Patrick Collison and his brother John conceived of financial services company Stripe as something that they thought should already exist.
On the way back from a UC Berkeley-hosted startup event in 2009, John, who is president of the company, suggested that they turn their idea for an “easy way to move money online” into a reality by creating their first prototype. Now, Stripe is used by millions of businesses across almost 50 countries, from startups to global enterprises like Ford and Amazon, having reached a $1 trillion in total payment volume.
Co-founder and CEO Patrick Collison shared his journey founding Stripe and the company’s mission to “grow the GDP of the internet” at a recent Dean’s Speaker Series, co-hosted by the Berkeley Center for Workplace Culture and Innovation. Professors Jennifer Chatman and Sameer Srivastava interviewed Collison in a fireside chat format.
Born and raised in rural Ireland, Collison grew up without a consistent internet connection, only reading about it in books. At age 13, he made his first pitch to his parents for satellite internet. It was from here that he discovered programming, planting the first seeds for one of the most successful software-as-a-service companies today.
But the journey from startup to Fortune and Global 500 wasn’t always smooth. Between difficulty finding growth to feeling like he was misallocating time, Collison reflected that the first few years were a time of trial and error. He noted, however, that if the idea and the market are good, startups can actually be fairly resilient as management learns along the way.
“I think it’s much better to be right than to be consistent. And I think somehow getting yourself into the psychological frame where that’s OK, and you can swiftly recognize, ‘well, I tried this, it’s really not working, let’s do something else,’” Collison said. “As long as you get into that mindset, I think a lot of the actual errors themselves are recoverable.”
With the success of the company dependent on handling people’s money, Collison likewise stressed the importance of ensuring a meticulous and diligent atmosphere at Stripe.
“There’s this culture at Stripe of just really prizing the small details. And we talk a lot about craftsmanship and rigor and abstractions that can endure over decades and sort of really getting those right. So I would say precision is pretty deeply embedded,” he said.
Stripe is also known for its early adoption of a writing-oriented culture. An avid reader and self-described “misanthropic introvert,” Collison said this came about partially as an accident. But he likewise reflected on its benefits, citing written work as valuable for its ability to be reflected and improved upon.
In 2020, the company launched Stripe Climate, which allows users to direct a portion of their revenue toward scaling carbon removal technologies, leading to the implementation of the second-ever large-scale advanced market commitment (AMC) for carbon removal.
The company’s mission to do good hasn’t stopped there. Believing access to the internet is crucial for global development, Collison stressed the role Stripe has on aglobal scale. Citing his own experience witnessing the rapid expansion of Ireland’s economy, he stressed the “moral importance” of economic growth for parts of the world that have been left behind. .
“I think the internet is one of the most important technologies ever created with respect to the enablement of global development,” he said. “Stripe’s mission is to grow the GDP of the internet.”
Read the full transcript
– Good afternoon. It doesn’t seem like it’s on. Is it good? Good afternoon. Welcome to our Dean’s Speaker Series, co-hosted by the Berkeley Center for Workplace Culture and Innovation. My name is Jenny Chapman. I’m the associate dean here at Haas. Dean Harrison is actually traveling on school business, kind of around the world right now. She apologizes that she can’t be here, but we are here, and I’m so absolutely thrilled to introduce our guest today, Patrick Collison. Patrick has been an entrepreneur, I’m guessing since the day you were born. He created iPhone apps with his brother John and his teams, founding Auctomatic while at MIT, and co-founding his biggest venture Stripe, which we’re going to talk a lot about. Stripe was born when Patrick and John looked for a payment platform but couldn’t find all of the features they thought would be important. And Stripe debuted in 2010 and grew exponentially because the product is really simple for businesses to implement. Of course, the back end is anything but simple, but it’s easy on the front end for customers. Through Patrick’s leadership, Stripe has reached, listen to this, $1 trillion in total payment volume. So it’s really becoming a dominant methodology. And today—
– You can’t assume the causality, though, there. You can’t say, like, because of my leadership. It could be despite my leadership.
– I’m sure that that’s correct, but we will talk about the attribution, I’m sure. You will deflect it, and we will heap it on. So today, millions of businesses, from hypergrowth startups to global enterprises like Ford and Amazon, use Stripe to accept payments and payouts, manage complex business online. Patrick, we were back in our green room here, and Patrick was telling us about his biomedical research foundation, the Arc Institute, that he co-founded in 2021. Fascinating, fascinating work. Through the Arc Institute, Patrick and his co-founders are pioneering new model research in partnership with Stanford, UC San Francisco, and UC Berkeley, and hopes to enable passionate biomedical investigations to study and address complex diseases, which gives us all great comfort to know that brilliant minds are researching some of the most difficult health challenges that we face. So Patrick, we’re incredibly grateful to have you here today, to impart your learnings on our student body.
– No, thanks for having me. And it’s a particular honor to be here because the first time I tried to come here, you guys rejected me. So it’s… No, look, I might have rejected me as well.
– So did you land at your safety school, MIT?
– My first ever trip to America was to the Bay Area to… And I visited Stanford and Berkeley, and those are my first, literally, kind of two experiences of the U.S., and I just assumed all of the U.S. is like this. And so then, I, yes, decided to apply to college here, but I couldn’t get further than the East Coast.
