with Spencer Greenberg
the podcast about ideas that matter

Episode 152: How can you tell if you're cut out for entrepreneurship? (with Matt Clifford)

Enjoying the episode? Want to listen later? Subscribe on any of these apps or stores to be notified when we release new episodes:

April 6, 2023

What are "variance-amplifying" and "variance-dampening" institutions? Has the world been getting weirder recently? Should entrepreneurs aim for variance amplification or variance dampening? What percentage of people should be entrepreneurs? What traits and skills are necessary for successful entrepreneurship? How has ambition changed over the course of history? How can entrepreneurs know if they're really changing the world, or just doing something slightly before someone else did it, or just doing something that would have happened anyway? How can entrepreneurs avoid getting mired in "tar pit" ideas?

Matt Clifford MBE is cofounder and CEO of Entrepreneur First, the leading technology company builder that invests in top technical individuals to help them build world-class deep technology startups from scratch in six locations across Europe, Asia, and Canada. Since 2011, Entrepreneur First has created over 500 startups worth over $10bn including Magic Pony Technology, Tractable, and CloudNC. Matt is also Chairman of the UK's new Advanced Research and Invention Agency (ARIA), which aims to enable exceptional scientists and researchers to identify and fund transformational research that leads to new technologies, discoveries, products, and services. Matt sits on the board of Code First Girls, which he co-founded in 2013 to teach young women how to code, and is a member of the Innovate UK Council. Matt started his career at McKinsey & Co. and holds degrees from Cambridge and MIT, where he was a Kennedy Scholar. He was awarded an MBE for services to business in the 2016 Queen's Birthday Honours. Follow him on Twitter, interact with him on LinkedIn, or learn more about his work at Entrepreneur First.

JOSH: Hello, and welcome to Clearer Thinking with Spencer Greenberg, the podcast about ideas that matter. I'm Josh Castle, the producer of the podcast, and I'm so glad you've joined us today. In this episode, Spencer speaks with Matt Clifford about variance-amplifying institutions, traits of successful entrepreneurs, and ambition.

SPENCER: Matt, welcome.

MATT: Thanks so much for having me.

SPENCER: Now you're in a really unique position because you've worked with well over a thousand different entrepreneurs to help them reach their entrepreneur potential and potentially have an impact on the future through new technology. So today, I want to discuss with you a number of different topics, loosely connected to technology in the future. I think it's gonna be a really interesting discussion.

MATT: I'm excited to be here and excited to talk to you about it.

SPENCER: Awesome. So let's start with variance amplifying institutions. Do you want to tell us what that phrase means and why you think it's important?

MATT: This is an idea that I started thinking about, probably six or seven years ago, in response to the sense that I think a lot of people got that the world was getting weirder. You saw that in politics, you saw it in society, in culture, just everywhere you looked — you probably saw this on Twitter — people started talking about, like the Reuters (meaning like, it felt like we were living almost in a TV show at times). And so I started thinking about why the world got so weird. And the answer I came up with was this idea that most of the last few hundred years, what we might call modernity, has been dominated by the sort of removal of variance from our lives. And that's broadly a really good thing. Variance was the dominant thing for thousands of years before that. And that led to sort of nasty, brutish and short lives. As Hobbes said, the triumph of modernity is sort of actually reducing variance and getting to a place where our lives are more predictable. And so as a result, when the internet came along and provided a mechanism for not just reintroducing but amplifying variance in our lives, we've experienced this as weirdness. And so very briefly, what I mean by a ‘variance amplifying institution', is an institution that sort of selects for tail behavior — the most unusual behavior — and then amplifies it. I think the internet is probably the paradigmatic example because everything more or less that you read on the internet, you're reading because it's been selected by a large group of people for amplification. What do they select to amplify to you? Well, sort of the tail-like behavior that they see. Obviously, last year, there was this extraordinary phenomenon of Wall Street bets where the kind of Reddit took on Wall Street, and at least briefly won. You could argue that the whole of the Donald Trump phenomenon is a sort of internet-amplified phenomenon. And I think everywhere you look today, you see counter to all these forces that have reduced variance through modernity, the internet, picking up and amplifying things that look and feel crazy to us.

SPENCER: So is the idea that many, many normal things happen every day but somehow the internet likes to broadcast the weirdest of the weird, and that this somehow actually affects society is not just, “Oh, look at this cute curiosity,” but then there's sort of a feedback loop and it's actually changing society?

MATT: Exactly, right. So I think the first order effect is just when weird stuff happens, it gets amplified. But if that didn't also change people's behaviors and incentives, then it probably wouldn't be that interesting. It would sort of be the equivalent of the fact that viral videos that are cute or funny get amplified. Well, there's a change in culture, but it's not a very interesting one, in my view. The thing that's interesting is that we're all dynamic actors, we all respond to incentives. What the internet and related technologies/institutions have done is that they've actually increased the payoff to extreme behavior. Because in the past, extreme behavior didn't really scale and it could be dampened locally by, kind of if you like the forces of normality. But what the internet does is it takes that behavior and gives it a much bigger platform. And as a result, people that are sort of seeking maximum impact — seeking maximum stage time, if you like — are incentivized to act in an ever more extreme way, in order to be picked up and amplified in this way. One example that I think brings this to life is to compare the richest person in the world today, Elon Musk, to (say) the richest person (call it) 100 or something years ago, JP Morgan, and say like, “Look at who they were trying to impress? Look at what their behaviors that were amplified for general consumption by the public?” And this sort of comes to life that why Elon Musk behaves the way he does is largely because he's rewarded for it. And I think this idea that it's actually a pretty significant change if the most powerful people in the world are incentivized to embrace variance amplification. It's very different from the previous paradigm, where the most powerful people in the world were actually encouraged to embrace a variance dampening, because they wanted to preserve the existing structure of society.

SPENCER: This reminds me of an example I saw recently. So on YouTube and Tiktok, you have some great educators who have mental health challenges, who talk about their mental health challenges in a very helpful way. But then you also have this group of people with mental health challenges, who display their mental health challenges. So they're not educating you about it, they're actually just acting in totally bizarre seeming ways that lead to getting a lot of attention and actually help them build their brand.

MATT: Right, exactly. So you end up in this situation where almost in every walk of life, what we see today is the almost anomalous behavior, the extreme behavior. And therefore, even people with goals that we might all say are kind of praisable goals, what we see is the way to get most attention is to be at the extremes. And so, I think that this sort of taps into a broader topic that I'm interested in, which is how do ambitious people think about their lives? I think one consequence of the rise of variance amplifying institutions is that taking more risk is actually strongly encouraged by variance amplifying institutions, because you get this sort of idea of uncapped upside that comes through the internet. Broadly, I'm very in favor of that. A lot of what we're trying to do at Entrepreneur First is encourage people to seek uncapped upside rather than sort of structured (if you like) variance dampened careers. But I do think there are also these more harmful, or at least questionable, consequences as well.

