职位叫总工程师。
Lenny: 我还以为这期节目已经挖完金矿了。这太有意思了。我觉得这完全说得通,我也同意公司里应该更频繁地采用这种做法。哇。
Eric Ries: 对,这种结构性的东西确实很缺失。
人生格言
Lenny: 好的,还有两个问题。你有没有一个最喜欢的人生格言,会经常对自己重复、分享给别人,或者经常回顾以帮助自己做出正确选择的那种?
Eric Ries: 我不展开解释整个背景了。这个格言是:真实的东西不可被威胁,不真实的东西不存在。
Lenny: 这句话够我们再单独做一期节目来拆解了。太棒了。
未被广泛接受的术语
Lenny: 最后一个问题,你发明了很多术语和概念。有没有哪个你觉得没有火起来、没有像你预期的那样广泛传播的?
Eric Ries: 哦,有。有太多了,太多了。剪裁室地板上堆满了各种各样的东西。我经历过很长一段时间的尝试……那是在《精益创业》之前,我尝试用细胞生物学作为创业公司的比喻。本来应该是……我想要一个更偏生物学的、不那么机械的比喻。结果没有一个人听懂我在说什么。完全行不通。
但我跟你说最让我意外的那个。这个很小,但我真的以为它会起飞,人们会对此非常热衷。我写了很多关于它的东西,但就是没能让它扎下根来。大家都知道 viral loops(病毒式循环),其实还有 engagement loops(参与循环)。我当时想,“就像我们有关于产品如何增长的 viral loops 一样,我们也应该有关于人们如何留存的 engagement loops。”
留存是有科学的,人们也从”如何让你的产品更让人上瘾”这个角度写过相关内容。但这其实不是我想说的重点。我要说的是能够建模客户如何流动的动力学。《精益创业》里有一个版本,我称之为可持续增长定律(law of sustainable growth),即新客户作为老客户行为的自然副产品而到来的情况。
这显然对 viral loop 是成立的。大家都知道付费获客引擎,即每个客户的边际利润驱动你投资客户获取的能力。但还有一种我称之为粘性引擎(sticky engine of growth),这是这个想法在《精益创业》中留下的痕迹。但我其实认为这是一整个非常有趣的领域——你想想看,思考方式是这样的:一个人正在过自己的生活,过得挺开心的,然后因为某个原因使用了你的产品。他为什么用你的产品?什么让他回到你的产品?
实际上,原因没那么多。可能是因为你给了他一个人造通知,比如”嘿,请回到我的产品来”,然后他照做了。可能是因为某种有机通知发生了,比如他在你的平台上收到一封朋友发来的邮件——PayPal 有史以来最成功的客户邮件之一就是”你有钱了,点击这里取钱”。它的点击率非常高,也许仅次于 Facebook 的”有人在照片里圈了你,但我们不会告诉你是哪张照片。所以你最好点一下,对吧?”
所以是某个事件触发了回访。如果不是以上两种情况,那一定是因为定位。你大脑中的某个东西自主地告诉你去使用那个产品。你为什么去打《魔兽世界》?需要收到通知来提醒你吗?你喜欢玩,没人需要告诉你。你就是喜欢做这件事,或者生活中发生了某件事让你产生了需求——你饿了,所以你去那个地方弄吃的。
既然人们回到一个产品只有这几种方式,那它们都可以被测量、研究和理解,精确程度可以跟我们对 viral loops 的理解一样。我怎么也想不通为什么 viral loops 得到了所有的关注,而没人对参与的科学有哪怕一丝兴趣。曾经有一段时间,我认为 engagement loops 会让我比《精益创业》有名得多。我错得非常、非常离谱。
Lenny: 这也让我想到,你好像也是第一个把增长描述为循环、以及这些具体循环和增长引擎的人,现在很多人都这么说了。
Eric Ries: 对,增长这个概念是我的。
Lenny: 是的。哇,我们又发现了一个。我猜还有很多。Eric,这太精彩了。我确定这是我新的最喜欢的一期。非常感谢你来参加节目。
尾声
Lenny: 最后两个问题。如果大家想联系你,在哪里可以找到你?以及听众怎样能帮到你?