– Oh my gosh. Well, that was a miss on our part. I can’t say it would be the first or last one, but… Anyway, let me do two more housekeeping things, and then, we can properly welcome you. The first housekeeping issue is: You notice that you have note cards on your chairs, and a pencil. And it’s not for keeping your golf score. But please, if you have questions, please note them on the card. We have collectors, Sarah and Audrey, who will be collecting them throughout the session and bringing them up here. And at around 1:15, 1:10/1:15, we’re going to start answering questions from the audience. So please, be sure to capture those questions. And second, I want to introduce my colleague, Sameer Srivastava, my partner in crime. We are the co-founders of the Berkeley Center for Workplace Culture and Innovation. And many of you probably have had Sameer for the power and politics class. So Sameer, welcome too. So let’s give now a proper welcome to Patrick if we might. Thank you so much for coming up. So let’s start with a kind of softball question, which is, if you could walk us through your kind of career progression, I think it would be of great interest to our students.
– Oh gosh. Alright, well you already heard an important detail in this, but, so I grew up in very rural Ireland, and our house was kind of far from the phone exchange, so it’s more detailed maybe than the career than you want, but you can push fast-forward at any any point. But just the relevant kind of consequence of that was, we couldn’t really get a proper internet connection. And so, I first learned about the internet by reading books and borrowed from the library, and to think, “Wow, this internet thing sounds great.” And there was no TikTok or anything. And then, actually, the first pitch I ever wrote was to my parents when I was 13, trying to convince them to get this kind of pre-Starlink satellite internet connection from Germany, and very, very graciously, and I don’t know, forbearing exceeded, and we got it, and it was 100 euros a month or something, which was a big deal. So once we got that, then, I discovered programming, and kind of fell down that rabbit hole. And then, when I went to MIT, and maybe you guys have had kind of versions of this experience, but I considered myself pretty good at math and physics and sort of science and things like that when I was in high school. And so, MIT had three different intro kind of freshman physics classes. And I thought, with great ambition and virtue, that like I’d go take the hard one, and I did, and it was hard. And they had this kind of particular midterm, and that, I guess they use this kind of weed out, I don’t know, the something… And I remember the average score on that midterm was like, a 43 or something. And I was feeling, initially, I don’t know, well kind of mixed feelings, but somewhat proud I’d scored like a 70 or something like that. And then, I saw that this… I don’t know if this is good for one’s psychological well-being, and maybe now this policy has been reconsidered in the intervening years, but at the time, they published everyone’s score publicly. And then, I saw that this guy, Yufei Zhao, had gotten like a 97. And that was like a big moment in my life where, to be clear, I’ve never interacted with Yufei Zhao, I don’t think he’s heard of me.
– But you certainly remember his name, don’t you?
– He is large in my life. And I thought, well, “If Yufei Zhao exists and can be so spectacularly better than me at physics, well, maybe I shouldn’t become a physicist. Comparative advantage, Ricardo, the whole thing. So I did some soul-searching, and by the way, I looked him up, and he’s done very well now in math and physics. So, I think we… And he now trains MIT Putnam team, and they’re doing… Anyway, so he found his thing.
– He didn’t found Stripe.
– Well, look, I think he’s not doing… I think Yufei Zhao has self-actualized with tremendous effectiveness.
– You both have.
– But anyways, so… But I done all this programming in high school, and so… And then, right around the time of that midterm, this was fall of 2009. Sorry, fall of 2006, excuse me. Reddit was purchased by Conde Nast for I think $12 million. And I knew the Reddit guys a little bit because YC started out in Cambridge, and they were based there in Cambridge, and they were just like normal people, or so I thought, and the idea that they’d just sold like a website, like anyone can make a website, for $12 million, just there, was a big kind of shift in my perspective. And so, I thought, “Well, not cut out to be a physicist, maybe this startup thing could be interesting.” And I really liked programming. And so, my brother and I, we decided to start a company. It did OK. It was acquired for a reasonably small amount. Went back to school. And something we learned over the course of that first company was… Cloud computing was just starting back then. I think EC2 launched in ’07, maybe ’08. And it was now getting incredibly easy to just launch a website, and you could do that in an afternoon. You didn’t have to call someone up and rent a server. But whenever you wanted to move money, it was kind of jumping back in time a century. You’d go to a bank and fill out paperwork, and the forms were in Latin—or so it felt. And so, it was just this kind of weird dichotomy, where certain aspects of setting up a business were becoming kind of so streamlined. And then other parts were, yeah, were so… There was such a high activation energy barrier and kind of so much inhibition, and we thought something like Stripe must exist, and we were Googling for it. Like, there must be, like… How could there not be an easy way to move money online? Like, it’s not an obscure need. But we eventually concluded there wasn’t. And so, we’re walking back from dinner, not that far from here, a restaurant in San Francisco, and John turned to me… We’d come out here, actually we come out here for a startup event at Berkeley. You guys hosted YC startup school in 2009. And we were feeling very inspired. And so, after dinner, John, we’re walking back, John said to me, “Well, we should just build a prototype of this Stripe thing. You know, it can’t be that hard.” Here we are.
– Wow, that’s great.
– Great, so we wanted to ask you also about the early days of Stripe, ’cause every startup faces challenges and hurdles. So, could you describe for us, like, one big challenge or hurdle that you encountered in the early days of Stripe, and how you overcame it?
