Part IThe Story
Fifth and Broadway
The transmissions came in staccato bursts, and the boy in the smallest bedroom of the house in St. Louis wrote them down. He had chosen that room himself, the smallest one, over the objections of parents who did not quite know what to make of their eldest son. "He was always a minimalist," his mother, Marcia, would say decades later. The radio had been his father's. Tim Dorsey worked in the service department of a company that built mass spectrometers, and one of his favorite pastimes was driving down to the Mississippi to listen to the CB chatter of barges moving along the river. His son borrowed the set, added a police scanner, installed both in his room, and found the barges unbearable. "It's the most boring thing in the world," Jack Dorsey said. The scanner was different.
"They were reporting constantly," he remembered, "and they're reporting three things usually. No. 1, where they are. No. 2, where they're going. And, No. 3, what they're doing. So, for an ambulance in St. Louis: 'I'm at Fifth and Broadway, I'm going to St. John's Mercy, patient in cardiac arrest.'"
Hold on to that sentence. Almost everything Jack Dorsey has built in the forty years since (a social network that reshaped the politics of several continents, a payments company that processed $241 billion in a single year, a scatter of protocols and wallets and mesh-networked messaging apps that work without the internet at all) can be read as an attempt to complete it. Where they are. Where they're going. What they're doing. He was eight when his father bought him an IBM PCjr, in 1984. Three years later there was a Macintosh. St. Louis was, improbably, a technological city then: McDonnell Douglas was there, Southwestern Bell was there, the hacker magazine Phrack was published there, and Washington University let locals onto its computer network. Dorsey tapped in. He fell for Internet Relay Chat ("It was like a big Internet CB radio") and taught himself C by watching programmers trade code in the channels. Then he wrote a program that took the movements of the vehicles on the scanner and drew them on a screen.
"It's a rush," he said. "You forget to sleep. You forget to eat. I just felt so great, because, Oh, I can actually build something that enables me to see the city."
The map was beautiful and it was wrong, and he knew it was wrong almost immediately.
— Jack Dorsey, to The New Yorker, 2013I could see ambulances, I could see black cars, but I was missing the individuals.
The rest of this story concerns what happened when he put the individuals in, and then what happened when there turned out to be half a billion of them, and they would not stay on the map the way the ambulances did.
A word about the boy first, since he will not volunteer much. He had a speech impediment. "When you're doing something and you're doing it wrong, you usually either stop it or try to fix it," he said. "I did a bit of both." He developed an aversion to casual talk and then, characteristically, joined the debate team and the speech club, the two places in a Catholic high school where talking is least casual. When he was eight or nine his father took him to a Cardinals football game. "He brought a book," Tim recalled, adding, "We don't know where he came from!" At Bishop DuBourg High School in south St. Louis he ate lunch at the same table every day with the same two friends, and co-hosted a closed-circuit television show covering, as he put it, the "status of everything that was happening at the school." One of those friends, Charlie Kelly, never could work out how someone so quiet knew so much. "I'm a great listener, I guess," Dorsey said. "I like to know what's going on." Another, Tim Brouk, called him one of "the more popular unpopular guys in our class."
He also ran the class football pool. He printed the sheets on Monday, tallied the results on Sunday night, and distributed the forms in plastic bags his father brought home from work, sealed shut with a hot sealer. "I don't know why I loved doing it," he said, "but it was a little system that I could control and play with."
Control and play. Systems and status. A radio that reported positions and a teenager who reported on the school. Watching and listening. He didn't drink until he was twenty-two.
By the Numbers
The Dispatcher's Ledger
12thAccount created on Twitter, @jack; the first eleven were tests
140Original character limit, carved from SMS's 160 to leave room for a username
2.75%Square's single flat swipe fee, offered to every merchant in 2013
$1BSquare equity moved to Start Small in April 2020, about 28% of his wealth
$978MDorsey's projected payout from the 2022 Musk buyout, which he rolled into the private company
$24.2BBlock revenue in 2025
4,000+Block employees laid off in February 2026, out of roughly 10,000
A Hole in the Software
There is a particular kind of introvert whose way into the world is breaking into it. In 1995 Dorsey enrolled at the University of Missouri–Rolla to study computer science and math, and he found the theory dull. "I wanted to build more instead of learn more," he said. He kept working on the dispatch map. During his junior year he came across the website of Dispatch Management Services, a company that managed bicycle messengers in New York, hacked into its network to see who ran the place, and then contacted the chairman to report his findings. "You have a hole in your software," he told Greg Kidd.
Kidd belongs to a recognizable Silicon Valley type, the operator with a patron's instinct: he was the chairman of a startup trying to democratize bike-messenger work by letting any courier bid on any job, and when a stranger from rural Missouri broke into his systems, his response was to offer the intruder a job. He was more worried about the stranger's mother than his own security. "I was worried about his mother," Kidd recalled. He asked whether it was all right for Dorsey to leave school. Dorsey said the school would be fine, by which he meant that the computing class where he was a teaching assistant could get along without him. "It was the first of many miscommunications I had with Jack," Kidd said, laughing.
Kidd offered him lodging in one of two company leases: a house in Great Neck or an apartment on John Street in lower Manhattan whose windows faced an airshaft. Dorsey took the airshaft. "I got to take a subway every morning to work!" He was paid six hundred dollars a week and spent it at CBGB and Coney Island High. He enrolled at NYU that fall, a concession, he said, to "parental pressure," and wrote the software that ran Kidd's bidding system. The two became close in the way people become close to Dorsey, which is to say by spending time beside him. "I don't know what it means to hang out with Jack Dorsey," Kidd said. "He doesn't talk a lot—he listens and he's thoughtful."
Dorsey would later describe his attraction to that New York dispatch system in language that sounds, in hindsight, like a founding charter. "They're all reporting constantly where they are and what work they're doing," he said of the couriers. "I thought that abstraction was so cool that I wanted that same thing for my friends." It was at NYU, the record shows, that he first came up with the idea that would become Twitter. It would take him seven more years and a great deal of failure to build it.
The failure came first. Dorsey and Kidd wanted a bigger presence on the web, found the company resistant, and left together. At the end of 1999 they moved to San Francisco to start a web-based dispatch firm; by 2000 Dorsey was operating out of Oakland, building software to route couriers, taxis, and emergency services from the browser, sketching side projects that included networks of medical devices and something he called a "frictionless service market." He withdrew from NYU a semester short of a degree. He dyed his hair into dreadlocks that ranged, over time, from bright blue to peroxide orange, the look of the bike messengers he wrote software for. The board wanted sales. Dorsey wanted to hone the technology. The board won, Dorsey was fired, and the company died in the dot-com bust.
In July 2000, somewhere in all of this, he'd had another idea, and this time he wrote it down. He had been watching AOL Instant Messenger, in particular the little "Away" tags that told you when a friend wasn't available, and LiveJournal, and he had a new device, the RIM 850, an early BlackBerry that could do email. On a pad of lined paper he wrote two sample messages a person might broadcast to a list of contacts: "in bed" and "going to park." Under them he wrote a prompt: "know someone?" He coded it up and called it STAT.US. Then he walked out to Golden Gate Park and informed his friends, via the device, "I'm at the bison paddock."
"No one had a BlackBerry," he said, "and no one cared."
He followed women to Richmond, Virginia, and to Cambridge, Massachusetts. Neither relationship lasted. Asked about it, he offered the kind of self-diagnosis that is both candid and evasive at once: "I think it can be, in the worst case, isolating to girlfriends and friends." And then: "Mainly, the relationships ended more just from the feeling that this isn't really meant to be. I'm sure the independence thing and being in my head added up, though."