SPENCER: I see. So you think that the way that variance amplification works today actually should change the way people think about entrepreneurship?

MATT: I think it's certainly one way of thinking about how people, particularly people who are talented and ambitious, choose to spend their careers and their lives. One way I think about traditional prestigious careers — like going to work in a prestigious investment bank, or consulting firm, or even more recently, a big tech company — is that these sort of quite hierarchical institutions, you can think of them as sort of ambition shock absorbers. They take very talented, very ambitious people, and they give them a very structured, more or less predictable path to acquire and almost to convert their talent into influence, wealth, prestige. The upside is it means that you don't have those smart, talented, ambitious people engaging in destructive, competitive behavior with each other, or I would argue, at least generally. Entrepreneurship is quite different. Entrepreneurship is uncapped exercises; it's not hierarchical, it's not structured. And so what you get in entrepreneurship is people seeking the biggest, most extreme outcomes. I broadly think that also is a good thing. I think if we want to see big change in the world, it's not going to come from variance dampening institutions. And clearly, there are many aspects of human existence where we might want to see big change. But I do think it sort of creates a different kind of elite that is much less interested in conformity and kind of adherence to social norms — the Elon Musk's — which in turn, when these people become successful, they obviously really influence many aspects of life, as we're seeing now with Twitter.

SPENCER: If we think about this kind of variance dampening institutions, maybe you could go into a bit about what they were actually doing that was dampening. And then what happened to those institutions today? Do they still exist? Are they still trying to dampen variance and just being all successful at it?

MATT: I don't want to pretend that they've gone away. I think not only that they've not gone away, but if you zoom out and look at everything around, you probably see variance dampening institutions are still dominant. What do I mean by variance dampening institutions, wildly any institution that makes things more predictable for all the stakeholders concerned. So, you could argue that constitutional democracy is the variance dampening institution, like we have a more or less predictable way to achieve the peaceful transfer of power. And my guess is that the vast majority of people listening to this think that's a good thing. I certainly do. The alternative, the pre modern way of transfer of powers through battle or assassination or whatever, feels bad, right? That's one example. But I think across almost every aspect of life, you see as modernity emerges in the last few hundred years, you sort of see norms, rules, conventions that make life more predictable and reduce variance. I think the ultimate example of that, which broadly I think has been a real benefit for the world, is the idea of a career. In a way, the idea of a career as a variance dampening institution, like the fact that you create these paths where people compete for promotions in this sort of, broadly, peaceful way. These are really good things. I think, generally, our elites today are probably more peaceful than at any other point in history. There's loads more of these. You could think about the role that central banks have played in the modern economy, or the role that international institutions (multilateral institutions) have played. These are all variance dampening institutions. And in most ways that they're still the most powerful institutions in society, I think what's changed in the last decade, as a result largely of the Internet and adjacent technologies, is they're no longer completely dominant. I think today, it really is plausible that if you're a smart and ambitious individual, then starting a company rather than going to Goldman Sachs is actually, in expectation, the way to maximize your impact in the world,

SPENCER: It seems to me that the quality of the kind of standard good career paths has changed. For example, I think people going to law school today often don't really realize until they're in too deep that they ended up going to law school for a few years, they often end up owing a lot of money, if they're taking out loans for that. And then they're sort of forced to do corporate law rather than other high paying law jobs to kind of pay back that amount of money. And then they find that it's actually very grueling, very intense for quite a long period, before you start getting the rewards that you expect of that career path. So it's like, this idea that, “Oh, this is such a good career,” I think people have started to question that. And you can see kind of similar things like in medicine, where people just do this really long, really grueling path, just to get back to, “Okay, now, I'm breakeven again, and now I can begin earning money.”

MATT: I think one of the one of the most interesting sort of social economic phenomena of the last 50 years or so is that, I would say, sort of in the 1950s and 60s (this is probably a UK lens, but I'm pretty sure the same is true in the US) this sort of broad basket of prestigious, socially respected careers didn't have that higher variance in economic outcomes. Yeah, for sure, finance was probably still better paid than medicine, but it wasn't a lot better paid than medicine. And you sort of see this particularly, (and this probably is a very British perspective, but) for example, who can afford to send their children to prestigious private schools? Half a century ago — it's obviously an extremely narrow group with respect to society — but the diversity of those career paths gave that as an economic option. It was pretty broad — doctors and local lawyers — all these careers that were prestigious. And at the time, almost as well paid as doing anything else. I think one thing that happened in the second half of the 20th century is that the gap between professions that were well paid and were extraordinarily well paid, became much, much larger. And so there became this bigger gap between prestige, which I think remained. I think being a doctor remains very prestigious. And economic outcomes, where being a hedge fund manager is not just a little bit better paid than being a doctor, it's a lot better paid. And the reason that I think this is relevant to your question is that, I do think we sort of ended up creating this. You could call it like ‘the precarity trap for ambitious individuals', where because there's a relatively small number of institutions, universities, early career jobs, etc. that sort of gate keep passage into this increasingly unequal elite, the need to buy into those positional status goods of being able to send your kids to those places (or whatever) seems to have become sort of more and more desirable to people. And as a result, they feel like they have to earn more and more money in order to be able to buy into those positional goods, but that sort of creates this weird distortive effect. And I would say I've spent a lot of the last 10 years trying to push against this, where you get people who have absolutely zero interest in finance feeling that they have to be investment bankers as they graduate from university. And there's sort of no good reason for it except it's become the sort of catch all, “Well, surely this is what I should do in order to make sure I become a member of the sort of upper middle class elite that I think I'm sort of destined to be.” I think that's wildly disruptive for society to feel like that the path into having the kind of life that you thought you wanted is so narrow. Obviously I'm biased in that I think a lot of these people should be entrepreneurs. But I think more broadly, there is a real cost to them.

SPENCER: So one question I would ask you which is relevant to what you're just saying is, what percentage of people do you think should be entrepreneurs? Because I wonder if you and I agree on this.

MATT: I don't have a single answer to that. But let me make a couple of observations. So one, I think almost certainly more people should be entrepreneurs than are today. But that does not mean anything like all people should be entrepreneurs.

SPENCER: Do we know what percentage of people are? I don't have statistics handy there.