– Well, I had breakfast this morning with another startup founder, and we were kind of reminiscing about or discussing the fact that people will sometimes ask us now, I mean, she’s been working on our company for, I don’t know, five or six years now. And people will sometimes ask, like, “Is it still fun?” And we were kind of making fun of this question because the early stages of the company are never fun. And so, is it still fun? Like, which period was fun here? So, look, the early days of Stripe, like, in hindsight, it’s now, like many maybe intense experiences in life, you look back on them with some kind of fondness, but when you’re living them, I mean, you don’t know the outcome. I mean, they’re just stressful. I remember very vividly just how intent, intently, and intensively we worked. And I don’t know if we kind of had to, but we felt like we had to, and we did. Like, we started out down in Palo Alto, and we hired this guy who lived in San Francisco, and he ended up deciding to move from San Francisco to Palo Alto because the last Caltrain left Palo Alto to go north at 11 p.m., and he felt guilty leaving the office early every evening. And that’s, like, again, maybe that was… We were kind of misallocating our time or something, but just as a descriptive matter, that’s how it played out. And then, of course, there are all sorts of undulations and tribulations. Like, I remember our first serious outage very vividly where people are using Stripe to move money, and if Stripe is down, their business is down. So we felt this very intense responsibility, and a rack blew up at a data center, and Stripe therefore was unavailable. And we had a kind of… We didn’t have proper redundancy at the time. And I was paged at 2 a.m., and very unusually I’d left… I mean, Stripe was five people at the time, and very unusually, I’d left my laptop at the office, so I had to get on my bike, pedal to the office, start trying to fix things. It took six hours. Eventually, 8 a.m., 10 a.m., it’s back online. I remember feeling so horrifically dejected because I realized, as Stripe recovered, that we hadn’t received a single complaint. Nobody had even noticed. Stripe was that inconsequential in the world. And then, of course, subsequent to that, we had production issues that were… In fact, there was no shortage of complaints. And you’re taught to be kind of careful what you wish for. But I would say overall that… And by the way, I think this is not… I don’t think Stripe was in any way an unusual experience here. Like, I don’t think it’s in any way unusual where the experience of the first five years were… It was fulfilling, but, yeah, it was just like difficult AF. And I remember Jensen recently saying publicly, Jensen Huang from NVIDIA, that, if he’d known what starting NVIDIA would be like, even knowing the outcome today, the hottest company in the world, whatever, that if he’d known in the beginning what it would actually take, that he wouldn’t have started it. And I think, even when you condition on success, you get sort of a surprising number of responses like that.
– So we’ve been—
– I’m sorry, I’m really encouraging you guys to start a company. It’s great. Such a motivational fireside chat here. But look, actually, OK, if I knew what it would take, I would still start Stripe. There.
– You’ve heard it here first.
– Maybe NVIDIA was just harder or something, and I’m just a wimp, but…
– Yeah. Maybe he was having a bad day. So we’ve been really intentional here at the Haas School at creating a deliberate culture, which our students know well. We have four defining leader principles. One of my favorites is confidence without attitude, which is why I think our students are so open to learning and not overconfident. But I’m wondering about your intentionality in designing the culture at Stripe, and what you did initially, how you’ve scaled the company. I think now you have something like 7,000 employees. What looks different now? How is the culture doing? How are you thinking about it? We would be super interested to know.
– Well, just, you mentioned kind of the culture, the values here. And actually, we did notice pretty early on, and I won’t mention kind of any other schools, but just, we noticed that Berkeley students were more humble. And that was something that… But gee, you can’t let that go to your heads. It’d be self-defeating. So you can’t be proud of that. But yeah, I don’t know if it’s the Irish thing or something, but somehow… I mean, Ireland has kind of a hypertrophic pernicious version, I think, of humility, where we start to get resentful of anything that’s successful. Like U2 or any other tremendous Irish export, we take a very dim view of those things. But anyway, so we, early on… I don’t know. Well, partly because we knew that Stripe, even if it worked, would take a long time, just ’cause it’s an infrastructure business. With Snapchat or TikTok or something, everyone can just kind of decide overnight, we’re going to adopt it, it’s going to become super popular, within two years it’s an overnight success, whatever. If you look at the internet, or if you look at… Take a company like Amazon, kind of companies that are operating more at the sort of infrastructure level. Like Amazon, within its first couple of years, was not growing 100% year over year or something like that. Amazon, within four years, was growing at about 30 points a year. And what’s remarkable about Amazon is just the durability with which they’ve sustained that growth, but the kind of the rate itself is not particularly noteworthy. And similarly, if you take the internet, the internet has grown at a compounded rate of around 30% to 40% year over year. But in no year after the first two, I think, did the internet double year over year. Again, it was just this remarkably sustained growth over the course of, now, more than four decades. And so, anyway, we kind of knew that Stripe would… Again, even if it was going to work, would probably have roughly that kind of character. And so, yeah, we wanted to figure out a sort of a cultural orientation that we thought would kind of befit that, and that that requires recognizing that we’re not here to build cars, we’re here to build roads, and it’s the kind of personality not of someone who wants to build some hyper-successful app, but the kind of person who would like to build a TCP stack. And that’s not everyone, and that’s fine. We need cars for the roads. And so, there’s kind of a diversity of different skills required. And again, it is very hard to separate that which is adapted for Stripe and that which is just kind of personal preference or something. But because Stripe’s domain is really complicated, and where the details really matter. Like, if we make a mistake, just one mistake, there’s a very good chance that that’s, like somebody’s paycheck is wrong or something. It’s even in a single instance… There’s this, I think, a culture at Stripe of just really prizing the small details. And we talk a lot about craftsmanship and rigor and abstractions that can endure over decades and sort of really getting those right. So I would say precision is pretty deeply embedded.
– So could you talk a little bit about your own leadership journey during that process? It’s one thing to be the leader of a small startup, another to be running an organization of the size and scale that Stripe is now. And thinking, in particular, about this idea of rigor or precision, how do you really try to embody that and reinforce it through your own leadership?
