Part IThe Story
The pager went off at four in the morning. Stripe was about six months old, a handful of people writing code for other people who wrote code, and something had broken. An upstream switch at the data center had failed. The API was down, so no customer could take a payment. Patrick Collison went straight to the office. The data center needed eight hours to fix the switch, and for all eight he did what young founders do in that situation: he braced. He expected angry emails, a stream of complaints, the slow loss of trust that could end a company built on the promise that it would be there.
"This is not like a social media app or something," he said years later, telling the story to an audience of entrepreneurs in April 2017. "This is the conduit for our customer's revenue. So it was a big deal."
Then the hour came around.
— Patrick Collison, This Week in Startups, April 2017The darkest moment of course was because I was bracing myself for the deluge in complaints and dissatisfied customers and anger and vitriol and everything else. And 1pm rolled around and as far as I could tell, nobody had cared or notice[d]. On some level, that was good news. But as I reflected on it, I realized that it was actually kind of existentially bad news.
The phrase is existentially bad news. The young company's real problem was that it did not yet matter. A payments processor goes dark for a morning, and nothing in the world registers it, because nothing in the world depends on it yet. The goal implied by that silence is odd, and it has organized the rest of Patrick Collison's working life. The aim is to become something whose absence would be felt everywhere. The cost of getting there is to become something whose presence almost nobody notices. Infrastructure succeeds by disappearing. The water main, the switchboard, and the clearing system are all noticed only when they fail, and Collison set out to build something that would fail rarely and be missed badly when it did.
He had some practice at being unnoticed. He grew up at the edge of a network that did not know he was there, in a house where the internet came late and arrived by satellite. In the fine print of the websites he and his brothers eventually reached, they kept finding the same sentence telling them, politely, that they were standing on the wrong side of the glass.
By the Numbers
The Stripe Ledger
$159BStripe valuation in its February 2026 tender offer
$1.9TTotal payment volume processed in 2025, up 34% year over year
5MBusinesses whose transactions Stripe processed in 2025
~25%Share of new Delaware corporations formed via Stripe Atlas, mid-2026
19Patrick's age when he and John sold Auctomatic, on Good Friday 2008
14%Share of the workforce cut in November 2022, the 'hardest change' in company history
$17.5BPatrick Collison's estimated net worth, August 2026
Offer Not Available in the Republic of Ireland
Dromineer is a small village on the shore of Lough Derg in County Tipperary, about thirty miles from Limerick. Patrick Collison was born in Limerick on September 9, 1988, the eldest of three boys, and was raised in Dromineer. His mother, Lily, was a microbiologist. His father, Denis, was an electronic engineer. By their sons' account, both parents ran businesses they had started themselves, and both were busy enough that the boys had a great deal of freedom. "It would be fair to say," John Collison told Bloomberg in 2018, "that we were free range kids."
The life was rural and the distances were long. The drive to school took forty minutes. None of their friends lived nearby, so after school they could not run around with classmates and played with each other instead. Every day followed the same routine. "We would go to the library after school," Patrick said, describing it in a tone close to an engineer's: get some new books, read the books, "rinse, repeat for the next day." At some point one of those books was about programming. He read it, and it seemed, in his word, awesome.
— Patrick Collison, Bloomberg Studio 1.0, June 2018I read about the internet before we had the internet.
Read that sentence again. It describes a boy holding a theory of something before he had any experience of it, and that habit stayed with him. When the internet did reach Dromineer it was nothing like the frictionless highway in American advertising. The village was too remote for an ordinary connection, and the first usable broadband, according to the columnist John Naughton, came over an expensive satellite link. When the boys browsed, they learned to read the small print at the bottom of the page, and over and over the small print said the same thing: offer not available in the Republic of Ireland.
"So the sense was staring through the glass at this amazing world and internet out there," one of the brothers recalled on the Bloomberg program, "and not all of those opportunities being available or equally available." The same brother added a note of caution that the rest of this story has to keep in mind: the link between that experience and Stripe's later concern with global access "really was not conscious in any way." It became visible only looking back over the next twenty years.
The household set the tone. Patrick took his first computer course at eight, at the University of Limerick, and began programming at ten. He went to Gaelscoil Aonach Urmhumhan in Nenagh, an Irish-language primary school, and later to Castletroy College outside Limerick. John was two years younger and followed him into code as a teenager. Asked whether the brothers had hacked each other's websites, Patrick said, "I think that did happen." The third brother, Tommy, would enter the national Young Scientist exhibition in 2010 with a project on blogging. In 2011 John told an Irish tech reporter that the whole family blogged. One day their mother asked to be set up with a blog of her own. John installed WordPress on her machine, and she joined Twitter the same day.
Around them the country was changing. "Back in the 50s and 60s, Ireland was the poor man of Europe," Patrick told reporters in Sydney decades later. "Through the 80s, 90s, 2000s, we saw the Celtic Tiger and a complete transformation in Irish livelihoods." Then he made the point that matters most here: "You can't help but internalise the importance of that when you grew up in the midst of it." In 2014 John offered a drier reading of the same childhood. He told the Financial Times that he often wondered "if it's desirable to grow up somewhere boring because you're forced to find your own interests."
So there were two versions of the same place. In one, a country was being remade by access to markets and capital. In the other, a quiet village shore offered little to do except read. The brothers grew up inside both, and they took on the conviction of the first and the self-reliance of the second.
A Language Called Croma
Each January, Ireland's Young Scientist and Technology Exhibition fills a Dublin hall with teenagers and their projects. In practice it works as a national talent search, and in 2004, at the fortieth exhibition, a fifteen-year-old from Tipperary entered with an artificial-intelligence project he had named Isaac, after Isaac Newton, whom he admired. He finished as individual runner-up.
He came back. On January 14, 2005, at sixteen, Patrick Collison won the forty-first exhibition with Croma, a programming language in the Lisp family. Lisp was, in Naughton's description, "once the lingua franca of early AI programmers." It is an old and austere language, valued for its elegance by people who care about elegance and largely ignored by everyone else. President Mary McAleese presented him with a cheque and a Waterford Crystal trophy. (The size of the cheque varies between accounts. The record has also never settled the language's spelling, which appears as Croma in some sources and Chroma in others, a minor wobble in the record that would have bothered its author.)
The win mattered more than its scale suggests. Collison later credited his first company's success to it. It also earned him a place in the 2004 EU Contest for Young Scientists, where he met a Swiss teenager named Silvana Konermann. She went on to become a biochemist and a Stanford professor. In April 2022 they married, and an Irish Independent headline described her, with some pleasure, as the woman "who beat him to top science prize when they were both teenagers." The story of the most powerful payments executive of his generation has, near its start, a moment when he lost to the person he would marry.