MATT: I can give you some data on this but it's pretty misleading in that the way that this typically is measured is things like, people who are directors of limited companies in the UK, that they started or are founding directors. But of course, probably what you and I mean by entrepreneurship is people who start companies with the aspiration of those companies becoming large, which is very different from people who start companies as the sort of long-term alternative to employment. In the category that I'm talking about, it's comfortably below 1% of the population. Now, what do I think it should be? Well, I think there are two important things I would say about that. One, I think that it is unethical to encourage the general population to be entrepreneurs. I think all the evidence in the academic literature, but also in my experience is that for the median person, employment is a much better path than entrepreneurship. That's true, whatever you measure — whether you measure economic outcomes, whether you measure reports of life satisfaction — it is clear that for the medium person, employment is better. But what I'm interested in is the idea that for people with quite exceptional talents, people who might be in the top few percent by some measure of ability, actually, for them, the evidence is much less clear. In fact, for them the evidence suggests the opposite, that in expectation, entrepreneurship may be a much better path than employment for them. That's the first thing I would say. And what's interesting is — I've been saying that for about a decade and I think a lot of people thought it was potentially an obnoxious thing to say — that it was wrong in some profound almost metaphysical sense that entrepreneurship is something you are born to be. And the idea that these people that go work at Goldman Sachs or McKinsey could be entrepreneurs (like) violate some sort of law of the universe, that they're clearly not entrepreneurial, otherwise they never would have done that. I think what's really interesting is that over the last decade, a ton of evidence has come out that suggests that actually, a lot of these people would be great founders. And so, two of my favorite academic papers in the last few years, both basically looked one in the US one in China at the same phenomenon, which is, there's this sort of not quite natural experiment, but interesting discontinuity that can be exploited by researchers, which is that in times of recession, big financial institutions recruit fewer graduates coming out of universities, fewer undergrads, fewer MBAs. And so what the researchers do is they basically say, first, when that happens, do more people become founders, or this sort of target group become founders? And second, what are their outcomes? What happens to their companies? Now, I think conventional wisdom would be, “Yes, more people become founders. But no, they do worse than people that choose to become founders sort of more voluntarily (if you like) because they're not born entrepreneurs.” And what's interesting is that the opposite is true. Yes, more people become founders; that is right. But actually, the people that would, by hypothesis, otherwise have become bankers do better on average, than people that found companies in these non recession years. And I went through all the robustness exercises that the authors do, but their conclusion, which I share, is that actually probably, on average, in most ecosystems around the world (probably not actually in Silicon Valley), but on average, the typical person that gets a banking job out of university is higher skill than the typical person that starts company out of university. And so when those people shift into starting companies, lo and behold, there isn't anything special about entrepreneurship that is kind of orthogonal to scale. In fact, if high school people become entrepreneurs, they do better. And that suggests to me, not necessarily that vastly more people should be entrepreneurs, but probably vastly more of our most talented and ambitious people in society should be entrepreneurs.

SPENCER: That's really interesting. I feel though that we might be flattening skill a little bit here.

MATT: We definitely are.

SPENCER: Let's unpack this a little bit, because clearly, there's gonna be some people who are really talented, but they're not the right person to be an entrepreneur. Whereas other people might be talented in sort of the right way. So let's unpack a bit. Let's say someone is considering entrepreneurship. And let's say they have an accurate self-assessment, like they know their own strengths and weaknesses. What should they be looking at to help decide if they have the right strengths?

MATT: Here, I would come to like the selection criteria that we've created for EF over the last decade that we think are reasonably well-evidenced at this stage. I always say to people that they're quite boring, as in there's nothing very surprising there. But as you say, you need an accurate self-assessment of where you stand on them. So I do think the first one is: I think a lot of entrepreneurship is applied problem-solving. I think that sort of roughly approximates what most people mean when they say smart. I don't know why people find this like a sort of controversial idea, but I think in general, the best entrepreneurs are really smart. And if you are really smart, it's not a sign you should be an entrepreneur, but it's one of the filters we would use. I think it's important that entrepreneurs are very smart.

SPENCER: Are you basically talking about IQ there? Or are you talking about something different?

MATT: Yeah, I'm talking about like a generalized ability to solve complex problems, which yeah, I'm sort of wary of getting into the IQ debate, but yeah, broadly, the thing that most people mean when they say that person is really smart, I think is the thing that entrepreneurs need. I guess the reason I've been reticent is, I don't think that, for example, it would make sense for EF to ask a random sample of the population to take an IQ test, then just fund the top 2%. Equally, I'd be super surprised if having gone through our selection process, if we then did an IQ test, we didn't find that the vast majority of the people that we had chosen to fund were in the top handful percent for IQ. Maybe that's the way I would think about it. So I think that is important. I think the second thing is very important is some measure of resilience. Entrepreneurs need to be extremely determined. And one of the maybe obvious things to most listeners, but I think when you see it up close, it's just striking how true this is, is that there are so many points along the way in starting a company where the easier thing would be to give up. Almost by default, you would give up in the absence of feeling very strongly that you shouldn't. So people who are extremely determined and resilient, I think that's something where, if you feel you're in the top handful of the top few percent of the population for that, that's probably a really good sign as well.


SPENCER: One thing that I have noticed there, if you read the stories of a lot of successful companies, they very often will have one or more times when it really seemed like they're going to fail. And you think that most people in that situation would have given up. But they just kept persisting despite all the evidence that they're going to fail. And these are the successful companies, right?

MATT: Right. I cannot emphasize enough how true that is. And I think one of the reasons it's such an important message, particularly if there are any prospective founders listening, is that I think one false inference that some founders make is if they encounter one of these near-death experiences for the company in the first six to 12 months, and it's very easy to take that as a signal that either you are not good enough, or the idea is not good enough, or the company is not good enough in one of these ways. And actually, it's just definitely not always, definitely not conclusively true. I would say nearly every founder that I have worked with who I consider really exceptional has had one of these experiences. And it's not like something that you avoid by being a better founder. It's almost the opposite, in that you will encounter these moments. And one of the things that will determine whether or not you survive to be confounders is whether you give up.

SPENCER: Great. Do you want to go on the third trait?

MATT: Yes. The third thing we would look for is — and this is where I think your point about flattening skill is so important — we do find that, (and it's one of the things we look for) people that turn out to be exceptionally good founders usually have some functional skill at which they are exceptional. And that obviously, like some of those skills are going to be correlated with being with what we've called smart. But what I mean is, if you are going to build, for example, a deep tech company, you really want at least one of the founders to be exceptionally skilled in the tech. I mean, this sounds so banal, but it's actually really important. You really would want one of the founders to be exceptionally skilled in that area. Equally, if you're going to build a really sales-driven company, and you're the CEO, you would want some of the skills or characteristics that we most associate with great salespeople to be exceptionally strong, like this sort of rather tenuous thing we call charisma. A lot of great founders are really charismatic. Now, we don't say that every great founder needs to be charismatic. But one of the things we, Alice (my co-founder) and I, recently published a book called. “How to be a founder.” And one of the things we talk about in this is this idea of having an edge, having a personal competitive advantage. And I wouldn't say it's orthogonal to smart, but it's definitely not the same thing. And a lot of people who are generically smart and determined end up not being good founders, because they haven't really got this edge. So that would be the third thing I would really emphasize.