– Well, the good news about startups is, as far as I can tell, if the initial idea is good, and if kind of the market’s good, like, kind of those core characteristics, and if you’re willing to recognize your mistakes, the startups… So startups I think are actually quite resilient, and they can endure a lot of managerial malfeasance as you learn along the way. And I definitely didn’t come to Stripe with any kind of enlightened leadership expertise or sort of genetic muscle memory or something. Like, Stripe didn’t have any managers until we were 70 or 80 people. And that’s not a best practice. If we were doing it all again, I would definitely kind of invert colors on that one. And also, there’s lots of other things that I think in hindsight were ill-advised and mildly unhelpful but empirically survivable. And so, I think at a meta level, the question is, yeah, much more sort of the rate of adaptation, the rate of learning. And John and I, we’re… Stripe conducted layoffs in 2022, where… During the pandemic. I just said a moment ago that the internet, Amazon, whatever, didn’t grow that quickly. During the pandemic, it was such a crazy time that even though, again, Stripe broadly has that character, we roughly doubled in 2020 and 2021. And so, kind of our forecast, how big we would be and what we would need to sustain the service and everything got pretty out of whack. And so, in 2022, we were trying to kind of rectify this, obviously acknowledging a misprediction on our part and significant mistake. And John and I had this conversation, where we decided, imagine that we’re sort of marauding private equity raiders who’ve just purchased Stripe, and we’re horrified at the decisions the prior management has made. And the want and errors and mistakes and sort of grievous instances of mismanagement that have been committed. And somehow, I think it’s much better to be right than to be consistent. And I think somehow getting yourself into the psychological frame where that’s OK, and you can swiftly recognize, well, I tried this, it’s really not working, let’s do something else. I think, as long as you get into that mindset, I think a lot of the actual errors themselves are recoverable.
– Yeah. Better to be right. We agree with that. So…
– But it sounds like a… It’s easy to say. I think a lot of people, and to be clear, even myself, like, I think we all feel surprisingly strong kind of psychological pull to kind of intertemporal internal consistency.
– Well, there are huge biases. Actually, one of the deep experts in what’s called the escalation of commitment cycle, Barry Staw, was a professor here for many years, and it’s a very, very treacherous bias. That consistency, people really worry about having an external image of consistency, and even maintaining consistency internally, because in some ways consistency describes the essence of who you are.
– Yeah, and look, I mean, presumably there is some set of things about which one ought to be consistent, right? And so, it’s maybe wrong to toss consistency overboard wholesale.
– Right.
– But tying your self-conception and identity to specific management practices seems like maybe over-constraining the action space.
– Yeah. Well, I mean, the countervailing force is an experimental mindset, right? Where you’re actually discarding ideas that aren’t working because they’re not working.
– Yeah. And again, maybe the fact that we kind of came from the middle of nowhere in Ireland, like, it was… One thing that is, I think, helpful, a lot of different cultures around the world have kind of some self-conception of grandeur, right? The French people, English people, say Americans, one could occasionally accuse them of that.
– What?
– One thing that’s great about Ireland is, we never have delusions of grandeur. And Ireland never thought of itself as, like, the best country in the world by objective criteria. And very fond of it, to be clear. And so, anyway, it wasn’t that hard for us to think, “Well, we know nothing about this domain, and we should assume that 70% of the things we try will turn out, ex post factor, would be mistakes.”
– Yeah. Yeah, yeah. Well, an interesting issue. So switching topics a little bit, there’s a lot of discussion about… Well, we used to think about hiring people for culture fit. In fact, my dissertation was on culture fit and understanding all about culture fit. And in addition to finding people who can do the job, you want to find people who resonate with the culture that you’ve created. But people are now talking about culture add, and ensuring that you’re not just fueling a kind of homogeneous mindset in an organization. And so, I’m just interested to find out where Stripe is on this balance.
– Well, I think this question’s really interesting at the meta level, where, “How do we have more variegated and kind of heterogeneous cultures across an industry and across a society around the world so that people can find the place where, yeah, that that place is me, right?” And actually, one of the things that I really like about the internet… Like, Stripe’s mission is to increase the GDP of the internet. There’s sort of a fundamental question of like, “Why should you be excited about that?” Like, kind of, who cares? And I think part of it is, that enables increased access to goods generally, where, for most people in most parts of the world, it’s pretty difficult for them to benefit from the set of goods and services that, say, we here in the Bay Area might be able to purchase. And again, growing up in Ireland, there were so many magazines or newspapers or whatever we would get, and there’d be little fine print, “Offer not applicable in the Republic of Ireland,” where it was printed for the U.K., but they hadn’t figured out the shipping mechanics for Ireland or something. So something I think about the, just about this, kind of this global extension, global access. But maybe a second order aspect of the internet that personally appeals a great deal to me is, I think we want a… I think a richer society, not in the kind of pecuniary financial sense, but in kind of a… In the second sense of the word, is one where there’s a greater array and more complexity of goods and services, and that variety is in more abundance. And if you can aggregate more demand via the internet, well, then it can make sense to serve some very narrow niches, right? And part of what I love about Stripe is, I’ll stumble upon so many businesses where I think, wow, like, I would never have thought that that could even be a business, right? You know, people, a marketplace for sort of user designed and user-created action figures, right? Or CRM for Boy Scout troops. And these things would never make sense if you just had a village or you just had a town. It only makes sense where you can aggregate sort of the entirety of the internet’s demand. So anyway, I think that this also, with respect to cultures and if one can only work in one of 10 businesses, well, they’ll probably all… Like, by necessity, they’ll probably all have something approximating the same kind of culture just by virtue… Like, there can’t be that much self-selection going on if there’s only 10 places. And I love finding organizations with just very unusual and places where the central limit theorem doesn’t apply. And I visited the folks at Jane Street not that long ago, and I really think Jane Street would not be for most people. I mean, it probably wouldn’t be for me, but like, for the people at Jane Street, it seems awesome, and they stay there so long. Right? And my wife’s a scientist, and as you get a sense for the cultures of different departments at a university, or even across universities, you realize, well, some of these have very unusual cultures. And again, probably not even for most scientists in that area, but right for a little group of people who can affiliate. So anyway, just to your kind of point about culture add, I think figuring out mechanisms by which we can enable more of that structural diversity kind of across the board. I think that’s really helpful.