Croma connects to everything that came later, and so does the misjudgment it contained. Patrick wanted to understand systems from the bottom up, and he was drawn to tools that were beautiful in themselves. In a 2013 interview with Fortune, asked which technologists he admired, he divided them into two camps. One camp held visionaries like Douglas Engelbart, Ted Nelson, Alan Kay, and J.C.R. Licklider, "who inspired others to do amazing things." The other held implementers like Linus Torvalds, Dennis Ritchie, Vint Cerf, and John Carmack, "who've pulled off incredible technical accomplishments." He went on: "I'm still not sure which group is more important. I guess you need both." Then he named a bias: "I have a bias towards people who think about and have built basic infrastructure."
The best advice he said he had ever received came from Richard Hamming's lecture You and Your Research: "If you do not work on an important problem, it's unlikely you'll do important work." A sixteen-year-old who builds a programming language for a science fair is not yet working on an important problem. He is working on a beautiful one. Telling those two apart would take him about four more years, one company, and a database he wrote himself and nobody used.
The World's Most Elegantly Designed Userless Company
In 2007, back in Limerick, Patrick and John started a software company called Shuppa. The name is a phonetic spelling of siopa, the Irish word for shop. The brothers applied to Enterprise Ireland, the state agency charged with funding exactly this kind of venture, and Enterprise Ireland turned them down. "It was difficult for us to raise money in Ireland," Patrick told the BBC the following spring. "It was not just finding the money, it was also a matter of the help investors can provide for you."
Y Combinator, the Silicon Valley accelerator, showed interest, so the brothers moved to California. There they merged with a company run by two Oxford graduates, the brothers Harjeet and Kulveer Taggar, and Shuppa became Auctomatic. It sold tools to eBay "power sellers," the high-volume merchants who needed listing templates, scheduling, bulk uploads, image hosting, and inventory management. Mashable reported that Auctomatic raised more than $400,000. Its backers included Paul Buchheit, the creator and lead developer of Gmail, and the investor Chris Sacca. Patrick, the same report noted, had deferred a place at MIT to move to San Francisco with his younger brother. John was sixteen when he went. "Our parents are pretty open minded," he said in 2011, "most evidenced by – but still can't figure out how it happened – me going off to San Francisco when I was 16."
Auctomatic was a technical showpiece. Speaking at Stripe Sessions in 2024, Patrick described it with the affection of someone looking back at an old mistake. "Everything was so elegantly and carefully built. We wrote all the software in Smalltalk." John cut in to explain for the audience that Smalltalk is "an obscure programming language, even to those who appreciate obscure programming languages." Patrick went on: "And we didn't use any silly mainstream, relational database. We wrote our own distributed object data store." They had a system for saving a virtual-machine image, so that if a user hit a problem the team could restart the program later in exactly the same state.
"Nobody ever benefited from it," Patrick said at a Retool conference, "because Auctomatic had no users." He called it "the world's most elegantly designed userless company."
That self-assessment is too harsh to be fully fair, because the company did sell. On March 26, 2008, a Vancouver domain-name company called Communicate.com announced that it was buying Auctomatic for $5 million in cash and stock and renaming itself Live Current Media. The buyer was an odd one. It held about 800 domain names, including perfume.com and cricket.com, and earlier that year it had refused $6 million for cricket.com alone. Its market capitalization was a little under $60 million. Its chief executive, Geoffrey Hampson, said Auctomatic's shopping-engine technology would be central to building branded properties on those domains, beginning with Perfume.com. The Guardian reported that two large American internet companies had bid against Live Current. By Wikipedia's account, the deal closed on Good Friday. Patrick was nineteen and John was seventeen, and both were now millionaires.
Michael Arrington, who founded TechCrunch after practicing securities law and who would later become an investor, covered the deal with mild skepticism. "Will this work? I don't know and I'm not even sure I'll remember to check in on them," he wrote. He added a hope: that the Auctomatic founders, "who must leave Silicon Valley and move to Vancouver as part of the transaction, will be able to pry themselves loose at some point in the future and start another company."
In May 2008 Patrick moved to Vancouver as director of engineering. John went home to sit the Leaving Certificate, the Irish school-leaving exam, which he passed with eight A1s and two A2s. That Christmas, RTÉ put both brothers on a televised list of rich young Irish people.
One more side project from these years matters to what followed. The brothers built an iPhone app that stored a copy of Wikipedia on the phone so it could be read without a connection. They called it, with the reference intended, The Hitchhiker's Guide to the Galaxy. Each user had to download a two-gigabyte dump from their servers, and bandwidth cost money, so they charged for it. "It was more just a fun side project," Patrick told the interviewer Andrew Warner. "Then it ended up becoming really popular." Taking payments outside Apple's walled garden, for this app and for others, proved surprisingly hard: merchant accounts, gateways, delays of several working days, high fees. "It was in the course of building Auctomatic," Patrick said in 2024, "that we started to get exposed directly ourselves and also for the first time hearing stories from others about the difficulties with accepting online payments. So it's where we discovered the actual Stripe use case."
The userless company had turned up a problem that a great many users had.
The Grand Plan
Before he built the company that would make his name, Patrick Collison made a public argument to his own country, and it is revealing to read it now.
Under the headline "The grand plan," The Irish Times published an essay in which "wunderkind Patrick Collison" argued that "Ireland's tech boom was a mirage." The essay is confident and somewhat brash, and in its analysis it is close to cold. "Ireland has an abysmal record of encouraging technology companies, especially start-ups," he wrote. "This is masked by our stellar ability to attract companies for financial reasons." He set the reader a test: "Quick – try to think of an Irish technology company with revenues of more than, say, €50 million."
His central claim was that Silicon Valley's real advantage was migration, not native genius. "Not one of the founders of these companies grew up in the Valley," he wrote of Google, Yahoo, Sun, Facebook, Twitter, and Cisco. "A majority aren't even from the US." So Ireland was asking the wrong question. The right question was not "how can we encourage more Irish people to start companies?" but "how can we encourage founders (or people likely to become founders) to move to Ireland?" He borrowed a proposal from Paul Graham, the essayist and investor who had backed Auctomatic: give $1 million to each of fifty start-ups just to get them to relocate. Then he made the comparison an Irish reader would feel most. The cost, he noted, was "less than a quarter of the cost of the regeneration of Croke Park."