In 2002, at twenty-six, he went home.
A Residency in Failure
"I felt like a failure," he said. "Here I was, back in St. Louis."
What a failure does in St. Louis, it turns out, is a great many things at once, none of them quite a job. He coded for his father, who had by then started his own mass-spectrometer business. He enrolled in botanical illustration. He spent afternoons at the Missouri Botanical Garden doing drawing studies of a banyan tree, a plant whose roots fork down from the branches and then become trunks themselves, a structure in which every part supports the whole. He would tell an interviewer a decade later that he loved ferns because "they're a single structure that tends to repeat itself. They're fractal." He loved aspens for a related reason. "Aspen trees grow in groups. If one of them dies, they all suffer. I think humans have the same thing, though it's not as much on the surface."
His wrists ached "from years of programming." He tried keyboards designed to minimize finger movement; they didn't help. So he decided to become a massage therapist. "When something interests me, I don't just read about it," he said. "I do it. I go all in." The plan was a chair-massage service for programmers in San Francisco, a place that would give them, in his words, "code therapy and massage therapy." He promised he was serious. When he got back to San Francisco in 2005, he discovered that "everyone was a massage therapist," and dropped it.
Around then he got a tattoo on his left forearm, an elongated S that stood for three things at once: the f-hole of a violin; the collarbone, which he considers the most beautiful bone in the body; and the integral symbol from calculus. He also got a nose ring. "That really hurt," he said. "That was a mistake."
He did not want to go back to programming full-time. Greg Kidd offered him a deal: room and board in exchange for looking after Kidd's young daughter. Dorsey moved into a shed behind Kidd's house on the Berkeley–Oakland border. It is hard to picture a less likely nanny than a reticent twenty-eight-year-old with a nose ring and an integral sign on his arm, and Kidd says he was excellent at it. "He had that ability to be very present, and really small kids know whether you're present." At night, in the shed, Dorsey started coding again, rediscovering what he called "the purity of making cool little things." He found a gig on Craigslist rewriting the fraud-plagued ticketing system for the tourist ferry to Alcatraz, and wrote a more secure version.
The little things kept circling the same absence. He built a program that told Google users who else was searching for the same thing at the same moment and let them talk to each other. Kidd was fascinated: "It suggested you could use Google not just to find information but to find people." He built postcard x, which handed you the name and address of a stranger who wanted to receive a postcard. "About ten thousand people did it." He wrote a program that let you leave a note that vanished once the recipient read it. "I loved the impermanence of that," he said.
He took figure drawing because he'd read that Scott Morrison, a jeans maker he admired, preferred designers with that training. "He approached denim like it was a living journal," Dorsey said. He took fashion-design classes at Apparel Arts in the Dogpatch. He designed a pencil skirt, and an asymmetrical one. He put them in a closet. Eventually he threw them out.
It is tempting to treat this period as a detour, the wilderness years before the real story. It wasn't. The banyan and the fern, the vanishing note, the postcard to a stranger, the "living journal" of denim: these are the materials, and he was laying them out on the table. The adult Dorsey would say that Twitter and Square were each born in a slump: "We created Twitter in a depression, or funk, in the Odeo office, and Square in the heat of a financial crisis. It allows you to be creative, a depressive moment." He has a habit of turning his life into aphorism, and he is more interesting than his aphorisms. But that one he earned in the shed.
Discovery, Not Invention
One day in 2005, sitting in a café in San Francisco's South Park, he saw Evan Williams.
Williams was then thirty-three, a farm kid from Nebraska who had co-founded Blogger, sold it to Google two years earlier, and was now helping run a podcasting startup called Odeo with a man named Noah Glass. In the Valley he was already a figure, someone whose name was attached to a genuine shift in how people wrote online. "I could care less about podcasting, but I liked his style," Dorsey said. "He had an eye for good Web design. And he came from the Midwest—a farm boy from Nebraska." Dorsey sent in a résumé listing his name only as "Jack." Williams was not especially impressed, but he needed a coder.
Dom Sagolla, another Odeo engineer, remembered Dorsey's status precisely. "He was a grunt, like me. He wasn't even a lead developer. He was a line-level engineer." Dorsey spent a month trying to seem interested in podcasting. Then Apple announced a version of iTunes with a built-in podcast store, and Odeo began to look doomed.
Williams called a series of hackathons, open invitations for the programmers to build anything that might give the company a new direction. A colleague had introduced Dorsey to SMS, the text-messaging protocol, and Dorsey fell for it for exactly the reasons a dispatcher would. "It was rough around the edges, cheap, and it was on every single device out there." Texting had never caught on in America the way it had elsewhere, but that was changing fast. MySpace was booming, as was LiveJournal, where Dorsey had done contract work. "It was the beginning of the participatory Internet," Biz Stone, then at Odeo, remembered.
The founding scene, which Dorsey has told many times, takes place on a playground slide in South Park. He was sitting with Sagolla and Florian Weber, a German programmer, and he put it to them: "What if we used S.M.S. to report what you're doing, and also to receive news of what everyone else is doing?" Sagolla asked what this could do that audio couldn't. "What if you're in a loud night club?" Dorsey said. He was candid, later, about whose problem he was solving. "I wasn't considering what everyone else wanted. I was considering what I wanted."
Because Sagolla was the better public speaker, he presented the idea to the company. Not everyone saw why it beat the other SMS proposals floating around, including one with the unforgettable name Off Da Chainz. Dorsey considered leaving. What saved the project was Noah Glass, Odeo's co-founder, whom Sagolla remembered as "handsome, well spoken, erudite, witty, and loud." Glass threw his weight behind it, and under his supervision Dorsey and Weber built the code in two weeks.
The team thought about calling it Jitter, or Twitch, after the buzz a phone makes when a text arrives. Glass pulled out a dictionary. Dorsey later recalled what they found there: "The definition was 'a short burst of inconsequential information,' and 'chirps from birds.' And that's exactly what the product was." The domain belonged to a bird website; Williams eventually bought it for $7,500, but the service launched under the vowelless name twttr, in deference to Flickr. Because SMS capped messages at 160 characters and each post needed room for a username, posts were capped at 140.
On March 21, 2006, the first message went out from Dorsey's account: "just setting up my twttr." Twelve minutes later came the second: "inviting coworkers." Lowercase, he explained, because he "was into e.e. cummings." His handle was the twelfth account on the system, but the first eleven were tests, which made him, technically, Twitter's first user. At an Odeo presentation he described the product as "R.S.S. for S.M.S.," which was accurate and unlovely and exactly what a dispatcher would say.
Here the story divides, as founding stories do, into several stories, each told by a man who thinks he was the hero of it. Evan Henshaw-Plath, an Odeo engineer everyone called Rabble, and his colleague Blaine Cook were anarchists who had built TXTMob, software for coordinating protesters at the 2004 Republican National Convention; they wanted the new thing to work like that. Williams, according to Rabble, "wanted it to work like Blogger, Jack wanted the dispatch stuff, and Noah was passionate about the way we communicate." Glass's version, as told to the Times reporter Nick Bilton for Hatching Twitter, has him sobering up in a car after a long night out with Dorsey, in the middle of a painful divorce, thinking about how alone he felt, and realizing that the cold concept of status was really about connection. Others who were there are skeptical. "Noah never said Twitter should be about conversations!" Weber said. Glass's own tweets were mostly status reports. July 31, 2006: "at bestbuy … buying."