– So turning to unique cultures, I want to come back to Stripe. And one of the cultural tenets that you have talked about as being, “Continually paranoid at the prospect that we might be forgetting something important.” So I’m wondering if you could give us an example of that tenant in practice, but also, how do you keep it from going too far, from being too paranoid?
– Can you be too paranoid? What do you think? Well, if one can be too paranoid, then we might have to reconsider some things. So… The thing that makes me so paranoid, and maybe this is kind of more idiosyncratic to Stripe, is again, just the… Like, we handle around $1 trillion a year, that works out to around 1% of global GDP. And again, there’s just kind of basic point that, if Stripe is unavailable, that’s a lot of people and a lot of businesses and a lot of activity that isn’t happening. And during the pandemic, DoorDash, Instacart, Zoom, and Amazon, and so many of what kind of felt like load-bearing pillars of society, they were handling their transactions with Stripe. And so, we feel this extremely… I don’t know, solemn kind of custodial responsibility to be able to provide uninterrupted service to them, the way that they ought to expect. And so, it’s less a competitive paranoia, and more paranoia that we’ll screw something up with respect to our obligations to them. And again, I don’t know if you could take that too seriously, but for us, it’s very weighty.
– So I have a question. I used to teach a case about John Reed, who decades ago was the CEO of Citigroup, and he was one of the first to use a kind of metaphor of the back office being like a factory. You use a metaphor of front office, back office, in finance being more like a bicycle, which I find appealing. I’m a cyclist. So what does that actually mean, and how does that influence how you make decisions and how roles are thought about within Stripe?
– Well, I’ve forgotten I said that, so thank you for… It’s a good metaphor. But… The thing we spend a lot of time thinking about is necessarily in organizations you… Or maybe necessarily is a strong word, but the bureaucracies haven’t worked that well. So it appears necessary with current organizational practices to have some kind of a hierarchy. And I don’t have a better idea. So let’s just kind of take that as a premise. That necessarily involves some logarithmic game of Chinese whispers and information loss sort of through transmission. And so, the picture that the people making the most important decisions may have, might be meaningfully divergent from that which is actually true. And so, the thing I’m always wondering about is, just, “How do I know what’s actually true?” And not just how do I know, but how do leaders generally across Stripe know what’s actually true? And so, we built our own internal project management software. And it’s less because we want to be able to kind of customize the animations. It’s more that we want, like… Because it is so important that we know what reality is saying that, we want to be able to kind of figure out the optimal way of surfacing that context. And I mean, it’s true to some extent in every organization, but I think it’s especially true in knowledge work and in the creation of… Like, software is an interesting thing, where there’s mining, where maybe there’s just some kind of linear elasticity between the number of people extracting the rock from the ground and how much rock is produced in the economic value, whatever. It’s very kind of tailor-en. And then, you have movie making or novel writing, where, if you’re a publishing house, you can’t just kind of measure your likely prospective success on the basis of just number of writers, right? Like, it’s so sensitive to the specific efficacy of every individual person. I think creating software is sort of, in an interesting way, halfway between mining and novel writing, where look, there is just… There is some scaling in the amount of work and one person could not build all of Facebook, or choose your service, but it’s definitely not linear. And for us at Stripe, even though decisions have to get made, the work is not being done by the decision-makers. The software engineers and the designers and everybody involved in… The partnerships people, whatever. In the creation of the product, they’re our authors. They’re the people actually creating Stripe, and everyone else is in some sense playing a supporting role. And so, I think it’s important to have that kind of inverted mental model in mind. And it’s not sort of a feel good thing, it’s just, it’s a deep truth. Like, I don’t know how many support staff work behind J.K. Rowling to enable the books to get published. But J.K. Rowling is the one doing the writing.
– Yeah, right.
– So I’m going to turn to another facet of the culture at Stripe, which has to do—
– Can we talk about J.K. Rowling, or is she canceled? Alright, OK.
– Not yet.
– We’ll let it slide this time. So—
– We’re at a business school.
– Yes, exactly. So the other facet of the Stripe culture I want to talk about is the writing dimension. And, of course, it’s becoming more pervasive now, but my sense is you were one of the early adopters of a writing culture. So tell us a little bit about where that comes from, and then, how you reinforce it.
– It’s kind of funny. I know what you mean. It’s kind of funny to consider oneself an early adopter of writing culture for true to form tablets. What do I say? Well, partly I think it comes from… We were just kind of misanthropic introverts in the beginning and we, even when there’s only four or five of us working on Stripe, we would just communicate a lot in written form because it’s kind of less oppressive than having to talk to each other. So part of it was just that kind of predisposition of early people. And to be fair, if I must include myself. Like, John, my co-founder is much more extroverted and charming, and usually, he’s kind of… It’s rare that I’m wheeled out for public engagements.
– He’s the front man.