On July 18, 2009, aged twenty and following the publication of the McCarthy Report on public spending, he went on RTÉ's Saturday Night with Miriam to set out his ideas for Ireland's future. A young emigrant, recently enriched, explaining to the people at home on Saturday-night television why the country kept losing people like him: the scene has an ambivalence that his career never fully resolves. He loved the place, and he left it, and he kept telling it how to stop others from leaving.
That August, John left for Harvard. Lily Collison gave her middle son a sentence of advice that The Currency would later call infamous: "Do anything you want, study whatever you want, but just don't drop out of college like your brother."
Patrick had gone back to MIT after the Vancouver year to study mathematics and physics. The decision that ended his time there was made almost casually. In 2009 the brothers attended Y Combinator's Startup School in Berkeley, got sushi in Potrero Hill afterward, and decided on the walk home to start a payments company. His reasoning, as he remembers it, was that "we might as well because it probably won't be that hard." Two months after their mother's warning, the brothers were writing the first lines of code. Patrick took a second leave from MIT, and John left Harvard the following year.
"It was all downhill from there," Patrick said.
His view of that decision has changed. At Startup School in 2026, interviewed by YC's Harj Taggar, the same Taggar whose company had merged with Shuppa nineteen years earlier, Collison told students they should not rush to leave college. "When I was dropping out, people thought it was super weird," he said. "I think, overall, if you enjoy college, I think there's no harm in finishing. I felt this real sense of urgency, which I think, in hindsight, was a bit unnecessary." He had believed the opportunities were "ephemeral and fleeting," and he had operated in what he called a "speedrun" mode. "In hindsight, I think that was a poor intuition. It's been pretty robustly and reliably the case over many decades that Silicon Valley has a surfeit of opportunities." He said dropping out was "not totally a trapdoor." And with the plainness of someone who no longer has anything to prove: "as far as I can tell, nobody has ever cared."
The word is cared again. It is the same register as the outage: the fear that the world is watching closely, and the discovery that mostly it isn't.
Two Years in the Dark
In 2010 the brothers, aged twenty-one and nineteen, pitched Peter Thiel. Thiel had co-founded PayPal and was one of Facebook's earliest backers. The Financial Times later recounted that they had a prototype hacked together while on holiday and no experience of merchant gateways or the rest of the financial machinery. They told him their company would "increase the GDP of the internet." They also told him what was wrong with PayPal, the company he had built.
"I remember being very critical of PayPal," Patrick said two years later. "Halfway through the meeting I was like 'hmmm, maybe that's not the best strategy'."
Thiel invested anyway, and so did Elon Musk, his PayPal co-founder. Michael Moritz of Sequoia Capital, the former journalist who had become an early backer of Google, invested too. "It was obvious they were unusual," Moritz said. "It was a first on all counts." In 2011 the company announced a $2 million round with Thiel, Musk, Sequoia, Andreessen Horowitz, SV Angel, the Irish entrepreneur Liam Casey, and Max Levchin, the third PayPal veteran. In effect, the founders of PayPal were paying two Irish teenagers to make PayPal obsolete.
The money was the easy part. The Collisons had come to dislike an industry built on paper forms, merchant underwriting, and delays measured in "five working days." Naughton quotes one founder of that period describing "a growing wait-list of people that wanted to give us money but couldn't." Patrick put the opportunity in characteristically compressed terms: "the online payments industry was an unusually compelling example of an entire industry that is going to have its lunch eaten." But to move money you need banks, and the banks were not interested.
Collison has described his first bank meeting. He was told "in no uncertain terms" that there was "no possibility" the bank would work with Stripe. Similar refusals continued for two years. "I really feel this imperative to emphasize the long period—two years is long time when you're in it—of working away with so many of these roadblocks and headwinds and people telling us that it couldn't work or shouldn't work or was a bad idea," he said. Then he added the part founder stories usually leave out: "It could mean that you can't actually do it or the idea is bad. Or it might not."
The admission is honest. Refusal doesn't tell you anything by itself. The same silence can mean you are early or that you are wrong, and from inside it the two feel the same.
Meanwhile they were building, and building differently from before. The Smalltalk lesson had been learned. "Every time there's a super elegant way to do things and a practical, pragmatic way to do things," Patrick has said of Stripe's founding philosophy, "we're just gonna cut the corner—at least until we validate that there's actual user value here." Stripe was written in Ruby, not Lisp or Smalltalk. It ran on MongoDB, despite the database's known limitations. Fifteen years later, in a conversation with Michael Truell of Cursor, Collison said Stripe still lives with both choices. In the same conversation he named the one place he would have been less pragmatic: he would have spent even more time on API design.
That exception explains the whole approach. Internally, anything could be done the cheap way. The surface developers touched could not. The integration that became shorthand for the company, captured in the title of Ashlee Vance's 2017 Bloomberg Businessweek profile, "How Two Brothers Turned Seven Lines of Code Into a $9.2 Billion Startup," was the product. Everything underneath it could be patched later.
Stripe launched publicly in September 2011 after a long private beta, roughly two years after the walk home from the sushi place in Potrero Hill. The company that "probably won't be that hard" had taken twice as long as anyone planned. Early in that period came the morning the switch failed, the eight hours of waiting, and the silence at 1pm that told him he still mattered to no one.
The Atmosphere in Which Every Other Business Exists
There is an obvious way to describe what happened next, and it fits on a slide. Stripe moved from Palo Alto to San Francisco in 2012. That year it launched Connect, a system that let software platforms embed payments natively, which made Stripe the payments layer under other companies' businesses. By the end of 2013 it had eighty employees and had raised more than $40 million. Then came Radar for fraud, Billing, Issuing for cards, Terminal for in-person readers, Capital for merchant advances, Tax, Identity, Treasury. In November 2016 a round from CapitalG and General Catalyst valued the company at $9.2 billion and made the brothers the world's youngest self-made billionaires. By 2019 the valuation was $35 billion. That year MIT, hosting Patrick for a talk titled "The Impatient Pursuit of Progress," reported that more than 80 percent of people across the United States had made a purchase using Stripe in the preceding year.
The more interesting description concerns what Collison thought he was building. In the 2013 Fortune interview he rejected the idea that software was an industry at all. "I don't think it's really accurate to look at software as a 'sector;'" he said. "Software and the Internet are increasingly the atmosphere in which every other business exists." If that is right, then the payments layer is not a feature of the atmosphere. It is closer to its chemistry, the reaction that lets anything grow.