Dorsey's account of the car ride is that Glass "finally got" what he and Weber had been doing. On the specific question of conversation, Dorsey has the better evidence: the service had no way to address another person until November 2006, when a user named Robert Andersen began putting an @ in front of names. "If you look at all the early tweets, there are no conversations until people started using the @ symbol," Dorsey said. The hashtag came from a user, too: Chris Messina, in August 2007. The word "tweet" came from a third-party app developer and Blaine Cook. A commenter on the Y Combinator blog put it in five words that should be engraved somewhere: Twitter was "more of a discovery than an invention."
Hold that thought as well. It will matter in 2022.
Glass, meanwhile, came apart. He asked Williams if he could spin the project out as its own company; Williams said no. Sagolla remembered him screaming over a proposed logo: "No birds. No more fucking birds!" Zachary, an investor in both Odeo and Twitter, recalled him "going from elated to being despondent pretty quickly." Bilton reported that Dorsey threatened to quit unless Glass went. Dorsey says otherwise. "I didn't give an ultimatum. . . . I didn't have that leverage. Ev made his decision." His recollection is that Williams asked him whether they should let Glass go, and he answered, "I don't think I can work with him in his current state." By mid-August 2006, Glass was out. Williams later gave him some stock. His Twitter bio, for years afterward, read: "I started this."
Sometime later, Twitter's homepage featured a cute white bird.
By fall 2006 the service had fewer than five thousand registered users. In October, Williams, Stone, Dorsey and others formed Obvious Corporation and bought Odeo and its assets, Twitter included, back from the investors. Then, in March 2007, at South by Southwest in Austin, the company mounted two 60-inch plasma screens in the conference hallways, streaming tweets. Attendees who signed up saw their own messages appear on the screens in large type. Insiders performing for insiders. Daily tweets went from twenty thousand to sixty thousand during the festival. "Watching myself on twitter feed," one attendee posted, and that was the whole pitch. A month earlier, Dorsey had tweeted something that sounded grandiose: "One could change the world with one hundred and forty characters." In April 2007 Twitter spun off as its own company, and Williams named Dorsey chief executive, calling him "the genius behind the Twitter idea." According to one insider, Dorsey received twenty percent of the company.
It is worth pausing on what the ex-courier dispatcher had built. Not a social network, exactly; Twitter dropped its early "friends" function for a simpler, stranger mechanic in which anyone could follow anyone, strangers included. "The way we saw it," Dorsey said, "is that you write on the wall and they choose to follow what you're doing for a while, and they can always leave." This wasn't a circle of friends. It was a scanner, and everyone had one, and everyone was broadcasting. The question on the homepage was the dispatcher's question: What are you doing?
His mother loved it. "He's not the kind of guy who calls every day," she said. "With Twitter, we could see what he was doing."
The Count of Monte Cristo
The trouble with building a scanner the whole world wants to listen to is that the radio tower falls down.
Williams called the company in those years "a six-foot-tall sixth grader." The service crashed so often that its error page, a whale hoisted aloft by a flock of little birds, became famous in its own right as the Fail Whale. "We weren't changing the world when we were doing this thing," Stone said. "We were in an office that had rats in the basement, our team was fighting with each other, and everyone thought it was stupid." Dorsey remembered the two anarchist engineers refusing to sit during sit-down meetings and refusing to stand during stand-ups. Nobody knew what the thing was, open platform or closed system, utility or business. "For a long time, the world didn't know what Twitter was," one founder said, "and neither did we, exactly." Over the years it would go back to investors for some seven hundred million dollars.
Dorsey, as CEO, did two consequential things and one fatal thing. He raised money: a June 2007 round led by Fred Wilson of Union Square Ventures, five million dollars at a twenty-million-dollar valuation, and a second round in June 2008 that brought in Spark Capital and, through Bezos Expeditions, Jeff Bezos. He made uptime the top priority, ahead of revenue, which Twitter was not, as of 2008, designed to earn; he talked vaguely of commercial accounts and the API as paths to paid features. And he named three guiding principles that he said the company shared: simplicity, constraint, and craftsmanship.
The fatal thing was that he kept leaving early. The thirty-odd employees were irritated, in particular, by his habit of heading off to yoga classes. (Dorsey insisted he always went back to work later in the evening.) He had a CEO coach for a while. But his appetite for life as what D. T. Max of The New Yorker called "a shaggy polymath" kept running into the job. "There were a lot of things we could have done better," Dorsey would concede. "The system was going down a lot. We didn't have any instrumentation." He followed that with an explanation that, as Max noted drily, edited out his own failings: "A lot of that was because we were coming out of another company."
The financial crisis hit in September 2008. On October 16, the board, urged on by Williams, its largest shareholder, removed Dorsey and installed Williams. Dorsey kept the title of chairman, but it was purely ceremonial; he ceded his voting rights to Williams. He was, by every account, deeply hurt. "He didn't want to be C.E.O., and then he did," he said of Williams. Williams said he'd acted because Dorsey simply wasn't ready. It turned out Williams wasn't either. In October 2010 he stepped aside for his chief operating officer, Dick Costolo, and in March 2011 Dorsey returned as executive chairman.
What happened in between is one of the more interesting passages in the record, because it is the closest Dorsey comes to admitting to an emotion. Williams's allies believed Dorsey had engineered the reversal. "They think he worked on the board for two years, like the Count of Monte Cristo, to get his revenge," Stone said. Dorsey denied causing it; Williams was the biggest shareholder, he pointed out, and "I didn't have that much control." But he allowed that he might have been working the angles. "I was chairman. Many people were coming to me. I would say, 'You should bring that up with the board, not just to me.'" And then, unprompted:
— Jack Dorsey, on Evan Williams's ouster, to The New Yorker, 2013Was I thinking, Screw Ev? Emotionally, was I asking that? I don't know. Maybe.
It was in those exile years, as a figurehead without a desk, that the thing he'd built began reporting news no newsroom had. In January 2009 a ferry passenger named Janis Krums photographed a US Airways jet floating in the Hudson, passengers standing on its wings, and wrote: "There's a plane in the Hudson. I'm on the ferry going to pick up the people. Crazy." Dorsey said that was when he finally understood Twitter as a news source. Read Krums's tweet again with the police scanner in mind. I'm on the ferry. Going to pick up the people. Crazy. Where he is. Where he's going. What he's doing. The ambulance had become an individual.
The State Department took notice before Wall Street did. In April 2009 Dorsey went to Baghdad on a tech delegation led by a young official named Jared Cohen; in February 2010 he went to Moscow on another. In June 2009, as Iranians protested a disputed election and coordinated on Twitter, Cohen asked the company to postpone a scheduled maintenance shutdown. Twitter did. Because President Obama had pledged not to meddle in Iran, the move set off a minor diplomatic furor, and gave the world its first glimpse of a question that would consume Dorsey a decade later: who, exactly, was responsible for the radio?
Meanwhile, he went back to his life. He traded his 450-square-foot apartment near the office for a 1,300-square-foot loft in the same building. He dated a dancer in the San Francisco Ballet. He considered joining a venture firm and decided against it. Skeptics had a tidy story about him: the handsome frontman who got lucky with one simple idea, bumbled along while it surged, and then claimed more credit than he deserved. "People would say to me," he recalled, "'It's just a tool so people can update about what they had for breakfast.'"
He needed a second hit. He needed, specifically, a problem that affected him personally. "I'm somewhat selfish in that regard," he said. "I want to use these things in the world, and you make the bet that other people want to use them."