– You can see why. So if John didn’t exist, it’s unclear whether Stripe would have any customers. But I do like writing, and I like reading. I think it’s like a… I found very… Bruno Latour has this piece about… And he overcomplicates it because you have to for that kind of work. But he has this piece about kind of immutable, I guess, mobiles since he’s French. And he kind of makes this point that the printing press is maybe correctly associated with the advent of the scientific revolution. But maybe the sort of simplistic sense, in which we might perceive the causality there as, “OK, we have the printing press, we can distribute more stuff, and now just people have more information, whatever.” They come to more insight. He makes the point that before the printing press, manuscripts were necessarily copied by hand. And in the act of copying by hand, obviously there’s the prospect of the introduction of error. And so that means, when you kind of encounter or confront some observation where there’s a disparity between that claimed by the work or by the theory or whatever, and what is it you see, you can’t really tell. Is it because the theory is wrong, or is it because there was some boring mistake made along the way, right? And it’s kind of when knowledge became more rigid that it became easier to break in a way that is conducive to the rejection of false theories and inadequate explanations. I found that very thought-provoking. And I think there is something to that. And obviously, this also gets the difference between oral cultures and literary cultures, where, I guess back to this idea of assuming that 70% of what we believe is wrong, if we don’t write it down, it’s going to be harder to remember what specifically we thought and what specifically we believed because our minds will play subtle tricks on us. And so, I think part of the value in writing things down is, our past selves look stupider. And that’s actually very adaptive because we’re like, “Oh wait, we had these beliefs, and just, clearly these two are not true.” And I think, yeah, that robustness through time makes it easier to find our flaws.
– I was going to say, as a sociologist at a business school, I’ve talked to lots of CEOs. This is the first time someone has brought up Bruno Latour in a response to a question. So very impressive.
– Yeah. I’m going to skip around here ’cause we want to get to the audience questions. So actually, Sameer and I have a podcast because, like, who doesn’t? And our second episode just dropped today, and it’s about… CEOs have been, lots of managers have been asking us about work schedules and remote work and hybrid work and what’s working. So since we don’t know the answer, we’re going to ask you. So I understand that Stripe’s remote workforce increased from about 20% pre-Covid to almost 40% now. Is that about right? Yeah. So I mean, how has this influenced the company? What differences do you notice? Is it what you’re expecting for the long term?
– Well, I think the right answer is probably quite scale dependent and like, with Stripe now being 7,000 or 8,000 people, we can’t all fit in the same room regardless, no matter how draconian our in-office policy, right? And furthermore, we serve businesses and employ people all around the world. And so, we’re necessarily somewhat remote in that sense. Like, there is no universe in which we’re not having lots of Zoom meetings, just because of the globally distributed nature. And then, so the question is more something around the exact nature of the interactions we want people to have in the course of their day, but not, again, whether they’re conducting a lot of work that is distant from their colleagues. And so, I think for a large organization for whom that’s the case, there are considerable benefits to in-person work, just for boring reasons… I mean, everyone, this becomes such a religious debate, but I think there’s… A bunch of the kind of precepts are, I think, fairly uncontroversial. Like, there are some people who are really effective and really enjoy working in their cave. I think I would probably be one of those people, and sadly, I’m not in a role where that makes sense. But some people are, and that’s great. Other people, they get really bored, and they go, they get cabin fever, and they really want to be around other people. And just like, those people exist as well. And society probably has some mechanism for the provision of employment for both categories. It’s probably a significant efficiency gain to have more options for the kind of, the cave dwellers, the people who just want to sort of sit in the room and do the work by themselves. Again, like me. And then I think there… You have to think about it kind of longitudinally, where there’s a question of skill and culture and knowledge transmission. And so, I think some of the analyses that look at kind of short-term efficacy, they… I mean, that’s interesting, but I think you have to take… Ideally, you have to take a kind of a full life cycle view. And I think that sort of the cohorted change over a workforce. And Berkeley has a pretty strong culture, as I understand it. And they’re like, if Berkeley went remote, maybe Berkeley would be fine like next week or next month. But the idea of like four, 20 generations of Berkeley students being remote, I have to think that culture would at least be different, right? So yeah, I think all those considerations apply, but maybe relevantly for this room, I don’t know, I do notice that the 10-person, the 20-person, the 50-person companies, for whom being in room together is an option, the ones that exercise that option really seem to do better. And I think we all even intuitively kind of know it, where… I often ask parents, like, if your kid was considering two different jobs, one is kind of fully in person, resolutely five days a week, and one is kind of loosey goosey or fully remote or something, which, with your kid’s best interest in mind, which would you advise them to take? And no parent that I’ve asked has ever hesitated in answering that question.
– That’s right. OK, so, yeah.
– So I have the honor of doing the audience questions as well as my own. So—
– Austin, you want to introduce yourself?
– Sure, yeah. So I’m Austin Schoff, I’m a second-year MBA, I’m also a member of the Dean’s Speaker Series board. I think the first question we want to start with is the future of money movement. What do you think will happen? Will it be more real-time payments? Will it be more crypto type transactions? Will it still be T+2, and we’ll be stuck using Swift for eternity? What do you think is the future of money transactions and money movement?