The mission statement has always sounded slightly ridiculous, and Collison knows it. "Our mission is to grow the GDP of the internet and that's kind of an abstract idea," he told reporters in Sydney, "and certainly not the world's catchiest slogan. But we do think that there's not enough prosperity in the world. The average global income is on the order of $10,000 a year." At a Retool conference he cited a correlation of about 0.75 between self-reported happiness and log GDP per capita. "You very rarely in sociology see correlations that are that high." In this way of thinking, growth is the variable most other good things depend on, which makes it close to a moral matter.
Atlas shows most clearly how that belief shapes the products. Launched on February 14, 2016, it let entrepreneurs anywhere incorporate a Delaware company, with a U.S. bank account and a Stripe merchant account set up automatically. It was aimed specifically at founders outside the United States. In March 2016 Cuba was added to the list of eligible countries. By mid-2026, according to Collison's own website, Atlas was incorporating around a quarter of all new Delaware corporations.
It is hard not to read Atlas as the fine print reversed. A boy in Tipperary was told by a thousand websites that the offer did not extend to him. As an adult he built a machine that tells a founder in Lagos or Lahore or Havana that it does. Stripe invested in payment processors in Nigeria (Paystack, 2018, then acquired in October 2020 for a reported $200 million-plus), the Philippines (PayMongo, 2019), and Pakistan (Safepay, 2021). "Several billion people recently immigrated to the world's most vibrant city," Collison wrote to the economist Noah Smith in 2021, describing the internet, "and the system hasn't yet equilibrated." In the same exchange he called the internet "nitrogen fertilizer" for "the landscape of the global commons."
Migration is the metaphor he reaches for: billions of people moving to a city. The man who argued in The Irish Times that the right question is how to get founders to move describes the internet as the largest act of migration in history and Stripe as the immigration office that clears new arrivals for work.
The rest of his ideas sit near this. Collison believes in infrastructure the way some people believe in providence. Asked in 2013 which companies he admired, he gave a list that reads as a description of Stripe as it would become: companies "that use time horizons as a competitive advantage," that "still do hard, basic research," that "pay attention to the details even when they're big," that "retain a specific mission that's broader than their business," and that "try to figure out how an industry should work from first principles."
The Brother Who Plays Piano
Profiles usually go wrong on one point here, so it should be said directly: Stripe has two founders. The question of which brother "is" Stripe has as little meaning as asking which blade of a pair of scissors does the cutting.
John Collison was born in August 1990 and followed his older brother into everything early. He was in San Francisco at sixteen, a millionaire at seventeen, and back in Limerick that summer to finish secondary school. Naughton later wrote that John earned the highest scores ever recorded on the Irish Leaving Certificate. Harvard announced his arrival in September 2009, and he left the following year despite his mother's instruction. He is a pilot and a pianist. He is the president of Stripe and the more public of the two. In 2025 he started a podcast called A Cheeky Pint, in which he interviews technology leaders over a beer. In 2021 and 2022 he bought the Abbeyleix Estate and the derelict Millbrook House next to it, in County Laois, and said he would spend several million euros over some years restoring Millbrook as a family home. The Times put the estimated renovation cost at €6 million.
That last detail rhymes with the rest of the story. One brother writes essays about why Ireland loses its founders. The other buys a ruined Irish country house and starts rebuilding it room by room.
On stage, their way of working together looks like a long-running comedy routine. At the 2025 Stripe Sessions keynote they opened with a staged bit. John, at a keyboard, refused to come out because he was "vibe coding." "Patrick, you're so TradFi," he said. "I want to write code that feels good to me, not just CI." Patrick, playing the straight man, explained that Andrej Karpathy's tweet about vibe coding "wasn't talking about the particulars of your sensory experience here. He was talking about the tool chain." The routine works because the audience recognizes the roles: Patrick as the anxious systems thinker, John as the cheerful one pushing things forward.
The roles are real. In the 2022 layoff email Patrick wrote "John and I are fully responsible," and "we, the founders, made this decision." When Fidji Simo, the Instacart CEO, moderated a Sessions AMA in 2024 and asked what Smalltalk was, it was John who answered for the audience. When Claire Hughes Johnson, Stripe's former chief operating officer and later a corporate advisor, opened the 2025 AMA by threatening the founders with a performance review, Patrick delivered the joke. "Feedback is a gift," he said. "And with Claire, one receives many gifts."
That year the Singleton Foundation gave its Singleton Prize, awarded to active CEOs, to John Malone and to both Collisons together, recognizing them as chief executives in the mold of Henry Singleton, the Teledyne founder known for capital allocation. The foundation could not separate them either.
In the Bloomberg interview in 2018, describing the childhood that formed them, one brother remembered that because no friends lived nearby, "we had to run around and play with each other." Thirty years later they are still doing it, only now with $1.9 trillion a year passing through the playground.
Less Bang for the Buck
On his personal website Patrick Collison keeps a page called "Bookshelf." It is a catalog of the physical books he owns, in no particular order. He admits he has read only about half of them, and he cites Umberto Eco's theory of the library to explain why that is acceptable. The list runs from The Rise and Fall of American Growth to Invisible Cities, from The Outbreak of the Peloponnesian War to Let My People Go Surfing to Orality and Literacy. As of mid-2026, a note at the top says, the page had not been updated in about ten years. It is "kept around for posterity," and he "may remove it at some stage."
Another page, "Labs," is a reading list about famous research institutions: Xerox PARC, Bell Labs, the MIT Rad Lab and Building 20, ARPA, the Manhattan Project, Lockheed's Skunk Works, the China Lake Navy lab that built the Sidewinder missile, and George Mueller's management of Apollo. At the top he asks a question that the rest of his non-Stripe work has tried to answer: "Does it just seem that their heyday is past, or has something structurally changed?" On a third page, a draft of a Silicon Valley canon, he quotes Alan Kay: computing is pop culture, and "pop culture holds a disdain for history."
Chris Sacca, who rarely makes introductions, once sent Tim Ferriss a message about Collison that Ferriss read aloud on his podcast in 2018. Sacca called him "quite literally one of the smartest people I've ever known. Like, he puts Larry Page on his heels smart. I don't know anyone who has 1) read more books and 2) has the near photographic memory for what he has read." Noah Smith tells a story that may be more revealing: Collison once threw a party "that consisted of people giving him seminar talks on topics of their choice."
The reading had a purpose. In November 2018, with the physicist and writer Michael Nielsen, Collison published "Science Is Getting Less Bang for Its Buck" in The Atlantic. It argued that rising investment in science had not produced matching increases in output: more scientists and more money, but no comparable acceleration in discovery. In July 2019, with Tyler Cowen, the George Mason economist whose Emergent Ventures grant program is built for speed, he followed with "We Need a New Science of Progress." The essay proposed an academic discipline, Progress Studies, to examine the cultural and institutional conditions that produce progress and rising living standards. "I don't understand why everyone isn't obsessed with this," he said later. "Are we on track for the flying cars or not? And if we're not, what can we do to fix it?"