Payments Are Very Intimate
Jim McKelvey had known Jack Dorsey since Dorsey was a teenager. McKelvey was the other kind of St. Louis kid: Ladue High School, an economics and computer science degree from Washington University, a self-published debugging handbook for UCSD and Apple Pascal at twenty-one. He spent the late 1980s contracting for IBM and teaching glassblowing at the same time, and in 1989 he and a group of Washington University engineers started a company in St. Louis whose first product, a commercial document-imaging system, was built partly by a summer intern named Jack Dorsey. By the 2000s McKelvey ran a glass studio and gallery called Third Degree Glass Factory and had published The Art of Fire: Beginning Glassblowing. He was, in short, an engineer who had decided to make beautiful things with his hands, and Dorsey, who had spent his own wilderness years doing figure drawing and designing skirts, recognized the type.
The story of Square begins with a sale McKelvey couldn't make. A buyer wanted a piece of his glasswork, a faucet and fittings, for two thousand dollars. She wanted to pay by credit card. McKelvey couldn't accept one. He lost the sale.
When Dorsey talked it over with Greg Kidd, Kidd recalled, "He said one of those Jack things. 'You know, Greg, payments are very intimate.' That's all he said."
It's a funny line, and it's also a real insight. Dorsey began to think of payments not as money but as "commerce, which is conversation," and the analysis that followed was a dispatcher's analysis. He and McKelvey studied how card payments worked and found the system muddled, expensive and ugly: hulking registers, mounds of paper receipts, hidden fees, credit checks that insulted the small merchants they screened. Dorsey had spent much of his youth in debt and "hated the credit-card companies," and the project gave him pause. "Do I really want to support more of this instrument that leads a lot of people to bad financial health?" But it was another field that could be democratized, the way Kidd had tried to democratize messenger work, and another chance to build something sturdy and elegant.
Square was incorporated on February 14, 2009, Valentine's Day, which seems right for a company whose founding premise was that payments are intimate. That same month the team realized it could attach a card reader to an iPhone through the headphone jack, piggybacking on the phone's processing power and sidestepping Apple's patented dock connector. "It was faster, more reliable, and royalty-free," McKelvey said. Dorsey worked on software, McKelvey on hardware. They paid the early team partly by selling some of Dorsey's Twitter shares. In May 2009 Dorsey tweeted: "Getting ready to embark on something new and entirely different. Excited!"
The first name was Squirrel, suggesting small-bore savings, and the first reader was shaped like an acorn. "A cute idea, but probably not a great one," Dorsey said. They renamed it and reshaped it. Once they had a prototype, Dorsey invited his old Odeo colleague Sagolla to Blue Bottle Coffee, grilled him about his spending habits, asked to see his credit card, plugged the reader into his phone, swiped the card, and announced that Sagolla had just bought him breakfast. He tried the same trick in New York on Mayor Michael Bloomberg. Bloomberg demanded his money back.
Venture capitalists were not eager. The country was in a deep recession, payments were not what anyone expected from a social-media guy, and Dorsey insisting on being CEO, not just founder, was its own risk. "People weren't really sure why I left Twitter in the first place," he said. "And they saw me as having created something that was simultaneously big but also, possibly, trivial." So he built a slide titled "140 Reasons Square Will Fail." It worked. Evan Williams, of all people, connected him to his first equity investor.
Dorsey liked that it was hard. He said the difficulty satisfied him the way romance does. "I thrive on tension," he said. "If I had a relationship where nothing was ever wrong and we were never debating or arguing, I don't think we would ever grow. There's no change."
Square launched in May 2010. It had ten employees in December 2009 and more than a hundred by June 2011. That year the white plastic reader, a little square with a slot, entered the collection of the Museum of Modern Art. McKelvey stepped down as chairman in 2010 and went back to St. Louis to start a coding nonprofit and, eventually, to chair the board of the Federal Reserve Bank of St. Louis. When Dorsey described the company's focus, he traced it back to his mother's café, where as a teenager he'd made his first cappuccino, spilling milk everywhere and short on espresso; the customer said it was the best cappuccino ever. Beat. "St. Louis didn't have a lot of cafés back then."
"Our niche," he said, "is countertops and farm stands."
He priced it like a dispatcher, too. Visa and MasterCard had a dizzying schedule of swipe rates. Square charged every merchant the same thing: 2.75 percent. He hired almost nobody from the industry he meant to replace. "Of the six hundred people here," he said in 2013, "only ten ever worked in finance." Product managers were called editors. Copies of a book on wabi-sabi, the Japanese aesthetic he had discovered as a teenager reading about tea ceremonies, sat on a lending shelf in the office. He liked to explain the idea with an image: "The monks rake up leaves, then they sprinkle a few leaves back."
By 2013 Square was processing fifteen billion dollars a year, not counting Starbucks, which was both its biggest customer and an investor, and the company had been valued at $3.25 billion. Dorsey owned about a third of it. Its newest product, SquareCash, let people send money to one another by email. It would become Cash App.
"Twitter is about moving words," Dorsey said. "Square is about moving money."
The Ginkgo Leaf and the Prada Suit
In the summer of 2013 D. T. Max rode the bus with him.
Dorsey took the No. 1 California line, and he liked to catch the 7:06. A ride cost two dollars, he explained, but with the monthly pass the real cost was closer to a dollar seventy-five. "If you buy it in bulk, it saves you a little bit of money." He loved the bus. "It's consistent, and it runs every few minutes. But it's also express. If I took another bus, it'd be stopping." He sat in the back and did silent market research: who was on a phone, who on a tablet, who just staring into space. "I saw the rise of Instagram here," he said. "I saw the rise of Vine and Snapchat." He had no desk at work and carried no briefcase. Over several months, Max never saw him handle a piece of paper. When people gave him books he gave them away, downloaded the ebook, and usually deleted it when he was done.
He kept a weekly restaurant schedule. "Zuni on Tuesday, Aziza on Monday, 54 Mint on Wednesday. Thursday, Tiburon. Friday, I don't go anywhere." There was a theory. "There's an approach of going really deep or going really broad. I'm always looking for the one thing for me, rather than tasting everything I can. If you have a table in a consistent space, you can actually notice more things within it."
It is the police scanner again, turned around: if the watcher stays put, the world reveals itself in passing. The same table, the same bus, the same three lines from an ambulance driver.
He had become, by then, a billionaire. Forbes ranked him America's sixth-youngest. Seven years earlier he had been in debt, and the change sat uneasily on him. He had just bought a modern house with a view of the Golden Gate Bridge in Sea Cliff for ten million dollars and marked the occasion with a tweet quoting Pablo Neruda: "I need the sea because it teaches me." He said he'd been perfectly happy in his old studio. "One of my ex-girlfriends said to me, 'Money is round. It's made to roll.' I definitely took that in," he said, and then, quickly: "It really weighs on me. I definitely feel the most fundamental issue is economic equality."
Biz Stone had known him from the rat-infested days, and he had the sharpest line about him. Stone was another dropout from Boston who'd started out moving boxes at Little, Brown, then designed book covers there; he walked away from Google stock to join a struggling podcast company and became the house master of the dopey tweet. ("Wow, Amtrak serves a vegan burger! Nice.")
— Biz Stone, on Jack Dorsey, 2013He still has this way about him that is, like, "The ginkgo leaf is the perfect leaf. That's why it's the only leaf I draw." And now he's able to say, like, "The BMW is the only car I drive, because it's the best automotive engineering on the planet," or whatever. It was a ginkgo leaf before, and now it's a Prada suit.