– I think the short answer to your question, like, to each specific sub-question you asked, I think the answer to each one is yes, except maybe the T+2. But so, first off, the U.S. is actually one of the places in which payments are changing the slowest, where across most other… Not most. Many other major markets around the world, there’s been enormous changes over the last 10 years or so. So obviously, UPI in India. Everyone knows WePay and Alipay in China. But those are kind of famous ones. But like, Pix was launched in Brazil in 2020, and within two years, the majority of the Brazilian adult population was a weekly active user on Pix. It’s kind of UPI for Brazil. There’s Swish and TWINT in Sweden and Switzerland, respectively. And a whole host of schemes like this across in… Malaysia has its own. Japan is now finally starting to change pretty rapidly. So it’s happening, is the short answer. The Ron Paul GIF. And then, crypto, I think that… I mean, there’s kind of a fundamental question for crypto, like, these different lenses of analysis of store of value or some kind of risk hedge or kind of are talking like a means of exchange. But as a means of exchange, I think you should probably further disaggregate between stable coins and like, crypto crypto. I think the stable coin thing is happening, and the absolute numbers are sold reasonably small, but not that small. I think the outstanding tether volume is now, like, in the order of a $100 billion or so, like, it’s not nothing. And then, given the existence of stable coins, it’s not totally clear to me why you would ever use Bitcoin or some other kind of less convenient currency for a medium exchange. And Swift will still be a useful technology for interbank settlement, as it was initially designed. Or forfeiture was initially designed. And T+2, it’ll probably shrink, but you do have to… I mean it’s interesting to look at with FedNow in the U.S., this kind of new real-time payment scheme in the U.S. that has only kind of partially rolled out support, but we’re already starting to see that having instantaneous transactions… What about scams? What about fraud? What kind of oversight is possible? So it’s not a totally free lunch. And Bitcoin, of course, acolytes not unjustly, because for certain use cases this is important, but they often celebrate the fact that transactions are instant and irreversible. We also see some of the downsides of that with respect to other activities. But it’s all happening.
– Next question I want to hit on, how do you deliver constructive feedback while maintaining trust, especially given that a mistake in your organization could be the loss of a couple billion dollars depending on scale?
– I think people really want to… I think they really want to know what is… What is their interlocutor’s authentic view of them? And I think loss of trust doesn’t come from constructive feedback but comes from a divergence between what’s true and what you’re saying is true. And so, I would almost kind of invert it where it’s like, well, if you’ve constructive feedback and you’re not telling them, I think you’re really undermining trust. And, look, this is… I don’t know if it’s possible to teach giving good feedback. I haven’t heard of it being taught well anywhere. Maybe it is, I’m not saying it’s not. But I think it would be a very valuable skill to teach if, in fact, it can be taught well, and some of the people at Stripe… There’s, in fact, I think, even a causal relationship, a causal positive relationship, where the people at Stripe with whom I feel the greatest trust are, to a significant extent, those who give me the most critical feedback. And maybe just even kind of knowing… Like, maybe that fact should just be like on a billboard. Not about me, but just that these things, in fact, can go together.
– Could you talk a little bit about Stripe Climate and how we should think about your commitment to carbon removal as part of Stripe’s long-term goals?
– Yeah, so… If you look at the IPCC forecasts of what’s going to happen to the Earth’s climate over the course of the 21st century, basically, we’re going to significantly overshoot any reasonable target. Two, two and a half degrees, whatever. Unless we remove a significant amount of CO2 from the atmosphere, because the problem where, even if we fully decarbonize the economy, all the CO2 molecules, they don’t know about that. Like, they’re still there. And so, there’s this kind of overshoot and overhang, and we were looking at these IPCC reports a few years ago, and all companies are sort of announcing their sort of fancy climate whatever. And many of those are good and come from a good place, but not all of them are, and a lot of companies are pursuing kind of carbon offsets that we think are… Like, are just fundamentally kind of, almost necessarily fraudulent. Like, I think they’re fraudulent, they almost couldn’t not be, even if you were trying, where you’re kind of paying for counterfactuals and you’re like, “OK, well, I will not cut down this forest, please pay me.” But it’s like, were, are you going to cut down that forest? Like, how could anyone even know that, right? So anyway, we’re reflecting on those IPCC reports, and then kind of thinking about corporate programs in the context of climate generally. And I mean, because of this kind of precision rigor thing, we hate doing something that’s… Inauthentic is not quite the right word, but, if we say something, we want it to be literally true. And doing something kind of just for show really kind of rubs us the wrong way. So anyway, we noticed that no company in the world as of 2018 had ever purchased commercial carbon removal. So removal is totally different to offsets. Removal is like, I will give someone some money, and they’re going to have to bring back some actual CO2 molecules. But there’s this kind of, this proof of work involved, and more importantly, the molecules are no longer in the atmosphere. So no company had bought from any of these companies today. There were, or back then, there were, I think, only two companies even in operation. We thought, “Well, this will have to become a sector.” Again, the IPCC reports show you that. Like, they say we’re only not screwed if we remove a lot of CO2. And so, we thought maybe it’ll be helpful if we kind of… I mean it’s not like Stripe is Microsoft or IBM or something. Like, we’re not the most blue chip, legitimate, or legitimizing buyer you could have, but we’re something at least, and so, maybe it’d be helpful to these companies if we started to purchase from them. So we started to buy from them in 2018. Our first transaction was for $1 million of CO2. And kind of based on that, and learning more about the sector, we were very fortunate to have some really terrific people join. And they cooked up this really incredible AMC, the second-ever large-scale AMC. So AMC stands for advanced [market] commitment. And the idea is, you pre-commit to purchase something, something that doesn’t currently exist, as this was first pioneered for vaccines where there are all sorts of conditions we want to vaccinate people for, but like, the vaccines literally don’t exist. Of course, the vaccine manufacturer, the scientists or whatever, there’s market risk for them where it’s like, well, if I invent it, will anyone buy it? And so, this is a way to try to kind of bridge that gap. Worked quite successfully there. Gavi was the program. And so, we decided, “Hey, let’s do the same thing for carbon removal, where we want more of these companies to exist. We’ll pre-commit upfront to purchase $1 billion of carbon removal, at any price, from whoever will come along and sell it to us.” And so, we assembled a coalition that includes Meta, Alphabet, McKinsey, JPMorgan, a host of other companies, but Stripe committed the largest amount to it. And gosh, we’ve now, I think, purchased from on the order of 40 companies or so. And the stat that I’m proudest of is that, in most cases, we are the company’s first-ever customer. And so, we’re not like coming along to somebody who, we’re their thousandth customer, it’s already validated. Kind of, we’re through frontier, this organization kind of sticking our reputation on, “Hey, we think this technology is legitimate, we think it has promise, we’re going to contract with you today, and hopefully you can now use that contract to sell to others, and so forth.” So anyway, it’s early days. Most of the CO2 is still in the atmosphere, so it would be premature to declare any kind of victory. But some of the companies are now actually removing it, and yeah, we’ll see how it goes.