SPENCER: So it's sort of the thing that you just have an unfair advantage at, you're just so good at this thing, where it's like you're gonna be able to lean into that and get an advantage over your competition. But it feels like you also have to pick your idea carefully so that it actually exploits this advantage that you have.

MATT: 100%. That's absolutely right. So one of the core arguments of how to be a founder is that this idea of your edge is not only a useful tool for deciding whether to be a founder, but actually it should be core to the ideation process. And I think a mistake that a lot of very smart people who do have an edge make is almost discounting their edge and ideating from a blank sheet of paper, or ideating based on a chance encounter in a conversation. Like, someone said so and so was a hot area, so they sort of pursue that. At least in our experience with EF founders, you wildly increase your odds if you work in a space where you can exploit your edge.

SPENCER: And so what's the last trait you look for?

MATT: The last one is sort of the one which maybe gets to why you ask the question in the first place, which is about your comfort challenging the status quo. So I think one thing we struggle with a lot at EF is we get tens of thousands of applicants, and most of them will be pretty interesting. And a lot of them look like they will be good investment bankers, or management consultants, or Google employees or whatever. And as they said, my sort of own premise is that there's actually not that much difference that makes you successful at a traditional career versus as an entrepreneur. But I think that there is this thing that's arisen. Going back to the idea of variance dampening institutions, some people's preference for various dampening becomes very strong. And they actually struggle to succeed outside structured environments, they struggle to succeed if there isn't a mark scheme, or a sort of performance review grid to fill in. And so we look in every founder we fund for evidence that they've challenged the status quo, they've sort of found ways to succeed outside a structured environment. And I think that is probably the one thing that we would sort of say to people who were like, “Should I be an entrepreneur, or should I be a management consultant,” is to test that way yourself? Like, are you comfortable, actually taking risks outside the structured environment? Have you historically found yourself drawn to opportunities to have impact, even if it looks nothing like starting a company, but to have impact in ways that no one told you to do, no one asked you to do? That does seem to be very important, at least in our dataset.

SPENCER: Those are a great list of qualities. But I wonder about a couple others that I'm curious to get your thoughts on and see where you fit them in, or if you just think they're not essential. One is that entrepreneurship is such a stressful lifestyle, where you don't know if you're going to succeed. I assume, as described by people as “the world is going to punch you in the face over and over, and you don't know how many times it's gonna happen. And if you ever give up, you lose.” It feels to me just like for some people, that would just be torture. And I'm not sure it's captured in the list of four attributes you mentioned, like who would just hate that?

MATT: Yeah, I think it would be highly correlated with the resilience and determination point. But I agree. I guess the reason we like to phrase these quite broadly is because we think the prevailing cultural sentiment is to think that entrepreneurs are just weirdly different. But one of the weird phenomena about modern, particularly high prestige careers, is how much they're almost structured as deliberate hazing exercises. I was never an investment banker. But the stories I hear from first year investment bankers are pretty horrifying. It's not the same as entrepreneurship; you're not on the hook for the outcome in the same way. But the idea of just constantly feeling like what you're doing is not enough, and you're working 100-hour weeks, but still somehow you feel you're failing. And I think a lot of those characteristics that make people push through that — and by the way, I'm not saying necessarily people should push through that — are actually quite similar to what I think you need to push through as a founder.

SPENCER: Another one that comes to mind for me, and maybe you would link this to challenging the status quo, is that as an entrepreneur, you constantly have to make decisions, but you never really know what to do. You're like constantly don't have enough information and yet you have to do crucial decision making.

MATT: Yeah, isn't that absolutely right? Maybe we can come to like a little bit about — because this thing is relevant to this and other questions like — how we, in practice, evaluate entrepreneurs? I agree that comfort with ambiguity is absolutely key. And I do think that probably comes a little bit in the ‘challenging the status quo' like, “Are you willing to put yourselves in more ambiguous situations?” But one of the things that we have learned doing this for a long time is that there are some things you can evaluate pretty well in an interview. And there are some things you need to watch people work and do in order to be able to evaluate. And one of the (I think) benefits of the EF model, both for our investors but also for entrepreneurs, is that if you like the way it works in the way that it's very different from a startup accelerator — like Y Combinator or something — is that people join us pre company as individuals, and we fund them as individuals, and then work with them for three months to help them try to find a co-founder and try to develop a venture-backable idea. And what's interesting, I think both for the founder and for us and I think it's mutually advantageous, is both we know a lot more about the founder at the end of that three months than we did at the end of the interview, but also plausibly to your original question, I think they know awful lot more about themselves than they did at the start. And sometimes people conclude, “You know what, I just don't think this is the right path for me.” And again, obviously, I'm massively biased to think that EF is great, but I do think if we go back to the idea of institutions, I think one of the reasons that I'm really proud of EF as an institution is, entrepreneurship is one of the career paths (if you like) that is really hard to test without doing. I'd like to think that by providing a semi-structured environment and peers and some funding, we sort of give people a chance to figure out what is the right path for me before they commit to other paths that will take them away from that possibly for a very long time.

SPENCER: Yeah, getting them to try in a not-so-constrained environment to make sure it's the right path seems like a really good idea. And also, of course, gives you lots more information.

MATT: Yeah, exactly.

SPENCER: Going back to the idea of these career paths that are hard but the path is sort of laid out — like investment banking, medicine, law — nobody would say those are easy, right? They are actually really difficult. But the path is very clear, right? You can buy a book on how to do the career path. And you know if you hit certain milestones that are laid out, it's very likely you'll succeed. And you compare that (to make a metaphor) like, I feel like that's kind of like climbing Mount Everest, right? Climbing Mount Everest is really difficult, but many people do it every year. It's very clear how to climb Mount Everest, even though it's really difficult. And I would contrast that with something like climbing the Dawn Wall. Tommy Caldwell was the first person to ever climb it. Nobody knew if it was even possible, let alone how to do it. And he had to figure out how on earth would you climb this thing. And I feel like that's what entrepreneurship is like; it's not necessarily that it's that much harder, though probably I think it is harder, but it's really just like there's no path, like you're just in the middle of the forest being like, “Oh, which way to go?”