Eric Ries: 你仍然可以在 theleanstartup.com 找到我。我在所有常见平台上的账号都是 @ericries,E-R-I-C-R-I-E-S,不过我现在真的不怎么用社交媒体了,所以加入 theleanstartup.com 的邮件列表可能是跟踪我生活中重大动态的最佳方式。另外,我确实偶尔会上社交媒体发一些公告。如果有人想联系我,直接发邮件到 eric@theleanstartup.com 就行。
Lenny: 听众怎样能帮到你?
Eric Ries: 哦,这太贴心了。说实话,大家只需要把这些想法拿去实践,从中学习,然后告诉我结果如何。这就是我唯一想要的回报。告诉我发生了什么就好。即使你……我从中学到最多的人,恰恰是那些尝试了我的想法,觉得这是他们听过最愚蠢的东西,完全行不通,甚至很生气的人。告诉我,我想知道所有的细节,因为说实话,我的工作——我是这样看待自己的——我的工作就是努力理解这些东西。我只想知道真相。如果我错了……我不喜欢犯错,我和其他人一样对自己的想法有感情,但我真的非常努力地去发现什么是真实的。所以如果你发现了什么新东西,请告诉我。这就是我唯一的请求——让我知道。
如果你不去尝试,那怎么可能发现任何新东西呢?每天我都会遇到刚听说精益创业的人。对我来说,这已经太久了……每天都有人第一次读这本书,然后有问题想问……所以如果你正处于这个阶段,刚刚接触这些想法,太好了。我写了几本书,你可以去读,请为这些想法去实践。如果你是那种已经倦怠的人,觉得”哦,我已经听腻了精益创业”,我在这次对话中提到了一些新的想法。去试试其中一个。即使不管用,或者管用了,甚至我从中学到很多的那些人,他们尝试了,在行不通的过程中,反而产生了更好的想法,然后去做了那个。
创业者的集体力量
那就去做吧。我唯一的请求就是你告诉我结果如何。我想从那些正在把这些变成现实的人身上学习。我希望当我们作为一个整体走到一起时——不仅仅是精益创业运动,而是作为一个创业运动,作为一群从根本上相信通过从零开始建设组织可以让世界变得更好的人——我们怀有这个信念:未来并不可怕。新事物不应被恐惧或排斥,而是存在一种负责任的、有条理的、创造价值的方式,来推动世界正面变革。这种精神并不是普遍被认同的,如果有人在听你的播客,我的假设是你属于我们这个群体。所以我们有一种集体责任去思考:“除了我的公司、我的创始人身份、我的故事之外,我们代表什么?作为一个群体,我们共同的价值观是什么?我们如何让这些价值观在世界中体现出来?”
我留给大家最后一个想法。我们作为一个群体拥有的力量大得令人难以置信。因为现有系统的运作方式,我对我们当前所谓的晚期资本主义、金融化的资本主义引擎所了解的是,它在很多方面都在毁灭价值。虽然市场会惩戒公司,要求它们创造的价值大于毁灭的价值才能存活,但新公司——也就是我们所建造的那些——是如此宝贵。
公司的平均寿命在缩短
支撑这一切的核心洞见如此有价值,以至于你所创造的那个存在体在崩塌之前能承受巨大的损害。然而,如果你看看上市公司的平均存续时间,看看 CEO 的平均任期,你所看到的所有指标都很严峻。我们摧毁公司的速度快于创建公司的速度,以至于美国上市公司的总数比我们有生之年的峰值下降了一半以上。那并不是很久以前的事。到底发生了什么?原因是这个系统需要源源不断的新鲜血液来补充它所摧毁的。
如果你在听这个播客,你可能就是那个供应者。你是这个系统赖以生存所需之物的制造者。所以事实上,我们作为一个群体,有权决定这件事以什么条件进行。如果我们不想再把自己的作品——我们的孩子——卖掉换钱,我们不必这么做。如果我们希望它们在世界中拥有某种精神,希望它们成为值得信赖的合作伙伴,为我们所关心的人做些什么,我们可以坚持这是我们将倾注一切的前提条件。
所以我不想说我知道所有答案,也不想说我个人的价值观应该胜出,但我确实认为我们有一套共同的价值观,我们应该更加热切地捍卫它们。希望你的听众能把这一点放在心上。
Lenny: 一个美好而有力量的结尾方式。Eric Ries,非常感谢你来参加节目。
Eric Ries: 我的荣幸。
Lenny: 大家再见。
感谢大家的收听。如果你觉得这期节目有价值,可以在 Apple Podcasts、Spotify 或你喜欢的播客应用上订阅节目。也请考虑给我们评分或留下评论,因为这真的能帮助其他听众找到这个播客。你可以在 lennyspodcast.com 找到所有往期节目或了解更多关于节目的信息。下期再见。
术语表
| 原文 | 中文 |
|---|---|
| 2021 Vintage | 2021 年份(指 2021 年融资的那批创业公司) |
| agency problems | 代理问题 |
| agent | agent(智能体) |
| Agile | Agile(敏捷开发) |
| Al Ries | Al Ries(定位理论创始人) |
| alignment | alignment(对齐) |
| Andor | 《Andor》(安多) |
| Ben Silbermann | Ben Silbermann |
| best practice | 最佳实践 |
| big room | 大房间 |
| board mission pledge | 董事会使命承诺 |
| cap table | 股权结构表 |
| chief engineer | 总工程师 |
| Clay Christensen | Clay Christensen(克莱顿·克里斯坦森) |