– How do you think technologies like Stripe can help businesses in developing markets, maybe even in places where internet connectivity is not as strong as it could be? What do you think the future is for the developing world?
– I think the… Well, I think the internet access problem is pretty quickly being solved, in that… Like, over Thanksgiving, I went to Brazil, and I was on a little boat on the Amazon, and I had perfect 3G reception. Across most… Like, India thanks to Reliance Jio, data is becoming super cheap in the most populous country in the world. And there’s kind of versions of this story playing out in so many places. So I don’t worry too much about the provision of internet problem. I think the… Look, I think the internet is… Is one of the most important technologies ever created with respect to the enablement of global development. And so, I was born in Ireland in 1988, and when my parents who were born in Ireland, Ireland was a kind of, was a deeply impoverished theocracy, and the Catholic Church kind of ruled with an iron fist, and we were a very kind of mercantilist socialist, closed little enclave. And there was a great deal of interest in sort of, how could Ireland be the… How could things in Ireland be working so badly? We’re right there next to England, why is Ireland so bad? And it turns out that it was bad policy. And some enlightened people like T.K. Whitaker and others, they proposed that, “hey, if we reformed and reconfigured ourselves in a kind of more free market direction, that good things could follow. And there was some propitious timing, where the EU came along, and we joined it, and American multinationals set up operations in Ireland. We benefited from that and so forth. But hey, point is, between when I was born in ’88 and when I left for college in 2006, Ireland had, I think, the fastest economic growth of any country in the world. And you can’t grow up around that and kind of fail to internalize the, like, the moral importance of economic growth. And it’s kind of avant garde, here in the U.S., the Bay Area, rich places, just sort of degrowth and the ills of capitalism, whatever. And for places that are already extraordinarily prosperous, I understand how it might be difficult to perceive the underlying imperative there. But having kind of seen some of the kind of… The longitudinal difference, it’s… And I’m sure many of you have kind of versions of this from your lives. The criticality is very apparent. And so, anyway, Stripe’s mission is to grow the GDP of the internet. And we think about this a lot with respect to those emerging markets where… There’s only one technology that we’ve ever found to enable impoverished places to become, or at least to get on the trajectory to kind of full developed world status. And that is a free market economy connected and integrated with the rest of the world. And Stripe’s obviously not going to solve that, but if we can play a very small role in enabling those transitions, we’ll feel good about that.
– Amazing.
– Fantastic. Are you done?
– We have one final question.
– OK, one final.
– Are there any founders that you strive to emulate? And then, conversely, are there any founders who give you the ick?
– We won’t tell.
– It’s not recorded or anything.
– Exactly, not being livestreamed. So… Look, it’s hard to single out individual founders. I think that, I mean, which, look, it’s kind of a bland answer, but also true. I mean, I’m kind of partial to… I mean, we culturally know a lot about the present day founders, right? I think it’s kind of interesting that the founders of kind of generations past, and like, not that many generations past, are so much less culturally conspicuous. And so, sorry, this is now going to become an infomercial, but there’s a book Stripe Press published called “The Big Score” about the semiconductor industry, primarily in Silicon Valley, in the ’70s and the ’80s. And those characters and firms like National Semiconductor, and so on, they’re kind of forgotten today. But they were incredibly impressive. Or Cypress, these sorts of businesses. And so, I found it fun to learn from them and to see kind of what’s similar, what’s different. And then, I don’t know, I really enjoy kind of founder-like personalities from non-startup domains, and just kind of, again, both the sort of the compare and the contrast. And so, a guy who really inspired me growing up was Ed Walsh, who started the University of Limerick when he was 31. And nobody… Limerick is a very small city in Ireland, and nobody at the time really thought that Limerick even deserved a university. And he couldn’t persuade anyone to kind of let Limerick have a university from Day One. And so, it was a national institute of higher education. My dad was actually in the first year of students there. But then, eventually acquired kind of full university status. But like, I don’t know, he was a 31-year-old, who before that was living in the U.S. and decided like, screw it, I think Limerick should have a university. And over the course of several decades, really realized that vision in an extraordinarily impressive way. So, I don’t, I like finding those personalities from the nontechnology domain.
– Yeah. Well, thank you Austin. Those were terrific questions. Patrick, a whirlwind of insight and things… Yeah. Citations—
– That was so disorganized that I’m having trouble even summarizing it.
– No, no, no. Well, you began with Ireland, and you ended with Ireland. I think it’s completely appropriate. We are so grateful for coming to share your thinking. We hope you will come again sometime soon. But let’s offer Patrick our deepest thanks for coming out.
– Thank you very much.
– Fantastic. And thank you all for coming. We really appreciate you attending. We hope you got a good lunch and lots of mushrooms, I think. We’ll see you again soon. Thanks everybody.
– Thank you. Of course, that was fun.
– Yeah, that was amazing. That was so fun.