MATT: Yeah, I think that's exactly right. I think one of the things that's quite strange in a way is that, arguably one of the big benefits of being an entrepreneur today versus (say) 25 years ago is that maybe the first stages of the path have become a little bit more like climbing Mount Everest, as in, I think there is more codified knowledge that is probably useful to most entrepreneurs. I think the idea of customer development and testing, testing and validating ideas, I think there's a lot of the open sourcing of how to raise capital. This has been a really good thing. But I think the bit that is irreducible is it's almost like climbing to base camp, and then it turns out the mountain at the top, no one has climbed before. And so I think there's like a greater legibility of some of the common building blocks of entrepreneurship. And I'd like to think that being part of a community, like EF, is maybe one way to acquire those. But I think you're right that the idiosyncratic risk in each startup is like climbing a mountain that no one's ever climbed before.

SPENCER: Another trait that I think about, and I think this relates to the resilience and extreme determination when you mentioned but is a bit different, is that an entrepreneur has to be in this funny place where they're extremely determined to accomplish their long term goals, and they're not going to give up no matter what. But at the same time, they have to constantly be paying attention to which tactics are working and which are not and be adapting rapidly. So it's like you're never giving up and you're incredibly stubborn about the long term goal, but you're incredibly flexible about how you're getting there and constantly adjusting your plans. And I think that's an interesting dichotomy.

MATT: Yeah, I think it is. I think it's exactly right. And I think one of the hardest questions to answer that we get asked a lot is sort of, “How do I balance determination and sort of knowing when to stop, or more positively, knowing when to pivot?” But I think you actually gave what I believe is the answer in your question, which is, I think it's really about whether you are attuned to signal and what the relevant signal will be for you in your journey. Another way to describe the same phenomenon that you just described is, you're going to be absolutely pelted with noise as a founder. There are going to be these events that you have to interpret every day, and most of it is noise. And most of it, even things that feel really fatal, is noise. And you just shouldn't put a lot of weight on it. But every so often, there's something really high signal and you have to be attuned to it, because it might tell you that you're going in the wrong direction. And I think a lot of the skill, as opposed to the talent (if you like) of being a founder is in getting really, really good at telling the signal from the noise. Almost always a risk of oversimplifying that the signal is about what it is that the people that are or should be your customers are telling you versus what is everything else. But that is a massive simplification. One of the many reasons entrepreneurship is hard is because sometimes your customers are wrong, sometimes the way you frame the decision for your customers means that the answers you're getting are not useful to you. The thing that both great entrepreneurs and I think great advisors to entrepreneurs are good at is this highly contextual detection of signal in the noise.

SPENCER: So there's one more trait I want to ask you about before we pivot our conversation, so to speak, which is this trait of getting people to follow what you're doing, where some people, they are doing a thing and everyone just wants to jump on board. They're like, “I want to be on your team; I want to give you money; I want to follow your work.” And other people don't have that sort of like (maybe you call it like) inspiringness or makes people want to be part of what you're doing. Do you want to elaborate on that a little bit?

MATT: Yeah, I just realized Alice would probably kill me because I'm pretty sure...I just remember there is a criteria in our selection process that we call followership. A relatively hard question is, if you want to fund people who are at the start of their career — as we do, generally, not exclusively; but most people at EF are in the first few years of their career — and you want to know whether they can be an inspiring leader to people in the ways you describe, what should you take as signal? Like, they probably have not run companies before. But then equally, do you care as they've been like captain of the sports team? Maybe, maybe not. And actually, I think a big part of what we're selecting for is less like formal leadership positions, and more what are the examples that a person has where either very talented or very senior or very busy people have decided to pay attention in some way. And we found that that's actually a pretty good proxy for what I think it is you're describing, but it might look absolutely nothing like an entrepreneurial experience. But we do think that great founders do seem to attract talent around them, whatever they do. That doesn't just mean people that come to work for them, but as you said, their investors, their advisors, even their customers. I think you're right, there is something about that. Now, I suspect that it can be sort of disaggregated into not just the other criteria that we've talked about, but I suspect, there is some correlation with those. But I agree, it's very striking when you see it.

SPENCER: Alright, so let's change topics a little bit. Let's talk about ambition more broadly. And you get some interesting ideas in sort of how ambition has changed throughout time.

MATT: This idea that the most ambitious people in the world should at least consider entrepreneurship, I think is weird in that, I think, in the circles that I mainly spend time with, it's probably not a very controversial idea, but I think in the world as a whole, it is. And as you said, I have spent a lot of time thinking about this, and how the sort of most popular career paths for ambitious people have changed through time. I suppose the reason I've dedicated the last 10 years of my life to building Entrepreneur First with Alice is that I really believe that we're entering an era where technology entrepreneurship is going to become more and more the default career path for the most ambitious people. The main reason for that, for me anyway, is actually a historical reason, which is something like, even if you zoom out over a thousand year period, it's pretty clear that in every era, ambitious people seek leverage. They seek ways to maximize the amount of impact that they can have. And in each place and time through history, you find that there are different (what I call) technologies of ambition: the technologies that allow people to increase the scale of the impact they have in the world. Now what those technologies are has changed wildly over time. And to give a very, very brief, stylized history, I like to think of perhaps the first really great technology of ambition of the modern era was literacy. Before literacy, before people could read and write, you really couldn't have a lot of impact outside your immediate physical person. What you could do is limited by who you saw, what you could touch, what you could move. And then literacy suddenly creates this opportunity for scale. Because you can write something down and people who are not physically proximate to you can read it and take some action based on it. And I think there's no surprise that (if you like) what I would consider the birth of the modern career, which is sort of open to people (I don't want to say on merit, but like) more on merit than pure who your dad was, it coincides with the rise of literacy. And there's lots of really interesting examples. I'm a real medieval history geek. There's a lot of interesting examples from medieval England. Maybe the most famous one is this guy, Thomas Wolsey, who was born the son of a butcher in this tiny town in the East of England became the most powerful person after the king in England. Cardinal Wolsey was responsible for, among other things, the start of the English reformation. But how did he get there? Well, basically, by exploiting this technology of ambition that we call literacy. He was bright and managed to get to go to a Cathedral School. And then he went to (sort of brand new at the time) Oxford college, and he harnessed literacy to become a super powerful person. Now, literacy is great, but over time, it becomes table stakes. And so new technologies of ambition emerge. And I would argue that, if you were to zoom into (say) the 19th century or late 18th century, you could say something like, military command became this great technology of ambition. This is what allowed Napoleon to become — again from not quite humble but modest origins — the most powerful person in Europe. And he did that because suddenly you have this technology — modern military command — that allows you to say a word in Paris and armies move across the world. And fast forward again, the 20th century is really about the emergence of this new technology of ambition, which is sort of finance, which is you write a check in New York and people and materials all over the world sort of snapped to attention. And finance looked, in many ways, like the ultimate technology of ambition for ambitious people. But then you look to the last 20 years, and it's pretty clear to me that technology entrepreneurship has given ambitious people a degree of leverage that would have been unthinkable to any medieval Cardinal or early modern general or even 20th century financier, which is, if you can build a product from your bedroom in a permissionless way that can reach hundreds of millions or billions of people, that is an opportunity for the most ambitious people that I think is unprecedented. And so that's how I think about why we're going to see more and more of the world's most ambitious people attracted to this sort of limitless potential of entrepreneurship.