| cold sales outbound | 冷启动销售外联 |
| Conway’s Law | Conway’s Law(康威定律) |
| deductive logic | 演绎逻辑 |
| Delaware law | 特拉华州法律 |
| descriptive | 描述性 |
| DevOps | DevOps |
| engagement loops | engagement loops(参与循环) |
| ESG | ESG(环境、社会和公司治理) |
| fiduciary duty | 受托责任 |
| gatekeeping | 把关 |
| Hershey | Hershey(好时) |
| hockey stick | 曲棍球棍曲线 |
| IMVU | IMVU |
| innovation accounting | 创新会计 |
| inverse kinematics | 反向运动学 |
| jobs to be done framework | jobs to be done 框架(待办任务框架) |
| just so stories | ”就是这样”的故事 |
| Karri Saarinen | Karri Saarinen |
| keiretsu | keiretsu(企业集团) |
| Ken Liu | Ken Liu(刘宇昆) |
| law of diminishing returns | 收益递减定律 |
| law of sustainable growth | 可持续增长定律 |
| Lean Startup | 精益创业 |
| leap of faith assumption | 关键信念假设 |
| Lenny | Lenny |
| Levoit | Levoit |
| Linear | Linear |
| local maxima | 局部最优 |
| long-term benefit trust | 长期利益信托 |
| LP | LP(有限合伙人) |
| LTSE | LTSE |
| magic pixie dust | 魔法仙尘 |
| Martha Wells | Martha Wells |
| Merchants of Death | 死亡商人 |
| Murderbot Diaries | 《Murderbot Diaries》(杀手机器人日记) |
| MVP | MVP(最小可行产品)(已在术语表中) |
| naval gazing | 自省/内省 |
| net positive | 净正面 |
| outtakes | 废弃版本 |
| paid engine of acquisition | 付费获客引擎 |
| Paul Graham | Paul Graham |
| Philip Morris | Philip Morris(菲利普·莫里斯,烟草公司) |
| pivot | pivot(转向) |
| prescriptive | 规范性 |
| procurement policy | 采购策略 |
| product-market fit | product-market fit(产品市场契合) |
| public benefit corp | 公益型企业 |
| reality distortion field | 现实扭曲力场 |
| referrer logs | 反向链接日志 |
| SEC | SEC(美国证券交易委员会) |
| shareholder primacy theory | 股东至上理论 |
| silk punk | 丝绸朋克 |
| singular control | 单方面控制权 |
| Six Sigma | Six Sigma(六西格玛) |
| Smithsonian | Smithsonian(史密森学会) |
| social purpose corporation | 社会目的公司 |
| span of control | 管理幅度 |
| spiritual holding company | 精神控股公司 |
| SPV | SPV(特殊目的实体) |
| Stage-Gate | Stage-Gate(门径管理) |
| Steve Jobs | Steve Jobs |
| sticky engine of growth | 粘性引擎(增长引擎之一) |
| The Dandelion Dynasty | 《The Dandelion Dynasty》(蒲公英王朝) |
| The Sims | 《模拟人生》 |
| Thomas Kuhn | Thomas Kuhn(托马斯·库恩) |
| turn the crank | 拧扳手式(机械式) |
| uncanny valley | 恐怖谷 |
| viral loops | viral loops(病毒式循环) |
| Warren Buffett | Warren Buffett(沃伦·巴菲特) |
| willfully ignorant | 故意的无知/故意装无知 |
| Y Combinator | Y Combinator |
| Zuck | Zuck |
此文档由 AI 分片翻译(translate_long_document)