Then came a test. In March 2020, as the first American lockdowns began, Collison and Cowen asked leading scientists what they needed and found, to their surprise, that funding for COVID-19 research was not readily available. "We expected the U.S.'s immense government funding systems to be unleashed, with decisions made in days if not in hours," they wrote afterward. "This is what happened during World War II, which killed fewer Americans." Instead, some of the world's leading virologists were "stuck on hold," waiting to hear whether they could repurpose existing grants.
So in early April they built Fast Grants. The application would take under thirty minutes to complete, a decision would come within forty-eight hours, and money would follow a few days later. They set it up under the Mercatus Center at George Mason University, raised initial funding, and built the website. It launched about ten days after the idea first came up. Eligibility was strict: only principal investigators already working on COVID-related research. They expected a few hundred applications. Within a week they had 4,000 serious ones, "with virtually no spam." Over 2020 they raised more than $50 million and made more than 260 grants, at under 3 percent overhead. A team of twenty mostly early-career reviewers handled more than 6,000 applications. Each funded application had at least three reviewers, but unanimity was not required. A successful NIH grant, they noted, is typically reviewed by ten to twenty scientists and program officers across three phases. "'Let's do it,'" they wrote, "was then the basic attitude."
Fast Grants was the Stripe argument applied to science. In payments the problem had been forms, underwriting, and five working days. In science it was the same problem with more prestige attached. The fix in both cases was to remove latency.
In 2021 the argument became a permanent institution. With Konermann and the bioengineer Patrick Hsu, Collison co-founded the Arc Institute, a nonprofit biomedical research organization intended, as his site puts it, to be "a new way to do biomedical research." The institution was meant to answer his own question about whether the great labs' heyday had passed. On August 6, 2026, Science published a paper by Arc researchers titled "Generative design of bacteriophages with genome language models," which Wikipedia describes as the first AI-designed viral genome. The woman who beat him at a teenage science contest in 2004 was now his co-founder in an organization designing genomes with language models.
His list of interests keeps growing. He is involved with Stripe Press, which spent two years on a new edition of Poor Charlie's Almanack that came out just after Charlie Munger's death. He backs Works in Progress, a magazine about how progress happens. He supports Frontier, Stripe's carbon-removal coalition, launched in April 2022 with Alphabet, Shopify, McKinsey, and Meta as a $925 million advance market commitment. He funds Intercept, which aims to eliminate respiratory infections. With Cowen he started New Aesthetics, a grant program prompted by the question on his website, "Why is contemporary architecture so ugly?" He has invested in California Forever and Esmeralda, projects attempting new cities. Under "Religion," his site observes: "We live in a post-Christian West, but the need for metaphysics hasn't diminished."
That is a lot for one person, and it can sound like the hobbies of a man who can afford any hobby. A more generous reading is that the projects share an editorial line: he is trying to find the bottleneck in everything, and he keeps finding it in institutions.
Embracing Reality as It Is
There is a version of this story in which nothing goes wrong. It is not the true version.
The pandemic shifted the world toward e-commerce almost overnight. Stripe's revenue and payment volume more than tripled during 2020 and 2021, and the company "transitioned into a new operating mode," in Patrick's words. In March 2021 it raised $600 million at a $95 billion valuation, making it the most valuable startup in the United States. The investors included Fidelity, Sequoia, and Ireland's National Treasury Management Agency. The detail is worth stopping on. In 2007 Enterprise Ireland had declined to fund the brothers' first company. Fourteen years later, another arm of the Irish state bought shares in their second at $95 billion.
Then the market turned. In July 2022 The Wall Street Journal reported that Stripe had cut its internal share price, lowering its implied valuation from $95 billion to $74 billion. On November 3, 2022, Patrick sent an email to every employee. Its first line was: "Today we're announcing the hardest change we have had to make at Stripe to date."
Approximately 14 percent of Stripe's employees would be let go. Affected employees would receive notification emails "within the next 15 minutes." Then came the sentence most chief executives hand to lawyers or bury in the passive voice: "For those of you leaving: we're very sorry to be taking this step and John and I are fully responsible for the decisions leading up to it." Further down: "We, the founders, made this decision. We overhired for the world we're in."
The email laid out the macroeconomic case in specific terms: stubborn inflation, energy shocks, higher interest rates, a former Treasury Secretary warning of "as complex a set of macroeconomic challenges as at any time in 75 years." It was equally specific about severance: fourteen weeks for everyone, with pay through at least February 21, 2023; the 2022 bonus paid regardless of departure date; payment for unused PTO "including in regions where that's not legally required"; six months of healthcare premiums; vesting accelerated to the February 2023 date; the one-year cliff waived for those who hadn't reached it; immigration support for visa holders; and "a new tier of extra large Stripe discounts for anyone who decides to start a new business now or in the future." The central principle took one sentence: "Doing right by our users and our shareholders (including you) means embracing reality as it is."
In March 2023 Stripe raised more than $6.5 billion in a Series I at a $50 billion valuation, roughly half its 2021 peak. The company said it did not need the money to run the business. The round was for employees, to provide liquidity and to cover withholding taxes on equity awards. That same year Stripe processed more than $1 trillion in payment volume.
Not every decision afterward went as smoothly as the 2022 email. In January 2025 Stripe laid off nearly 300 people, mostly in product, operations, and engineering, and somehow sent the departing employees a cartoon picture of a duck. The chief people officer, Rob McIntosh, apologized. A company that had built its reputation on getting the small details right got this one badly wrong.
Then the numbers went back up. The valuation was $70 billion in 2024 and $91.5 billion in February 2025. In February 2026 a tender offer for employees and shareholders valued Stripe at $159 billion, about 70 percent above the year before, with $1.9 trillion in volume for 2025. Stripe bought Bridge, a two-year-old stablecoin company, for $1.1 billion, then the crypto-wallet company Privy, then the usage-billing platform Metronome. In September 2025 it began powering Instant Checkout in ChatGPT and released an Agentic Commerce Protocol co-developed with OpenAI. On August 19, 2026, it announced the acquisition of the AI startup OpenRouter for $7.5 billion.
Six weeks earlier, on July 15, 2026, Stripe and the private-equity firm Advent International had announced a joint bid of $53 billion for PayPal.