SPENCER: It's really interesting to think about these leverage technologies changing over time. And it makes me wonder what the next one will be, right? Is technology entrepreneurship going to last a really, really long time? Or could there be something even after that?

MATT: Yeah, it's really hard to predict. I spent a bit of time thinking about that question. And I think the thing is that it's any one of those moments, it would have been really hard to predict what's next because it's so driven by social and cultural and technological change. But I think one sort of provocative idea is can tell us in two very different directions. So one would be that if we do become, ultimately, an interplanetary species, then technologies that allow influence to be projected over much greater distances will become much more important. Let's imagine a sort of Star Wars-like future, where you have a confederation of planets. You can imagine, being able to influence that would be very attractive to ambitious people. But I think the more prosaic one — which is sort of really interesting and I guess getting a lot of hype right now — is sort of how will artificial intelligence change the way that we think about human leverage? What does artificial intelligence make cheap, and what does it make scarce? And my guess is that, as we're seeing, at least at the time of recording, people are very excited about language generation and large language models. If that technology continues to improve, I think you probably end up with a world where just distribution becomes what scarce relative to the ability to create, for example. I suspect that one thing we are already observing in the EF applicant pool is how many of our applicants have made big investments in building personal distribution networks through social media or newsletters or whatever, well in advance to starting a company. And I suspect that that sort of idea of building an audience is going to become more and more important for ambitious people.

SPENCER: Yeah, it's really interesting. You can also imagine as AIs get better and better, that AI is making more and more communication on behalf of people. Like, I wonder how many years away we are from being able to train an AI on your own output, and then just have it generate more stuff that's similar to what you would generate?

MATT: I would say, probably not measured in the plural years. I don't know if you saw this example that was moderately viral on Twitter this week of someone who trained a model on diaries that their younger self had written, and then had a conversation with their model trained on their younger self. And it's quite — I have to say — it was quite moving and beautiful as a thing. But I think you're right to say, well, we're probably not that far off of being able to build low fidelity versions of ourselves, at least when it comes to language generation, which obviously is a small part of what it means to be human. But I think you're absolutely right.


SPENCER: So before we wrap up, I wanted to at least have one more topic. Let's talk about counterfactual impact. I think a really interesting question comes up when you're thinking about either being a founder of a company or an investor, of whether you're really changing the world or you're just doing something slightly before someone else did it, or you're doing something that would have happened anyway. A lot of times investors are trying to invest in things that they think will be successful even if they don't invest. They want to invest in the thing that's close to guaranteed success as possible. Obviously, it is never guaranteed in investing, but if the company would fail without their investment, that's actually unappealing to them, right? But that means that they may be having less counterfactual impact. They may be actually changing the world less, which is for someone who wants to improve the world, that's not a good thing. Or if you're a founder and you're starting a company, there's your question, “Well, maybe if you didn't do it, someone else would have done it six months later. And so you're not really providing a new service to the world, you're just providing it slightly earlier than it would have been provided.” So yeah, I'd love to hear your thoughts on this.

MATT: I think this is a really interesting question and a risk of alienating your VC listeners (many of my best friends are VCs). One of the things that has never appealed to me about being a sort of traditional venture capitalist — I guess, technically, I am a venture capitalist, but a traditional one — is that I have often worried that, broadly, what you're really trying to do is win allocation in deals, rather than form companies that no one else will fund. In other words, almost by hypothesis, the deals you want to do are the ones that everyone else wants to do, too. (Not everyone else, but almost always more than one other person.) And as a result, what you're really doing is deciding who gets the slice of the pie, not actually changing it. And at least to me, counterfactual impact is really important. I think it's one of my core motivations in life, actually. And so, one of the things I love about what we do at Entrepreneur First is, I think it's quite plausible that a lot of the companies we fund otherwise wouldn't exist. They exist because the founders meet at EF. And going back to the early conversation about founder edge, I think a lot of the time the ideas come out of a meeting of two people with distinct edges who would never otherwise meet. And the thing that comes into being is something that probably in the wild wouldn't exist. You could say that's the very sort of self-aggrandizing version of the story. Because, of course, it is certainly true that one of the weird phenomena that you notice very quickly when you've been a VC, even for a short time, is that whenever an idea gets to be right (if you like), whenever the conditions for an idea to succeed are obtained, very quickly, you see lots of companies trying to do the same thing. And one of the sort of mildly amusing things about reading thousands of application forms is that you don't hear about an idea at all. And then suddenly, in the same cycle, you got a dozen people, all of whom have had the same idea, and all of whom insist that no one else in the world is working on it. So, you could argue that counterfactual impact's hard because there's something just overdetermined about what makes companies come into being. The thing I've more recently come to believe is that I think there are some ideas that actually are accelerated dramatically by a single founder or set of founders going after them. I think it's interesting through history that there are some ideas that weirdly seem to take far too long to emerge. My friend, Jason Crawford, has a great post on why it took so long for the bicycle to be invented. And I really recommend everyone read that. It makes you wonder what else has taken too long. But the more important thing that I would want to highlight is that I think often, particularly with very powerful technologies or very important companies, the character and values of the person that starts them actually turn out to be really, really important. We've just been talking about AI, I'm a little bit biased because he's a friend and an investor in EF, but I think it's a really good thing for the world that Demis Hassabis started DeepMind, which I still think is probably the company/organization most likely to develop general AI. And it was really good that it was him and not some pure power-maximizing psychopath that started DeepMind. Because I think that AI is going to be generationally a more important technology. And the values that we embed in it and the way it's deployed and distributed are going to really, really matter for the species. Now, you could say that's a unique example, but I actually don't think it is. It is certainly true that it is only relevant if companies become very successful. But without naming names, if you were to look at the five biggest kinds of tech successes of the last 20 years, I would say that the character of the people that started them has ended up having really great importance and influence.

SPENCER: Right. So it's not just whether the company comes into existence or how quickly, it's also who is the person running it and sort of what are their values and things like that?