Stone, like Dorsey, talked in aphorisms, and he had one for this too: "Money makes people just more of what they were before."
What Dorsey had always been was a man who preferred to watch. "I'm perfectly happy being by myself, walking and thinking, and going to movies by myself," he said. "A lot of people are not comfortable doing things completely alone, but I am." Max watched him run a standing design meeting at Square, feet planted, body still, looking away from people with blue eyes that might otherwise have been inviting. At first he seemed satisfied with the work. Then he spoke, so softly that Max had to lean in: "Apple is moving in this direction. . . . We can't be a version of what they're doing, and right now we are." And: "Get away from white as much as you can." The designers objected, then folded. Robert Andersen, a Square creative director (and, as it happens, the same Robert Andersen who had invented the @-reply on Twitter in 2006), put it simply: "If he says something at a meeting, you better listen—he would not have spoken if he didn't have something important to say."
He had learned, in other words, to make his silence into a kind of authority. When he had dinners with Mark Zuckerberg, Zuckerberg cooked, and they circled each other warily. "He'd ask me if I had any ideas. Am I seeing anything in video? And, immediately, I'd turn back the question: 'What are you guys seeing?'" Zuckerberg, for his part, praised him: "Jack has been clear and disciplined." Williams, who'd fired him, said he wished he'd invested in Square. "What I have seen in the last few years as a Jack observer is his intense focus on his work and on himself to become better. It's amazing."
A friend described the transformation another way: "Jack doing A/B tests on his personality." Stone teased him about the friends he had begun to mention in passing ("Oh, my friend is, like, whatever, the guy who owns Time Warner"), then conceded the transformation was real and rare: a "geeky programmer kid" learning to "embody his own persona in a graceful and leadership-y kind of way." On December 24, 2013, Disney named him to its board.
The persona had a public channel, and the channel had critics. Dorsey posted so many self-shot Vines of himself staring into the camera that Valleywag wrote, "Is staring blankly into a camera every day the best way to get people to use your app?" (Vine, the six-second video service, was a company Twitter had bought in late 2012 at Dorsey's urging. Costolo remembered his pitch: "It's the perfect way of thinking about constrained media as a complement to constrained text.") Dorsey said he was humanizing the pedestal: "Some of those small details of life—here's a beautiful sunset—make everything feel a lot smaller, and more human." Max's verdict was that the effect ran the other way, that the grainy Vines of Dorsey and his girlfriend in Big Sur and Marfa made him seem larger, "and maybe even a bit less human."
He wanted, at the time, to be mayor of New York. He admired Bloomberg's mastery of the city's systems; he spoke of the city as an engineering problem whose interface Bloomberg had improved. At Twitter's seventh-anniversary party he asked Bloomberg for advice on succeeding him. Bloomberg's answer, by Stone's account: "Become a billionaire!" When Max pointed out the obvious obstacles (no public experience, a life in California, a discomfort with human contact so evident that the Los Angeles Times had described him moderating a Twitter town hall with President Obama as a "stiff, sweaty, and serious emcee"), Dorsey answered in the only key he had. Being mayor would come with "a lot of constraints," he said, "but I do well with constraints."
Near the end of their time together he told Max that Square had achieved only "one per cent" of its potential, then paused. "I'm at a point in my life when I want to go deeper," he said. "I could just move down to Marfa and become an artist."
He did not move to Marfa.
Two Desks
On June 11, 2015, Dick Costolo announced he was stepping down as Twitter's CEO, and Dorsey took over as interim chief. On October 5 the job became permanent. On October 14, Square filed to go public; Dorsey owned 24.4 percent. He was now CEO of two public-facing companies at once, and the first question every investor asked was the one the Twitter staff had asked in 2008: was he showing up?
Square priced its IPO at nine dollars a share, valuing it at $2.9 billion, less than half its last private valuation. It began trading on the New York Stock Exchange on November 19, 2015, his thirty-ninth birthday.
At Twitter, the numbers that mattered were going the wrong way. Daily tweets had peaked at 661 million in August 2014 and fallen to about 300 million by January 2016. In early 2016, the company replaced the reverse-chronological feed, the purest form of the scanner, with an algorithmic one that ranked tweets by predicted importance. In May he announced that photos and links would no longer count against the 140-character limit. That October Twitter cut 350 jobs, about nine percent of its workforce, and announced it was shutting down Vine. Suitors including Salesforce and Disney had looked at buying the company and walked away, partly over its reputation for abuse and harassment. That December Fox Business named him the fourth-worst CEO of the year, citing stagnant growth, falling stock and a part-time commitment; in 2017 24/7 Wall St. put him among the twenty worst CEOs in America. In November 2016 he was briefly suspended from his own account, which then had 3.9 million followers. He called it an "internal mistake."
What had changed was not the product so much as the traffic on it. The radio that had helped Iranians coordinate had become the main channel of an American president. In 2017 the character limit doubled to 280. In September 2018 Dorsey testified before the Senate Intelligence Committee, alongside Sheryl Sandberg, about Russian interference in the 2016 election; Twitter's shares fell six percent that day. On April 23, 2019, he sat with Donald Trump in the Oval Office while Trump complained that Twitter was suppressing his follower count; Dorsey tweeted afterward that they'd discussed making Twitter "healthier and more civil." A week earlier at TED he had said harassment was his biggest worry. That August a group calling itself the Chuckling Squad hijacked his personal account for nearly an hour and posted racist tweets.
On October 23, 2019, Twitter's stock fell nearly 21 percent, to $30.75, after an earnings miss, and a week later he announced a ban on all political advertising, effective November 22. In late February 2020, the activist hedge fund Elliott Management, led by Paul Singer, moved to oust him; they settled within days, and he stayed on. Square, meanwhile, moved fifty million dollars of its treasury into bitcoin in October 2020 and another $170 million in February 2021. Dorsey said publicly that if he were not running Twitter and Square, he'd be working on bitcoin. "I don't think there is anything more important in my lifetime to work on," he said in June 2021.
Then came January 2021. After the Capitol riot on January 6, Twitter locked Trump's account for twelve hours; on January 8 it suspended him permanently. Dorsey defended the decision and in the same breath disowned it, writing on January 14 that the ban "sets a precedent I feel is dangerous." Angela Merkel's spokesman said much the same thing.
He would later name the mistake precisely. Writing in 2022, after he had left, he said:
— Jack Dorsey, 2022The biggest mistake I made was continuing to invest in building tools for us to manage the public conversation, versus building tools for the people using Twitter to easily manage it for themselves.
It is the Y Combinator commenter's line from 2007, reached the hard way. Twitter had been a discovery, something its users built on top of, adding the @, the hashtag, the retweet. Somewhere along the line it had turned into a thing to be administered from headquarters. He had, it should be said, helped drive that shift. In 2013 he had defended revoking independent developers' access to the platform because, he said, "When people go to the App Store and type in 'Twitter,' they don't expect a million different apps. They expect to find one—from the company." The dispatcher had taken over the scanner.
On November 29, 2021, he resigned as Twitter's CEO, handing the job to his chief technology officer, Parag Agrawal. Two days later, on December 1, Square announced it would rename itself Block, a nod to the blockchain and to a company that had outgrown its card reader. The name took effect on December 10. H&R Block sued six days later. (They settled in 2023.) His title at Block became "Block Head." In May 2022 he left Twitter's board entirely.