Sixteen years after two young men sat in Peter Thiel's office criticizing PayPal and realizing halfway through that it might be a poor strategy, their company made an offer to buy it. The obvious question from the press, which Fortune called "the inexorable drumbeat," continued alongside: when will Stripe go public? As of this writing it has not. The company that once worried nobody was watching now has everyone watching, and it still keeps most of what it does private.
Stab City
On a Friday in April 2021, Forbes published a contributor's piece about the Collison brothers. Its first line was: "Some call it 'stab city'."
"Many folks think Ireland is all rolling green hills and five-star golf courses," the writer, Stephen McBride, continued. "But in the middle of the Irish countryside is a city called Limerick – known as the 'murder capital' of Europe." The piece called the city a "warzone." It claimed that shootings, pipe-bomb attacks, and stabbings happened there nightly and that some neighborhoods were "walled off by a dirty graffitied 10-foot-high barrier, like the Berlin Wall." It ended with the triumph it had been building toward: "Not only did they escape 'stab city', they moved to Silicon Valley."
The reaction in Ireland was fast. Patrick O'Donovan, a Limerick representative and minister of state, demanded an apology "for the insult and hurt caused" and invited the magazine to visit: "Please let me know when suits to visit." His colleague Niall Collins called it a "disgraceful description of Limerick, home to so many fine and decent people." John called the article "daft." Patrick wrote a sentence that reads like a short statement of identity:
— Patrick Collison, on Twitter, April 2021Not only mistaken about Limerick but the idea of 'overcoming' anything is crazy. We are who we are because we grew up where we did.
The article was removed from the site on April 9, 2021. One irony went largely unremarked except by The Irish Times, which noted dryly that the brothers had actually grown up in Dromineer, about thirty miles away. The supposed escape from stab city had in fact started at a lakeside village where the main hazard was boredom, and where the boys had been, by John's own account, "forced to find your own interests."
Origins tend to get flattened as they travel. Silicon Valley wanted a story of escape, and the Collisons refused it. Then, on December 12, 2024, Patrick appeared before the Canadian House of Commons Industry Committee and opened with this: "My co-founder and brother John and I are originally from Dublin in Ireland." It is in the official record. Perhaps the claim was a courtesy to foreigners for whom Dublin is simply Ireland. Perhaps Stripe's second headquarters, which opened in Dublin the following October, had already begun to rewrite the story. In any case, the man who had defended his origin against a magazine's caricature had, in a parliament's transcript, moved it himself.
By then his public life had acquired the weight that comes with this much money and influence. In 2018 Stripe gave $1 million to California YIMBY, the pro-housing lobby, and opposed San Francisco's Proposition C, a tax on large companies to fund homelessness services. Marc Benioff of Salesforce responded by accusing Stripe of refusing to give back at scale to the city where it had made $20 billion. Collison replied in a long statement: "With problems as complex as homelessness, there are rarely clear-cut answers. Any of us could turn out to be wrong. Today, the world is pulling us towards polarised discourse and emotionally-charged, soundbite analysis. We're all familiar with the forces at play. We think this is important to resist." In November 2024 he posted from Israel during the war in Gaza, and the post drew boycott calls against Stripe. In March 2026 the Federal Trade Commission warned Stripe and three other payment processors against "debanking" law-abiding customers. In July 2026 it was reported that he had given $7 million to a group opposing a proposed one-time wealth tax on California's billionaires. He joined Meta's board in April 2025. In 2026 he co-founded the Rhine Group, a policy forum and think tank, with Mario Draghi, the former prime minister of Italy. His own site states the premise bluntly: "Europe. It's currently off-track, falling behind the US in key technologies and general prosperity. What can be done?"
So the twenty-year-old who went on Saturday-night Irish television to tell his country how to keep its founders is now, at thirty-seven, telling a continent the same thing, and he is better placed than nearly anyone to be listened to. He lives in California and runs a company headquartered in South San Francisco and in Dublin. His brother is restoring a ruined house in Laois. His wife, the Swiss teenager who beat him at a science contest, runs research at an institute they founded together that is designing viral genomes. When the switch fails now, everyone notices.
His website opens with three plain sentences. I live in California and mainly work on Stripe. I cofounded the Arc Institute. I grew up in Ireland and previously studied at MIT.
Below them, in the place where websites put their small print, is a single line:
Email: patrick@collison.ie.
Part IIThe Playbook
The principles below come from the pattern of Patrick Collison's decisions, from Croma through the PayPal bid. Very few of them appear in his speeches. Most are inferred from what he built, what he dropped, what he admitted getting wrong, and what he repeated. Taken together, they describe an operator who values infrastructure over attention, latency over polish, and long time horizons over speed, while running one of the faster-moving companies of his era.
The principles
- 1.Learn the shape of a system from its exclusions.
- 2.Cut every elegant corner except the one the user touches.
- 3.Treat invisibility as both the goal and the threat.
- 4.Read the long silence without flattering yourself.
- 5.Sell to the builder, not the buyer.
- 6.Name the unflattering parts of your culture out loud.
- 7.Use the time horizon as the moat.
- 8.Measure institutions by their latency.
- 9.Own the cut, then pay for it generously.
- 10.Build the adjacent institutions your thesis requires.
- 11.Split the founder role along temperament, not title.
- 12.Don't mistake your urgency for the opportunity's scarcity.
Principle 1
Learn the shape of a system from its exclusions
The Collisons' first lesson about the internet was where it ended. "Offer not available in the Republic of Ireland" kept appearing in the fine print of websites viewed over a satellite link from Dromineer. Most people who hit a wall like that resent it or ignore it. The Collisons kept track of it. Twenty years later Stripe's distinctive products, Atlas above all, are close to point-by-point negations of those exclusions: incorporation for non-Americans, Cuba added in March 2016, investments in processors in Nigeria, the Philippines, and Pakistan.
What makes this a skill and not just a grievance is the precision. An exclusion tells you exactly where a system's assumptions stop. Every "not available in your region" marks a spot where the incumbent decided the cost of serving someone was greater than the revenue. If your cost structure is different, that spot is a market.
Tactic: List every place your industry tells a customer "no" in its fine print, and price what it would cost you to say "yes" to each.
Principle 2
Cut every elegant corner except the one the user touches
Auctomatic was beautiful and had no users: Smalltalk, a custom distributed object store, saved VM images. Stripe ran on Ruby and MongoDB, choices Collison admits the company still lives with fifteen years later. The rule he took from the first company was explicit. "Every time there's a super elegant way to do things and a practical, pragmatic way to do things, we're just gonna cut the corner—at least until we validate that there's actual user value here."
The rule had one exception, and it is the important part. When Collison is asked what he would redo, he says he would have spent even more time on API design. The interface is the product, and everything behind it can be fixed later. Most teams get this backwards. They polish the internals they control and neglect the surface customers actually see. Stripe put its perfectionism on the one surface that compounds, the integration that came to be summed up as seven lines of code, and was pragmatic everywhere else.
Tactic: Draw a line between what your customer touches and what they don't, and move your perfectionism to their side of it.
Principle 3
Treat invisibility as both the goal and the threat
Six months into Stripe, the API went down for eight hours and nobody noticed. Collison called this "existentially bad news." The insight runs both ways. Mature infrastructure should be invisible: Stripe's ideal end state is one where, as MIT reported in 2019, most American online shoppers pay through it without knowing. Early infrastructure that goes unnoticed, though, is infrastructure nobody needs yet.
The diagnostic question is which kind of invisibility you have. Does nobody notice you because you work perfectly, or because nobody relies on you? The answers call for opposite responses. The first calls for protecting reliability. The second calls for pushing hard toward dependence, finding the customers whose revenue would actually stop if you did.
Tactic: After every outage, measure how many customers complained, and treat a low number early in your company's life as a warning, not a relief.
Principle 4
Read the long silence without flattering yourself
For two years banks told Stripe "in no uncertain terms" that there was "no possibility" of working together. The usual founder story treats this as heroic persistence vindicated. Collison's own account is more useful because it doesn't say that. Refusal "could mean that you can't actually do it or the idea is bad. Or it might not."
The discipline lies in keeping both interpretations alive and looking for evidence that separates them. Stripe had something to look at that the banks didn't: developers who wanted the product and a waiting list of businesses that "wanted to give us money but couldn't." The banks' refusals were evidence about the banks. Developer demand was evidence about the market. Persistence made sense because the second kind of evidence was strong, not because persistence is a virtue in itself.
Tactic: When gatekeepers reject you, separately track whether end users still want the product; persist only while that second signal holds.
§
Conventional Wisdom vs. the Collison Approach
Where the Stripe playbook departs from standard startup practice.
| Conventional wisdom | Collison approach | Domain |
|---|---|---|
| Build the most elegant system you can. | Cut corners internally; obsess only over the API surface. | Product |
| Sell payments to finance departments. | Sell integration to developers, who become the champions. | Distribution |
| Describe culture in aspirational abstractions. | State the demanding parts plainly so the wrong people self-select out. | Culture |
| Go public when the market allows. | Stay private; use tenders and liquidity rounds to serve employees. | Capital |
| Blame the macro environment for layoffs. | "We overhired." Founders take responsibility by name. | Risk |
| Philanthropy is separate from the business thesis. | Build institutions (Arc, Fast Grants, Frontier) that test the same latency thesis. | Institutions |
Principle 5
Sell to the builder, not the buyer
Collison has said that most people never run into the problem of accepting payments online, because most internet users have nothing to sell. Web developers do. The Collisons had hit the problem themselves with an offline Wikipedia app that cost real money in bandwidth. That firsthand knowledge decided who the product was for. Stripe was built for the engineer who would add it to a codebase, not for the CFO who would sign a contract with a processor.
This turns the enterprise sales motion upside down. Developers adopt the tool in an afternoon, the company grows around it, and by the time a finance department notices, switching would mean rewriting code. Stripe Connect in 2012 extended the logic to platforms, so that the builders of marketplaces carried Stripe into businesses that had never chosen it directly. The buyer arrives later and finds the decision already made.
Tactic: Identify the person who implements your product, not the person who approves the budget, and design the first hour of their experience before anything else.
Principle 6
Name the unflattering parts of your culture out loud
Talking to Elad Gil about culture, Collison identified the typical failure: companies are "too precious about it, being too apologetic about it, and not treating it as dynamic and subject to revision." His alternative was to state the demanding parts bluntly. "If you work with us, you're going to have to be okay with your work being repeatedly designated as inadequate, and okay with it being redone several times over."
The reasoning is selection, not motivation. A vague value like "commitment" attracts everyone and filters nobody. A specific and somewhat off-putting description filters hard. Collison ties the standard to the stakes of the domain. At Berkeley Haas in 2024 he said, "if we make a mistake—just one mistake—there's a very good chance that somebody's paycheck is wrong." The duck cartoon sent to laid-off employees in 2025 is the counterexample that proves the rule: when a culture known for detail slips, everyone sees it.
Tactic: Rewrite your values document so that at least one line would make a reasonable candidate decide not to apply.
Principle 7
Use the time horizon as the moat
In 2013, with eighty employees, Collison said he admired companies "that use time horizons as a competitive advantage — doing things that pay off over a longer timeline than competitors are willing to wait." Stripe's structure follows from this. It has stayed private through the "inexorable drumbeat" of questions about an IPO. It raised a $6.5 billion round that it said it didn't need, to provide employee liquidity. Frontier was designed as a nine-year advance market commitment for carbon removal. Arc is an institute whose results will be measured over decades.
Staying private has a cost, and Collison pays it on purpose. A public market demands quarterly answers. A private company with patient investors can build Atlas, Issuing, Treasury, and stablecoin infrastructure before any of them earns its keep. A long horizon is not the same as moving slowly: Stripe ships constantly. What it buys is the freedom to be wrong for longer than competitors can afford.
Tactic: Pick one initiative whose payoff lies beyond your competitors' planning horizon, and protect its budget from quarterly review.
Principle 8
Measure institutions by their latency
Fast Grants is Collison's clearest statement of what he thinks is broken in the world, and it took ten days to build. The diagnosis was not a lack of money in science but slow movement of money. The NIH had large budgets, and scientists were "stuck on hold." Fast Grants specified latency first: under thirty minutes to apply, forty-eight hours to a decision. It achieved that with twenty reviewers, no unanimity requirement, and under 3 percent overhead.
The same diagnosis runs through Stripe. Before Stripe, merchant accounts took "five working days" and pages of forms. The common thread is that latency is a hidden tax on every downstream activity. A scientist waiting months for a grant decision and a founder waiting weeks to accept a payment are paying the same tax in different currencies. Collison's view of the world, from the Atlantic essays to Progress Studies, treats institutional slowness as a variable that can be measured and fixed, not as an unavoidable fact.
Tactic: For your company's three most important decisions or approvals, measure elapsed time from request to answer, and set a target that is one-tenth of the current figure.
⏱
A Decade-by-Decade Ledger of Decisions
Key choices that shaped the Collison trajectory.
2005
Wins Ireland's Young Scientist exhibition with Croma, a Lisp-family language.
2007
Enterprise Ireland declines to fund Shuppa; the brothers move to California via Y Combinator.
2008
Auctomatic sold to Live Current Media for $5 million in cash and stock on Good Friday.
2009
Startup School, sushi in Potrero Hill, and the walk home on which Stripe is decided.
2011
Public launch in September after two years of bank refusals.
2016
Atlas launches for foreign founders; $9.2 billion valuation makes the brothers billionaires.
2020
Fast Grants launched about ten days after the idea; Paystack acquired.
2021
$95 billion valuation; Arc Institute co-founded with Konermann and Hsu.
2022
Internal valuation cut; 14% of staff laid off; "John and I are fully responsible."
2026
$159 billion tender; $53 billion joint bid for PayPal; OpenRouter acquired for $7.5 billion.
Principle 9
Own the cut, then pay for it generously
The November 2022 layoff email holds up as a model because it refuses the usual evasions. It names the founders as responsible ("We, the founders, made this decision"). It names the cause as their own mistake ("We overhired for the world we're in") and not only the macro environment, which it describes in detail elsewhere. It gives affected employees a fifteen-minute window, so nobody is left guessing.
Then it spells out the cost in full: fourteen weeks of severance, the full bonus, PTO paid out "including in regions where that's not legally required," vesting cliffs waived, immigration support. Taking responsibility with no generosity attached is empty, and generosity with no responsibility looks like a bribe. Together they protect the trust of the people who stay, who are reading the email as a preview of how they will be treated. The phrase "embracing reality as it is" works as an operating principle only when the reality being embraced includes your own error.
Tactic: In any painful announcement, put the sentence naming who is responsible in the first paragraph, and make the support package more generous than the law requires in at least one specific way.
Principle 10
Build the adjacent institutions your thesis requires
Collison's thesis is that economic growth is the most important variable and that institutions now constrain it. A thesis like that can't be carried out by a payments company alone. So around Stripe he has built, funded, or co-founded a set of institutions: Stripe Press and Works in Progress to shape ideas, Progress Studies to give them an academic home, Fast Grants and Arc to test institutional reform in science, Frontier to create a market for carbon removal, New Aesthetics to fund beauty, and city-building investments to test urbanism.
These are not distractions from the business. They form an ecosystem for its thesis. More founders, more science, and more growth mean more economic activity on the internet, and more of that activity runs through Stripe. Collison describes this plainly: Stripe takes a cut of each transaction, so "the more startups and founders that are building and selling," the better for business. Here the mission and the margin point the same way.
Tactic: Write down the three conditions outside your company that would most accelerate your market, and fund or build one of them directly.
Principle 11
Split the founder role along temperament, not title
The Collison partnership works because the brothers don't divide work by function so much as by temperament. Patrick is the systems thinker, the one who writes long emails and takes the institutional responsibility. John is the public extrovert, the podcaster, the one who plays the role of pushing things forward in the onstage skits. They grew up forty minutes from any friends and "had to run around and play with each other," and that childhood shows in how easily they trade lines.
The broader lesson is that co-founder conflict often comes from two people with the same temperament fighting over the same territory. The Collisons share a mission and differ in style, which lets each cover what the other doesn't. Claire Hughes Johnson's teasing about "performance reviews" and Patrick's reply about "many gifts" of feedback show a partnership comfortable enough to joke about its own management.
Tactic: Have each co-founder list the work they would do for free and the work they dread, and assign roles where those lists complement each other.
Principle 12
Don't mistake your urgency for the opportunity's scarcity
The most useful thing Collison has said recently is a correction of his own legend. He dropped out of MIT twice in what he now calls "speedrun" mode, convinced that Silicon Valley's opportunities were "ephemeral and fleeting." "In hindsight," he told Startup School in 2026, "I think that was a poor intuition." Opportunity, he concluded, has been "robustly and reliably" plentiful.
Urgency is useful when it comes from the problem: COVID scientists stuck waiting for funding, merchants stuck waiting for payments. It is harmful when it comes from the founder's fear that the world will move on without them. The first kind produced Fast Grants in ten days. The second kind produces rushed decisions dressed up as boldness. The test is whether the deadline is real or invented. Collison's own career shows that the "it probably won't be that hard" kind of urgency carries a cost, since the idea took two years longer than he thought it would.
Tactic: Before acting on a sense of urgency, write down the specific external event that would close the window, and if you cannot name one, slow down.
Free playbook
Get The Business Model Playbook
58 business models, one visual page each: how the money flows, the metrics that matter, and who runs it. Free when you join the Faster Than Normal email.
Free. No spam. Unsubscribe anytime.
Part IIIQuotes and Maxims
In their words
— Patrick Collison, quoting Richard Hamming, Fortune, December 2013If you do not work on an important problem, it's unlikely you'll do important work.
— Patrick Collison, The Tim Ferriss Show, December 2018If people around you don't think what you're doing is a bit strange, maybe it's not strange enough.
— Patrick Collison, Stripe Sessions opening keynote, 2025It's not nice ideas cooked up in the science lab that changed the world; it's good ideas that are painstakingly and steadily deployed and distributed and scaled. And this process of entrepreneurship, it's the greatest force for welfare improvement in human history.
— Patrick Collison, email to Stripe employees, November 2022Doing right by our users and our shareholders (including you) means embracing reality as it is.
— Patrick Collison, patrickcollison.com/adviceIf you think something is important but people older than you don't hold it in high regard, there's a reasonable chance that you're right and they're wrong. Status lags by a generation or more.
Maxims
- Read the fine print as a map. Every exclusion an incumbent publishes marks exactly where a new market starts.
- Elegance is for the interface. Cut corners everywhere the user can't see, and don't cut any where they can.
- Silence after failure is data. If nobody notices when you break, nobody depends on you yet.
- Rejection is ambiguous. A gatekeeper's no tells you about the gatekeeper. Only customer demand tells you about the idea.
- Sell to the person who implements. Win the builder in an afternoon and the buyer follows years later, with switching costs already in place.
- Say the hard part of the culture plainly. Vague values attract everyone. Specific demands select the right people.
- Patience is a competitive weapon. Pursue the projects that pay off after your competitors have stopped waiting.
- Latency is a tax on everything downstream. Measure how long your institution takes to say yes, then cut that time by an order of magnitude.
- Name yourself in the bad news. Responsibility stated in the first paragraph and backed by generous support is what keeps the trust of those who stay.
- We are who we are because of where we grew up. Don't let anyone recast your origin as something you escaped.