MATT: Yeah, exactly. And I think that really matters. And so I think, particularly if you're very ambitious, and particularly if you're ambitious and you have an idea that you think is important, it could be very important that you are the person that starts it, rather than just saying, “Oh, well, these things are overdetermined; it will happen.” There's sort of a link here to a book that we recommend our cohort read. It is Peter Thiel's “Zero to one,” which I think is very good. He has this idea of definite optimism versus indefinite optimism, where indefinite optimism is a somewhat complacent belief that the world will get better. And definite optimism is, “and it will get better in these ways because of me, and all because of us.” And although I disagree with Peter Thiel on a lot of things, I think we do need more definite optimism in the world. And that's one way in which you can have a real counterfactual impact, I think.

SPENCER: Going back to the question of counterfactual impact of VCs, I think an interesting way to look at what a VC (venture capitalists) is doing is how do they actually excel and make money. And there's only a few things that they can do to have an edge, right? One of them is that they can hear about more deals than other people, or hear about more of the better deals (let's say), or more promising deals. I've seen them do this in two ways successfully. The first is where they're just really, really well known. Like Founders Fund, just pure deals fund. They're just so well known, everyone wants to do a deal with them. And so they hear about the best deal flow. The other way is where they specialize. Let's say they're just in healthcare, and so they try to see every healthcare deal happening, and they just get known for being a specialist in that area. So the first thing is just hearing about deals, they can hear about more of them. The second is they can be better at selecting deals. Like, among all the deal flow they're seeing, they can maybe have an edge at picking the right deals to be part of. And I feel like that's a tricky one, because sort of everyone thinks that they're better than average at that. But it's really hard to know. And presumably, there are some people that are better, maybe they have more experience, maybe they have just better intuitions, or maybe they have some kind of framework they apply. The third piece there is that they can help their investment companies succeed. They can give them advice, they can give them intros. Some VC firms offer more specialty services, like maybe they help them with marketing, or they help them with sales or whatever. And then the final way I think they can have an edge is they can be smart about reinvesting. So if they have a deal that's promising that they got an early, they can have a strategy where they'll increase their investment. They have a way of assessing, “Oh, is this a good one to put more money with.” And those are sort of the four main ways I've seen VCs have an edge. Is there anything you'd add to that list or does that seem right to you?

MATT: Yeah, that's right. I think we talk about that a lot at EF and we always say it's like, what do you see, what do you pick, what do you win, and how do you help? So I guess a very similar view. And I think probably (this will not be a surprise to anyone listening who knows anything about VC), but I think something that's very surprising to outsiders is that the thing that seems most intuitively important, like what do you pick, is actually I don't think a huge driver of value for most VCs. It's true that occasionally you get deals with companies that turn out to be wildly successful that are very hard to get funded. But even though canonical examples of that, like Airbnb — the famous sort of Brian Chesky thread of all the rejection emails they got — they still had more than one fund that wanted to invest. So yeah, picking matters. But I think, more and more, winning matters. As the space has become more competitive, I think more and more, it feels that the ability to actually get access to the companies you want to invest in is more important than being able to pick the ones that you want to invest in.

SPENCER: Right. And in the hottest deals, it may be just, will the company take your money and instead of someone else's, right? Can you differentiate?

MATT: One interesting phenomenon — which I don't know whether there's good recent data on, but it's certainly been true historically — is that compared to other asset classes, venture capital has a lot of return persistence. If you're a top quartile firm (kind of) last decade, it doesn't guarantee at all that you will be a top quartile firm next decade, but you're much more likely to be than someone that isn't currently a top quartile firm. And that's really because (I did not come up with this line, but I like it a lot, which is) venture capital is the only asset class where the asset chooses the manager rather than the manager choosing the asset. And as a result, the brand isn't permanent and impregnable, but it can be very persistent. To affirm approximation, people want Sequoia's money because of every company that Sequoia has invested in since the 70s, there's like a genealogy of success that makes people want their money.

SPENCER: Yeah, that's really interesting. It's also a bit surprising that there's so much persistence to me, because it's such an outlier driven investment strategy, right? Like, basically, you're looking for your one or two best picks to drive the fund performance. And so you might think that it's gonna be a lot of luck. Like, if you just didn't have that one company in it, you would have had a shit return, but like you had that one really homerun, and then you did well. So you might think, “Well, that's gonna make it very luck driven.” So it's quite interesting that you think there's a lot of persistence.

MATT: I think it will probably be more true if every investor could invest in every company they wanted to invest in. So all you need for returns persistence, even in an outlier driven model, is that you can somewhat detect the outliers, and people want your money. And so because approximately everyone that might be an outlier ends up pitching Sequoia, Sequoia only needs to be pretty good at picking in order to make sure that they get a lock in. But not only that, it's the repeated game for a fund like Sequoia or Andreessen Horowitz that can invest in most or every stage. They can often come in later, if they miss something, and people still want their money. Zoom would be a good example. Sequoia came in relatively late to Zoom, but ended up with a fantastic return. And so I think structurally, there's a lot of advantages to being one of these big storage firms.

SPENCER: Yeah, that makes sense. And then if we go back to thinking about counterfactual impact, and you kind of break down how VCs make money, most of those ways that they get an edge don't actually give them the ability to have much counterfactual impact, right?

MATT: I totally agree with that.

SPENCER: It seems like the biggest way they can have counterfactual impact is if they really can help the firm, right? Like if by investing, they're able to add other value added services, like give them strategic advice, or strategic interests or things like that. That really (maybe) could make a difference. Or, in some cases, if they're able to give them a lot more money maybe than they could have gotten otherwise. ]But again, on the hottest deals, maybe it's just more about getting into the round and sort of oversubscribed anyway.

MATT: And that's exactly right.

SPENCER: One more question I have before we wrap up that I'm curious about; So you mentioned how in different rounds of Entrepreneurs First, you'll get people applying with similar ideas, maybe it's sort of in the air and a hot idea right now. But I'm curious, are there ideas that you see sort of again and again, you've seen for years, where everyone sort of has the same idea and you kind of roll your eyes like, “Oh, not again, everyone seems to have the same idea and it never works.” I'm just curious, sometimes people call these tarpit ideas—these ideas that people are drawn to, but they never actually succeeded.

MATT: Absolutely. One of the reasons that we came up with this framework around people's edge was to try and help people, particularly in early in their career, have more distinctive ideas. Because I think a lot of these tarpit ideas (or whatever you want to call them) come out of people being drawn to ideas that they experience mildly and infrequently as consumers rather than deeply and painfully as sort of producers. And so, everyone eats and so there's a lot of food ideas. Everyone wants to — well, at some stage of life, I've been married for 10 years, so I don't — but everyone wants to go on dates. So you get a lot of people who want to start dating apps. Everyone likes music. So as that person you get a lot of music ideas. And so the idea of edge as a framework is to sort of help steer people to places where they have a distinct advantage, rather than I like music. Well, same with seven or eight billion other people. That said, the sort of contrarian takers, these ideas are so attractive because they're so universal. And so, for example, it wouldn't surprise me, if despite the fact that I groan every time I see a dating app idea. It wouldn't surprise me if there is a breakout new dating app in the next decade that becomes worth billions and billions of dollars.