The Singular Solution
Elon Musk began buying Twitter stock on January 31, 2022. By April 4 he owned 9.2 percent. In the days before that disclosure Musk had been talking with Dorsey about the future of social media, and Dorsey had texted him that he hoped Twitter could become open-source, and that he'd pushed, unsuccessfully, to bring Musk onto Twitter's board a year earlier, a failure he linked to his own decision to leave as CEO. "I trust you," he wrote. Musk asked Dorsey to stay on the board himself. Dorsey declined.
On April 25, Twitter's board unanimously accepted Musk's $44 billion offer, $54.20 a share. Dorsey's stake stood to be worth $978 million. He blessed the deal publicly in terms that sounded, for him, almost evangelical: "taking [Twitter] back from Wall Street is the correct first step," he said, and "Elon is the singular solution I trust." In August, in the litigation that followed Musk's attempt to back out, Musk subpoenaed him. When the deal closed on October 27, 2022, Musk fired Agrawal, Dorsey's chosen successor, along with the chief financial officer, the chief legal officer, and the general counsel, and had them walked out of the building by security. It was six years to the day after Twitter announced it was killing Vine. Dorsey did not sell. He rolled his roughly 2.4 percent stake into the new private company, saving Musk about a billion dollars in cash.
What followed is well documented. Within a year the company went from about 7,500 employees to about 1,500. Musk himself valued it at nineteen billion dollars on the first anniversary of the purchase. In April 2023 a court filing disclosed that Twitter, Inc., no longer existed. In July the bird was retired and the service became X. In March 2025 Musk's AI company, xAI, acquired X in an all-stock deal; in February 2026 SpaceX acquired xAI. In December 2022, as Musk's allies published internal documents they called the Twitter Files, Dorsey called the attacks on his former colleagues dangerous. In May 2024 he called X "freedom technology."
His heart, by then, was elsewhere, on a project that looked a lot like the one he'd failed to build in 2000. In February 2022 he had joined the board of Bluesky, a decentralized protocol Twitter had funded under his tenure, where users could choose their own algorithms. He left in May 2024, saying the project had drifted toward a conventional corporate structure and centralized moderation tools, exactly what he'd come to regret at Twitter. On December 16, 2022, he tweeted, "14 BTC deployed to @fiatjaf for #nostr," funding a social protocol with no company at its center at all. In 2023 he seeded OCEAN, a bitcoin mining pool built so miners could construct their own block templates instead of depending on pool operators. In July 2025 he released Bitchat, a messaging app that runs over Bluetooth mesh networks, phone to phone, with no internet connection, which is to say a citizens'-band radio. In November 2025 he backed diVine, a reboot of Vine built on its old video archive.
His politics followed the same centrifugal path. He gave money to Tulsi Gabbard and Andrew Yang in 2019; $15 million to mayors piloting guaranteed income in 2020; ten million dollars to Ibram X. Kendi's antiracism center at Boston University that August; an endorsement to Robert F. Kennedy Jr. in 2023; ten million dollars to the Freedom of the Press Foundation in January 2024, the largest gift in its history; and in October 2025, an endorsement to Thomas Massie, a Kentucky Republican, for the 2028 presidential nomination. The thread connecting these is not partisan. It is something closer to a deep, consistent distrust of the center of any network, whoever is sitting there.
On April 7, 2020, as the pandemic took hold, he tweeted that he was "moving $1B of my Square equity (~28% of my wealth) to #startsmall LLC to fund global COVID-19 relief," and added that the flows would be tracked publicly in a spreadsheet. He had already funded about six hundred Missouri public-school classroom projects on DonorsChoose in a single day in March 2016. In late 2017 he sat through a ten-day Vipassana retreat, no talking, no reading, no contact. In November 2018 he spent his birthday on another one in Myanmar.
Block, meanwhile, became large. It bought a majority of Jay-Z's Tidal for $297 million in 2021 and the Australian buy-now-pay-later lender Afterpay for $29 billion in stock in January 2022. On March 23, 2023, the short seller Hindenburg Research accused Cash App of inflating user counts and tolerating fraud; the stock fell 15 percent that day, and Block said it would explore legal action. It launched Bitkey, a self-custody bitcoin wallet, in December 2023 and began designing its own three-nanometer mining chips. It scaled back Tidal, wound down a decentralized-web unit called TBD, and changed its ticker in January 2025 from SQ to XYZ. It joined the S&P 500 on July 23, 2025. It has no designated headquarters. In 2024 Square was the U.S. market leader in point-of-sale systems, with four million sellers, and Cash App had 57 million users, many of them lower-income Americans who preferred it to the alternatives. Revenue in 2025 was $24.2 billion. On February 26, 2026, Block announced it would lay off more than four thousand of its roughly ten thousand employees, citing artificial intelligence as a primary reason. The stock rose 24 percent.
Forbes, that month, estimated his net worth at $5.9 billion.
It would be easy to read all this as a man running away from what he built, from the platform that slipped his grasp, from the corporate machinery of Block, from the center. It is just as easy to read it as a man still working on the same problem he had at eight: how to see the city, and how to put the people in. He has tried a central dispatcher and found that the dispatcher becomes the problem. He has tried handing the radio to a single owner he trusted. Now he is building radios with no tower at all.
His father never stopped going down to the river. Some time after his son invented Twitter, Tim Dorsey began tweeting the news from the Mississippi. When Vine came along, he posted the boats. Six seconds at a time, from the bank of the river in St. Louis, the old man was broadcasting the barges his son had found unbearable: where one was, where it was going, what it was doing.
Part IIThe Playbook
What follows is drawn from the record above, not from Dorsey's speeches, which tend toward aphorism, but from his decisions, which are messier and more instructive. He is an unusual case study: a founder who has been fired by his own company, returned to run it, run a second public company at the same time, been named one of the worst CEOs in America, sold to his own chosen buyer, and repudiated much of what that company became. The lessons are not uniformly flattering. Some of them are warnings.
The principles
- 1.Incubate one obsession for decades.
- 2.Find the abstraction under the industry.
- 3.Treat constraint as a product specification.
- 4.Use the fallow years as a residency.
- 5.Ship the primitive; let users finish it.
- 6.Build in the trough.
- 7.Pitch against yourself first.
- 8.Hire from outside the industry you intend to replace.
- 9.Make silence a management instrument.
- 10.Presence is the job; the polymath tax is real.
- 11.Rename the company when the thesis changes.
- 12.Distrust the center, including yourself.
⌁
The Dispatch Thread
One idea, re-expressed for four decades.
1984
At eight, gets an IBM PCjr; soon writes software to plot police-scanner vehicle movements across St. Louis.
1997
Hacks Dispatch Management Services, is hired by Greg Kidd, moves to New York to write courier-bidding software.
2000
Sketches "in bed" / "going to park" and builds STAT.US on a RIM 850. "No one cared."
2006
Sends "just setting up my twttr" on March 21; Twitter launches publicly July 15.
2008
Removed as Twitter CEO on October 16; becomes ceremonial chairman.
2009
Square founded February 14 after Jim McKelvey loses a $2,000 glass sale.
2015
Returns as Twitter CEO; Square IPOs on his 39th birthday.
2021
Resigns from Twitter November 29; Square renamed Block days later.
2022
Endorses Musk's $44B buyout; funds Nostr with 14 BTC.
2025
Ships Bitchat, messaging over Bluetooth mesh with no internet required.
Principle 1
Incubate one obsession for decades
Twitter is usually told as a two-week story: a playground slide, a hackathon, a prototype. The real gestation was twenty-two years. The police-scanner map in 1984, the courier system in 1997, the web dispatch startup in 1999, STAT.US in 2000, the vanishing-note and find-people-searching-Google experiments in the shed: each was the same question (where are the people, and what are they doing?) put to whatever technology happened to be at hand. When SMS arrived, Dorsey didn't have to think of the idea. He had to recognize the delivery mechanism.
The useful distinction is between an obsession and a project. Projects fail and are abandoned; STAT.US was a total failure. An obsession survives the failure of every individual project because it lives one level of abstraction up. Founders who cycle through unrelated ideas are effectively starting from zero each time. Dorsey started each one with a decade of accumulated intuition about the same problem, which is why, at the decisive moment, he could answer Sagolla's objection instantly: "What if you're in a loud night club?"
The risk, of course, is that an obsession can blind you to what your product has become. Dorsey said he didn't really grasp Twitter as a news medium until the Hudson landing in 2009, three years after launch.
Tactic: Write down the single question your last three projects were secretly trying to answer, and evaluate your next idea by whether it answers that question better.
Principle 2
Find the abstraction under the industry
Dorsey's most valuable intellectual move is to strip an industry down to its underlying verb. Couriers, taxis, and ambulances were all, to him, "reporting constantly where they are and what work they're doing," and he said the "abstraction was so cool that I wanted that same thing for my friends." Payments weren't money; they were "commerce, which is conversation," and, in the line Kidd remembered, "very intimate." Once he reframed payments as a conversation between two people, the clutter of the existing system (registers, paper receipts, tiered fees) read as noise on the line.
The abstraction also tells you whom to serve. If payment is conversation, the underserved party is the small merchant who couldn't afford to join the conversation at all: McKelvey losing a $2,000 sale, the farm stand, the coffee shop that reminded Dorsey of his mother's café. The positioning ("countertops and farm stands") came straight out of the reframe.
This is different from the usual "first principles" exhortation. Dorsey doesn't decompose industries into physics. He decomposes them into human signals, and then asks what the cheapest, most universal channel for that signal would be.
Tactic: Describe your market in a single verb that would make sense to someone outside it, then ask who is currently priced out of performing that verb.
Principle 3
Treat constraint as a product specification
"Constraint inspires creativity" is the credo Dorsey repeats most, and it is easy to dismiss as aesthetic posturing from a man with a fixed weekly restaurant schedule. The business record supports it. The 140-character limit was not chosen; it was inherited from SMS's 160-character cap minus room for a username, and the constraint became the identity of the medium. Square's single 2.75 percent rate was a constraint on the company's own pricing flexibility, and it became a marketing message to merchants drowning in Visa's rate tables. Vine's six seconds was pitched to Costolo explicitly as "constrained media as a complement to constrained text."
The point is that a hard constraint does work that a feature list can't. It tells users what the product is for, makes the product legible in a sentence, and forces the people building it to choose. When Twitter began relaxing its constraints (excluding links and photos from the count in 2016, doubling to 280 characters in 2017) it was also, not coincidentally, a period of stalled growth and identity confusion.
▢
Two Hits, One Grammar
The same design instincts applied to two very different industries.
| Dimension | Twitter (2006) | Square (2009) |
|---|---|---|
| Underlying verb | Reporting status | Conversation via commerce |
| Borrowed infrastructure | SMS, on "every single device" | The iPhone's headphone jack, royalty-free |
| Defining constraint | 140 characters | One flat 2.75% fee |
| Born in | Odeo's collapse | The 2008–09 financial crisis |
| Original user | Dorsey himself | A glassblower who lost a sale |
| Dorsey's role | CEO, then ousted | CEO from day one, "his do-over" |
Tactic: Pick one constraint your product will never relax, and make it the first sentence of your pitch.
Principle 4
Use the fallow years as a residency
At twenty-six Dorsey moved home to St. Louis and said he felt like a failure. Then he studied botanical illustration, drew a banyan tree at the botanical garden, trained toward massage therapy, babysat in exchange for a shed, took figure drawing because a denim designer he admired valued it, and designed two skirts he threw away. A conventional résumé reads this as three lost years. The record suggests otherwise. The aesthetic sensibility that made Square's reader a MoMA object, that named product managers "editors," that told designers to "get away from white," was assembled in those years, not at Odeo.
The underlying discipline is his rule: "When something interests me, I don't just read about it. I do it." Doing, rather than reading, means the fallow period produces skills and artifacts rather than opinions. Notice too that he kept a low-stakes coding practice alive at night ("the purity of making cool little things") so that when an opening appeared at Odeo, his hands were warm.
There is a selection effect here, and it would be dishonest to ignore it. Kidd's shed and his father's software work were a safety net many people don't have. The transferable part isn't the luxury of drifting. It's the habit of filling a gap with deliberate, hands-on apprenticeships in adjacent crafts.
Tactic: During any forced pause, enroll in one craft adjacent to your field and finish one small working thing a week in your own.
Principle 5
Ship the primitive; let users finish it
None of Twitter's signature conventions were invented by its founders. The @-reply came from Robert Andersen in November 2006, the hashtag from Chris Messina in August 2007, the word "tweet" from outside developers. Dorsey shipped something austere enough, "R.S.S. for S.M.S.," that users could build a grammar on top of it. That austerity is why the product absorbed revolutions, celebrity fandom, and a ferry passenger's photo of a plane in the river.
The cautionary half of this principle is Dorsey's own. By his account, his "biggest mistake" was investing in tools for the company "to manage the public conversation, versus building tools for the people using Twitter to easily manage it for themselves." He had also, in 2013, defended pushing out third-party developers because users "expect to find one [app]—from the company." Those two choices — closing the ecosystem, then centralizing the moderation burden — turned a discovery into an administration. Every Dorsey project since, from Bluesky to Nostr to Bitchat, has been an attempt to undo that drift.
Tactic: Before building any feature, check whether your heaviest users have already improvised it, and formalize their version instead of designing your own.
Principle 6
Build in the trough
"We created Twitter in a depression, or funk, in the Odeo office, and Square in the heat of a financial crisis." Both claims are accurate. Twitter came out of a hackathon called because Apple had just made Odeo's business irrelevant. Square was incorporated in February 2009, months after the financial crisis and his own firing, when, he says, venture capitalists weren't eager.
Downturns help founders in practical ways. Talent is idle, competitors retreat, and incumbents (the card networks, in Square's case) are distracted. A trough also gives cover for an unconventional bet, because no one expects the conventional ones to work. Dorsey adds a psychological claim, that he "thrives on tension," which is harder to generalize. The structural point holds regardless: his two biggest companies started at moments when the expected value of the status quo had collapsed.
Tactic: When your current project is visibly failing, run a time-boxed internal hackathon before you run a layoff.
Principle 7
Pitch against yourself first
In 2009 Dorsey was a fired CEO pitching a payments company to skeptical investors in a recession. His answer was a slide titled "140 Reasons Square Will Fail." It is a small move with a lot packed into it. The number is his own brand, which acknowledges the Twitter history instead of hiding it. The premise steals the objections before the room can raise them. And it signals that the founder has thought about the risks harder than anyone across the table.
He applied the same technique to himself in public. Asked whether he'd plotted Williams's downfall, he didn't deny the feeling: "Was I thinking, Screw Ev? . . . I don't know. Maybe." That candor is disarming in exactly the way the slide is. It's also a reminder that Dorsey has always been careful with his own story. He "sedulously worked his way back" into Twitter's founding myth, in Max's phrase, partly by resurfacing the "in bed" / "going to park" notepad on Twitter's fifth anniversary.
Tactic: Put your strongest objections into your own deck, numbered and titled, before you walk into the room.
Principle 8
Hire from outside the industry you intend to replace
"Of the six hundred people here, only ten ever worked in finance." Square was a payments company staffed almost entirely by people who did not know how payments were "supposed" to work. That is how a team decided to route card data through a headphone jack, sidestepping Apple's dock connector, and how it priced at a single flat rate in an industry built on tiered complexity.
The approach has a cost, and the record shows it. Industries develop their habits partly to manage risk, and newcomers learn those lessons the expensive way. Square's withholding of merchant funds in 2020, Hindenburg's 2023 short report on Cash App fraud, and the 2024 settlement negotiations over anti-money-laundering compliance are, at least in part, the bill for inexperience in a heavily regulated business. The right lesson isn't "never hire insiders." It's to keep insiders in the guardrail functions, compliance, risk and treasury, and keep them out of product.
Tactic: Staff product and design from outside your industry, and staff compliance and risk from deep inside it.
Principle 9
Make silence a management instrument
Most founders lead by talking. Dorsey leads by not talking, and then talking once. "He would not have spoken if he didn't have something important to say," said Robert Andersen. In the design meeting Max witnessed, Dorsey stood still, let the demos play, and then delivered one quiet sentence ("We can't be a version of what they're doing, and right now we are") that redirected the work. At dinners with Zuckerberg he turned every question back: "What are you guys seeing?"
Silence works as a tool because it rations attention. Every remark carries more weight, and people who would otherwise wait for direction have room to fill the gap. It also gathers information. Dorsey's lifelong mode, from the police scanner to the back of the No. 1 bus, is the listener's. Its failure mode showed up in his first stint at Twitter, where a quiet, often absent CEO read to the staff as a disengaged one. Silence only works when people can see you're present.
Tactic: In your next three decision meetings, speak only once, at the end, and make that sentence the decision.
Principle 10
Presence is the job; the polymath tax is real
Dorsey's 2008 removal is often told as boardroom politics. The underlying cause was simpler: the people working for him could see he wasn't fully there. His defense, that he went back to work at night after yoga, misses the point. A startup's staff takes its cues from what it can observe, and what thirty people observed was a CEO leaving early.
He paid the tax again as a two-company CEO from 2015 to 2021. Fox Business's 2016 ranking cited his "part-time commitment," Elliott Management moved against him in 2020 citing the same concern, and the New Yorker in 2013 had already reported that splitting time between Twitter and Square had exhausted him and "alienated some Twitter employees." The interesting contrast is Kidd's description of him as a babysitter, "very present," which the children could feel. Presence was never something he lacked. What he lacked was a way of making it visible to adults.
◐
The Portfolio, Tagged
Dorsey's ventures sorted by what they became.
| Venture | Year | Outcome | Stage |
|---|---|---|---|
| STAT.US | 2000 | "No one cared" | Scan |
| 2006 | Sold for $44B in 2022 | Scale | |
| Square / Cash App | 2009 / 2013 | $24.2B 2025 revenue | Scale |
| Vine | 2012 | Shut down 2016; rebooted as diVine 2025 | Risk |
| Tidal | 2021 | Scaled back 2024 | Risk |
| Afterpay | 2022 | $29B stock acquisition | Scale |
| Bluesky | 2022 | Left board 2024 | Risk |
| Nostr / OCEAN | 2022–23 | Funded; protocol-level bets | Try |
| Bitkey / Proto | 2023–24 | Shipping hardware | Try |
| Bitchat | 2025 | Live on App Store | Scan |
Tactic: If you have interests outside the company, schedule them where your team can't see them, and make your hours inside it unmistakable.
Principle 11
Rename the company when the thesis changes
Square's name came from its first device, and the device stopped being the company long before 2021. By then there was Cash App, a bank charter approved in 2020, bitcoin on the balance sheet, Tidal, and Afterpay on the way. On December 1, 2021, two days after leaving Twitter, Dorsey renamed the company Block, then changed the ticker from SQ to XYZ in 2025. Even the CEO's title changed to "Block Head." The name told employees, investors, and regulators that the device was now one product among several and that the organizing idea had moved to the protocol layer.
Renaming is expensive: H&R Block sued within a week. But a name that describes the past makes every new initiative look like a distraction, while one that describes the thesis makes it look like the strategy. Dorsey had already lived through this once at Twitter, whose homepage question changed from "What are you doing?" to "What's happening?" in 2009 because the old one no longer described how people used the service.
Tactic: If your company's name describes your first product rather than your current thesis, put a rename on the board agenda this year.
Principle 12
Distrust the center, including yourself
The thread running through Dorsey's last decade is a growing suspicion of any single point of control: the Twitter moderation apparatus he built and regretted, Wall Street ownership he wanted "taken back," Bluesky's turn toward "a traditional corporate structure," mining pools that dictate block templates, even the internet itself, which Bitchat routes around. His Trump decision is the cleanest example: he defended the ban and called the precedent "dangerous" in the same statement, recognizing that a power he had used correctly was still a power no one should hold.
The irony is that his most consequential act of decentralization ended in maximal centralization: he helped hand his company to a single owner who fired his successor on day one and dissolved the board. "Elon is the singular solution I trust" is a sentence about trust in a person, which is exactly the dependency his later projects are designed to remove. The lesson cuts both ways. Building systems that don't depend on anyone's good judgment, including your own, is sound, but you can't get there by trusting the right person.
Tactic: For every critical decision right your company holds, write down who could abuse it after you leave, and design a version that doesn't require trusting them.
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
— Jack Dorsey, on the police scanner of his childhood, to The New Yorker, 2013They were reporting constantly, and they're reporting three things usually. No. 1, where they are. No. 2, where they're going. And, No. 3, what they're doing.
— Jack Dorsey, on the origins of Twitter, to The New Yorker, 2013I wasn't considering what everyone else wanted. I was considering what I wanted.
— Jack Dorsey, to The New Yorker, 2013There's an approach of going really deep or going really broad. I'm always looking for the one thing for me, rather than tasting everything I can. If you have a table in a consistent space, you can actually notice more things within it.
— Jack Dorsey, on building Square, to The New Yorker, 2013I thrive on tension. If I had a relationship where nothing was ever wrong and we were never debating or arguing, I don't think we would ever grow. There's no change.
— Jack Dorsey, to The New Yorker, 2013When something interests me, I don't just read about it. I do it. I go all in.
Maxims
- Keep the question, change the radio. The same problem, put to SMS, then a headphone jack, then Bluetooth mesh, yields a different company each time.
- Name the verb, not the industry. Payments are conversation; dispatch is status; once you name the verb, you can see who's been left out of it.
- A constraint is a promise. 140 characters and 2.75 percent told users what the product was before any marketing did.
- Failure is a residency if you make things during it. Botanical drawings and a shed full of code were the apprenticeship for the second act.
- Ship something bare enough to be finished by strangers. The @, the hashtag, and the retweet came from users, and so did the value.
- Don't let the dispatcher take over the scanner. Tools for the company to manage the conversation became Dorsey's self-described biggest mistake.
- Write your own failure slide. "140 Reasons Square Will Fail" took the objections off the table before investors could raise them.
- Speak once. In a quiet leader's meeting, one sentence redirects more than an hour of opinions.
- Absence is a message. A CEO's yoga class is invisible to him and very visible to thirty employees.
- Trusting a person is still a single point of failure. Build systems that don't depend on anyone's judgment, including your own.