SPENCER: And we can almost guarantee there will be, right?

MATT: There will be, right. And yet any given idea in this space feels like, “Oh, you gotta be kidding. Surely not.” But this is why edge is so important, because I do see you need a right to win.

SPENCER: So someone came in with a certain kind of edge around dating, you could see like, “Oh, wait, maybe this person really has a secret sauce around dating. And that could make them much more likely to make that billion dollar idea.” But if you're just sort of a random person who likes going on dates, right?

MATT: Absolutely. The original founding or like motivating joke for the idea of edge was, we used to meet a lot of people be like, “Hey, I've got a PhD in quantum computing, and I really want to start dating apps.” It's like, “Well, cool. But you have a very clear edge as well.” But I often say when I give talks to groups of university students, if you have an idea that is one of these tarpit ideas, but you know that you have this super granular, deep, contrarian understanding of the problem, and how it can be done differently, then I want to hear from you, because I actually think there's a lot of power in that.

SPENCER: Yeah, that makes sense. Well, actually, I have a few ideas that people have proposed or tarpit ideas, or that I suspect might be. I'm curious to run by and see whether you agree. I actually heard people at Y Combinator say this, that there's this tarpit idea around recommendation systems like, “Oh, well, I just wish there was a place that would tell me exactly where to go eat, or to tell me exactly what movie to watch, or this kind of thing.”

MATT: Probably my favorite ever startup essay, or maybe the essay, at least that I recommend most people, was written by one of the partners of Andreessen Horowitz. I think either around or before the time he joined Andreessen Horowitz, Chris Dixon, wrote this essay called the idea maze. And in the idea maze, what he argues is that, in all of these ideas that have been tried many, many times, what both founders and investors come to see is that a little like a maze through a number of key decision points that you have to make, or rather that you face on the way we have to make an important call about which way to go. And although the range of ways to go is very large, there are probably going to be a dozen core questions that you have to answer. And the most important thing as a founder entering one of these like tarpit ideas is to really understand the idea base, like what are the decisions that previous founders in the space have made? And how did those decisions yield different outcomes? Now, I think the tempting thing is to do one of those like slightly bullshit slides that you see in a lot of funding decks, where people list out a bunch of features and have, lo and behold, their idea ticks all the boxes and the other competitors don't have all the same features, and therefore you're gonna win. That's not what I'm talking about. I'm talking about having a really deep understanding of why all those event apps fail? Like, what were the decisions that they made? And what are the different decisions along the idea maze that you're gonna make? So, I think it's wise in general, that if you are encountering or thinking about a tarpit idea for the first time, probably the simplest heuristic is don't move on, use your edge. But I think if it's an idea that you just keep coming back to, the key is to actually understand why so many people have tried and failed here? What are the decisions then made? Do you have a genuinely distinctive way to navigate the idea maze?

SPENCER: Yeah, I think that's such a good point. And there's this phrase I like to use, which is ‘value in plain sight', where there's some big pile of value that clearly has not been tapped. But sort of like it's not a secret, right? The value sitting there, but then the fact that the value sitting there should make us very concerned about why nobody has taken it yet. So an example of this would be like, why are so many people unhappy dating? Clearly, there's something broken here. Women constantly complain that they get a lot of low quality messages on dating apps. Men complain that they don't get responses at all. This is not a secret, right? We know that there's like hundreds of companies that have tried to solve this. So the fact that there's this big pile of value and hundreds of those companies, it really points to this idea like, “Oh, wait, there must be something really tricky about getting that value.” But that doesn't mean you shouldn't try, it means you should be incredibly careful. It's like you're Indiana Jones. You know, there's like 50 movie traps?

MATT: Exactly. And that's a great way of thinking about it. It's not, don't do it. But absolutely don't do it thinking that you're just smarter than those other people. And that they didn't see the skeletons. They probably did. As you said, the weird thing is it's almost inevitable that someone will come up with these things. It's just that many more people will try. And so just making sure that you really have that map (if you like) of the maze. It's so important.

SPENCER: Yeah, another one of these ideas that I'm curious about if you see this again and again. It's like, let's make a new social media app, but we're gonna correct the things that people hate about social media, like that it's just optimizing for time on site. I don't know, I've just seen tons of people kind of pitch this, and it's like, “Hmm, I don't know.” It feels like a structural disadvantage. Like, if anything, you're just saying, ‘We're gonna make a social media site, but we're gonna make it less compelling for people to use.” Now, maybe that's good for society. That sounds great. But that seems like a competitive disadvantage.

MATT: That's sort of like a special case of a more general thing, where it's like, there is this feature of existing solutions, which we don't like, but turns out to be absolutely core to the ability to make them viable businesses. What happens if we strip out that feature? Well, guess what, it looks good in the short term and pretty horrible in the long term. But again, the great thing about startups, the intellectually endlessly exciting thing about them, is that generalizations like median outcomes are a poor guide. So, in general, I think that's something founders should really avoid—this, like, “Oh, what if we just tweak this one feature that seems completely core to make it a viable business.” Generally avoid that, except that relaxing one assumption that everyone thought was integral and turns out to be incidental can be a really good path to finding a really big idea. I think the key is just making sure that you haven't taken out the sort of foundation stone that makes the whole thing stack up at all. As you say, time on site, sadly, probably is one of those. But again, it wouldn't surprise me if someone did come up with a different model. I mean, again, it's too soon to know if it's a success but the phenomenon of BeReal, which is probably the hottest new-ish social media thing, is sort of in a way that anti time on site mechanisms. So, it's hard to say. But in general, I think you're right that you should be wary of that impulse.

SPENCER: Matt, thank you so much for coming on. This was a great conversation.

MATT: Spencer, thanks so much for having me. I really enjoyed it.


JOSH: A listener asks, "Who are some very wise or enlightened thinkers that we should learn from but who aren't very famous?"

SPENCER: Three people that come to mind that I really respect their thinking who are not that well known are Amanda Askell, Luke Melhauser, and Julia Galef. I mean, Julia is definitely the best known of the three. Quite a number of people know who she is, but she's not famous per se. I think they're all very good thinkers, and I recommend following them on social media if you don't and reading things they've written.




Click here to return to the list of all episodes.


Sign up to receive one helpful idea and one brand-new podcast episode each week!

Contact Us

We'd love to hear from you! To give us your feedback on the podcast, or to tell us about how the ideas from the podcast have impacted you, send us an email at:

Or connect with us on social media: