Part IThe Story
In the spring of 2010, on his walk home through Palo Alto, a twenty-seven-year-old former Google customer-support employee named Ben Silbermann would stop at the Apple Store, walk along the tables of display machines, and change the open browser window on each one to a website that almost nobody had heard of. Then he would retreat to the back of the store and wait. When shoppers drifted over to the screens, he performed a small piece of theater for them, the kind a shy person rehearses in his head before saying aloud. Wow, he would say, to no one in particular. This Pinterest thing. It's really blowing up.
It was not. Four months after launch the site had roughly 3,000 registered accounts, and by one count only about 200 people actually using it, half of them his friends from Des Moines. His friends and family, emailed with the hopeful announcement every founder sends, had answered with the special politeness Midwesterners reserve for a nephew's garage band: Oh, looks interesting. Very interesting. Investors had passed. A New York magazine publisher, offered the chance to simply buy the thing, declined even to take the meeting. The company's only physical address was a two-bedroom apartment, one bedroom of which was rented to a person who did not work there.
Silbermann has told the Apple Store story several times since, always for a laugh, always filed under "pretty desperate things, honestly." It is funny. It is also, if you sit with it, a nearly perfect miniature of the man and the company. He was not hacking a growth loop. He was not gaming a ranking. He was standing at the back of a room, unremarkable in appearance, quietly arranging a display and hoping a stranger would find it beautiful. The whole of Pinterest is in that gesture — the curation, the patience, the faith in ordinary people's taste, the reluctance to raise his voice, and, underneath, a fierce and slightly embarrassed wish to be noticed by people who didn't know him.
Asked once what his "big plan" was, he gave an answer that sounds, in retrospect, almost absurdly small for a man who would one day control a public company worth more than $12 billion.
— Ben Silbermann, Y Combinator Startup School, 2012I just want to go somewhere and see somebody that I don't know using something that I made and have it be kind of useful.
He got his wish. Hundreds of millions of times over, he got it. What follows is the story of what that wish cost, what it built, and what it never quite managed to sell.
By the Numbers
The Collector's Ledger
3,000Registered accounts four months after Pinterest launched
50Fully functional layout versions Evan Sharp built before the grid was settled
11.7MU.S. unique visitors in January 2012 — the fastest site ever to cross 10 million
$1.5BValuation in Rakuten's May 2012 round, roughly two years after launch
$24.40First-day close after the April 18, 2019 IPO, priced at $19
433MMonthly active users when Silbermann stepped down as CEO in June 2022
$5 vs. $48.29Q1 2022 revenue per user in the U.S. and Canada: Pinterest vs. Facebook
A Specimen Drawer in Des Moines
Before there were boards there were drawers, and in the drawers there were bugs.
Ben Silbermann grew up in the suburbs of Des Moines in a family that had, by the time he was old enough to notice, essentially settled the question of what he would become. His parents, Jane Wang and Neil Silbermann, were ophthalmologists with a family practice in Urbandale. His sisters became doctors. His father, by his son's later and affectionate report, liked his VCR just fine and saw no particular reason to replace it. This was not a technology household. It was a household of diagnosis — of looking closely at small things and naming what was wrong with them — and the boy absorbed the habit even as he redirected it. He collected stamps. He collected leaves. He collected insects with a fervor he would later describe as "maniacal," pinning them in rows, labeling them, arranging them so that the arrangement itself said something.
The rest of his adolescence reads like the résumé of a son preparing, conscientiously, for the life laid out in front of him. He played cello in the Des Moines Youth Symphony. He split his days between Roosevelt High School and Central Academy, the district's program for gifted students. In the summer of 1998 he attended the Research Science Institute at MIT, a six-week program for high-school researchers founded by, of all people, Admiral Hyman Rickover. And he debated — Lincoln-Douglas, the one-on-one format built around values rather than evidence cards — under a coach named Dave Huston, who is now in the National Speech & Debate Association's Hall of Fame. In 1999, Silbermann took sixth place at nationals.
That last detail matters more than it seems. Lincoln-Douglas rewards a peculiar kind of persuasion: not volume, not aggression, but the patient construction of a framework in which your conclusion becomes the only reasonable one. It is the debate format for people who would rather be right than loud. Twenty-five years later, he and his wife would fund the NSDA's expansion efforts through their family foundation, and he would fly back to accept its Alumni Lifetime Achievement Award. Of all the honors available to a billionaire, he chose to keep showing up for the one that remembered him as a teenager in a borrowed suit, arguing about justice.
He arrived at Yale as pre-med. Somewhere around his junior year — the accounts vary between junior and senior — he concluded that he did not, in fact, want to be a doctor. He had been taking political science courses alongside the pre-med sequence; he finished with a degree in political science in the spring of 2003. He has also said, in passing, that what pulled him toward technology was something as mundane as access to high-speed internet in his college room. The family business of looking at things closely had found a different object.
The Story of His Time
The first job out of Yale was so far from Silicon Valley it might as well have been on the opposite coast, which it was. Silbermann went to Washington, D.C., to work in the technology group of a consulting firm, writing what he would later describe, without nostalgia, as corporate business-to-business disaster-recovery plans. It was Web 2.0's first flush. At some point he stumbled onto a blog called TechCrunch and thought it was a ridiculous name. He kept reading anyway.
What he felt reading it was not envy exactly but exclusion — the sensation of watching history happen through a window. "I felt like it was a story of my time that I wasn't a part of," he told Inc. years later. He found himself inspired by Kevin Rose of Digg, a founder whose public persona was then the very image of the tech dude Silbermann would never become. He drifted toward the internet's oddball discussion spaces, Reddit and the now-defunct Google Answers. And in the free hours he built small things with friends — "internet toys," he called them — including an online tool that let you try on eyeglasses virtually. The son of two eye doctors, building a tool to look at frames. The rhyme is too neat to have been planned.
In 2006 he moved to California and, in his phrase, "cajoled his way into Google." He was not an engineer. He was hired into online sales and operations — customer support, by several accounts, eventually working on display advertising products inside the ads group — and he stayed from December 2006 to November 2008. He has always spoken of Google with unusual warmth. He had never been inside a company that thought so big, where an idea was not dismissed as far-fetched but simply broken into its component problems. He found the place "very philosophical." He once offered, with the dry edge he rarely lets show, that people who complain about Google and Facebook have never worked anywhere else.
But a non-engineer at Google in 2007 was a man standing at a glass ceiling he could see straight through. His ideas, he concluded, would never be recognized there in the form he imagined them. He kept talking about starting something. He kept not starting it.
His wife ended the stalemate. Divya Bhaskaran had graduated from Clark University in 2004 with a degree in psychology, and her diagnosis of her husband's condition was brief and clinical. "She said, 'You should do it or shut up about it,'" Silbermann told an audience at Startup Grind in 2012, sounding, the reporter noted, slightly thunderstruck still. Across a decade of interviews he has credited two people above all others for keeping him going: his wife, and a mentor, Eventbrite's Kevin Hartz. The ultimatum is worth pausing on because it is the first of many moments in which Silbermann's story turns on someone else saying the blunt thing he would not say himself.
The partner was Paul Sciarra, a college friend who had been working at Radius Capital, a New York venture firm — a man who had sat on the money side of the table and decided he'd rather be across from it. In 2007, Sciarra asked Silbermann whether he wanted to explore making software for the newly released iPhone. In August 2008, both men quit. They called their company Cold Brew Labs.
The timing was exquisite in the worst sense. Within weeks, Lehman Brothers collapsed and the venture market froze into something close to permafrost.
Tote, or the Catalog Nobody Bought From
The first product was called Tote, and in hindsight it contains the entire argument of Pinterest — including, eerily, the part of the argument Pinterest would spend the next fourteen years losing.
Tote was a mobile shopping catalog for the iPhone, pitched as the first women's fashion catalog on the device: a virtual replacement for the glossy paper catalogs that arrive unbidden in American mailboxes. It would aggregate products from many retailers, personalize the browsing, and let you buy on the go. The trouble was the buying. Mobile payments in 2008 and 2009 were primitive; checking out on a phone was miserable, and so almost nobody did. Apple's approval process slowed every update to a crawl. The founders kept cramming in features to compensate. Investors weren't interested.
Silbermann later diagnosed the failure with a precision that became something like his creed: "Even though we had all these great features we were cramming in, we weren't great at one thing. There wasn't one thing that was special about it."
Getting even that far had required a hack. With the market frozen, Silbermann entered the NYU Stern Business Plan Competition, which he could enter despite not being an NYU student. The team placed second — not first — and the prize was a meeting with FirstMark Capital in New York. That meeting, he said, produced half their money. Brian Cohen, founder of New York Angels, met the pair at the competition and became, by his own proud accounting, Pinterest's very first investor. Cohen came away believing they were "a couple of young guys from NYU," which tells you something about how much the founders cared to correct strangers about their résumés. The original check, in Silbermann's telling, was about $200,000.
What Cohen remembered was not the pitch but the listening. "As the app was being used, he recognized that women were grabbing, tagging specific items that they were able to view later when they got home," Cohen said. People were not buying on Tote. They were saving. They were amassing collections of favorite things and showing them to friends. "He recognized that these women were sharing their tastes."
This is the hinge of the whole story, and it bears stating with care. Silbermann watched a commerce product fail at commerce and succeed at something adjacent to it — the assembly of desire, the public display of taste — and he chose, decisively, to follow the adjacent thing. He built a company on saving instead of buying. It was the right call. It was also the original sin that Wall Street would never quite forgive.
There were, by one account, other failures before this one; a 2014 profile counted five companies that died in quick succession before he found the one that lived, and Silbermann has mentioned a family medical-records tool he tried to build with PhD students, who dispersed the moment their degrees were done. "Most things worthwhile," he said, "take a while." He says this sort of thing often. It has the quality of a man reassuring himself in the third person.
Fifty Versions of a Grid
In November and December of 2009, Silbermann, Sciarra, and a third friend began building something new. The third friend was Evan Sharp, then an architecture student — a person trained to think about how bodies move through space and how the arrangement of walls tells you where to look. Sharp knew people on the East Coast; Silbermann knew people on the West. Neither group, it would turn out, understood what they'd made.
None of the three was really an engineer. Silbermann has been candid about this. "None of us were super technical so we would always just sort of chip in and learn what we needed to do to get to the next part," he told Y Combinator's Ali Rowghani in 2016. He picked up product design and simple front-end code in fragments, the way a collector learns taxonomy: by needing it for the specimen in his hand. He spent long hours at Hacker Dojo in Mountain View, a hothouse for programmers, where he was, in the room's own terms, a tourist. MIT Technology Review would later state it flatly: "he's no programmer." He directed people who were.
What he directed them toward was beauty, and he was almost pathologically specific about it. Sharp built fifty fully functional versions of the site's basic layout, varying spacing and image sizes by fractions of an inch — fifty working websites, not mockups, for a product with essentially no users to test them on. The founders ignored the gospel of shipping early and iterating in public. At South by Southwest in 2012, asked why, Silbermann gave an answer that is less a product philosophy than a moral one: "Collections have to be something that you're proud of. If the collections didn't look awesome, why would anyone spend time working on them?" And then, more pointedly: "You have to show that you put in as much effort as you expect [your users] to put in."
He refined the principle later into the advice he gave most often to young founders, at Startup School in 2012: "My advice is you should ship when you have one thing that you're proud of, like one thing that is worthy of someone's time." For Tote there had been no such thing. For Pinterest there was exactly one: it made your collection look cool. "If it didn't look cool, then no one is going to make a collection," he said, "because they don't want to show their friends because this thing they just made looks really lame."
The result was the grid — the cascading, uneven, ragged-bottomed columns of images that would be copied by half the internet within three years. It looked like nothing so much as a corkboard in a teenager's bedroom, or, if you squinted, a museum drawer of pinned specimens, each one sized to its own wings. "Much the way he'd pinned and displayed insects as a boy," the San Jose Mercury News put it in 2012, and Silbermann never objected to the comparison. Vanity Fair said the stamps and insects inspired "the aesthetic display." It is the rare origin myth that is also simply true.
There is a short children's biography of the founders — Rosa Waters's Pinterest: How Ben Silbermann & Evan Sharp Changed the Way We Share What We Love, part of a "Wizards of Technology" series aimed at young readers (also available as an e-book edition) — and it is telling that the story is most often told this way, as a fable for children. It has a fable's shape: a boy who loved small beautiful things, who grew up and made a place to keep them.
Flyover Country
The site launched in the first months of 2010 — accounts differ between January and March — as an invitation-only beta. And then, for most of a year, almost nothing happened.
"The thing that surprised me the most in starting a company," Silbermann told the Startup School audience in 2012, "after reading about, you know, Facebook hits Harvard, 95% penetration in two weeks, like Instagram shoots to a million people, is that it can take a really, really long time to build things that are worthwhile." He then read aloud an email he'd sent to his advisers and investors in March 2010, a few months after launch — a document of sober hope from a man who had not yet been rescued by anything.
The tech people didn't get it. That is the part of the story Silicon Valley tells about itself with a kind of self-flagellating relish. What is more interesting is who did. The small group who loved the site, Silbermann noticed, were "not who you think of as early adopters. They were folks that I grew up with, and they were using it for regular stuff in their life — what was my house going to look like? What kinds of food do I want to eat?" Half of the first 200 active users were his friends in Des Moines. The vision, as MIT Technology Review would later put it, gained momentum "not at the elite colleges and California coffee shops that often function as the Web's proving ground for new ideas but by word of mouth in Silbermann's home state of Iowa."
Here a different founder would have rebuilt the product for the people he wished were using it. Silbermann did something stranger and humbler. "Maybe," he thought, "I can just find more people like me." Where did these people congregate? Not on Hacker News. On design blogs and craft blogs and home blogs, many of them written by women. He went to a conference where those bloggers gathered — the Alt Summit, which he would later return to as a keynote speaker — and organized a marketing event in which each blogger introduced Pinterest to her audience. The company ran a campaign called "Pin It Forward" that chained bloggers together, each one inviting the next. There were meetups. "The thing about it that really worked," he said, "was we found this little group of people that were interested in the same thing, and we showed them how the service could be helpful to them."
And he took care of them with an intensity that bordered on the devotional. He put his personal cell phone number in every customer-support email. When the service went down — which, run by non-engineers out of an apartment, it did — his phone would light up: Hey, I can't get my pins. He wrote personally to the first 5,000 users, offering his number and, in some cases, meeting them. He told a London interviewer in 2014 that he had once "known all 500 users." He sat in coffee shops and asked strangers to try the site, and he ran a ritual so simple it sounds like a parlor game: "I would ask them to look at a button and push it. And right before, I'd say, What do you expect to see on the other side of that? And then right after, I'd be like, is that what you saw? And if it was different, I'd be like, We should fix that up." His explanation for the intensity was almost tender: "Since it took us so long to get those users, we cared about them so much."
Meanwhile the money men kept saying no. At TechCrunch Disrupt in New York in May 2010, the company had a booth — though, as Business Insider noted dryly, nowhere near the main stage, and only because FirstMark was a sponsor and had to pull strings. On May 31, 2010, a former IAC dealmaker named Shana Fisher called to say she loved the product and wanted to invest if they'd let her. Small checks trickled in from an angel roster that, read now, looks like a guest list for a future party: Jeremy Stoppelman of Yelp, Jack Abraham, Michael Birch, Scott Belsky, Kevin Hartz, Hank Vigil, Fritz Lanman, William Lohse, Brian Cohen. The team stretched every dollar to extend the runway. Silbermann would later say he came to view fundraising as a resource rather than a goal, and that he worried about founders "confusing money with purpose." He also learned, in his telling, three lessons from the long season of rejection, the sharpest of which was about leverage: if you need the money and they're the only one who can give it to you, you have none.
The other lesson had a knife's edge on it, softened only by his delivery. Venture capitalists, he told the Startup School crowd, have their IPO certificates on the wall. "There are things that they passed on, and those are the things that actually haunt them at night."
By March 2012 — after the curve had finally bent, after Silbermann had gone onstage at South by Southwest — the coastal establishment was looking at his user map with a kind of bafflement. Chris Dixon, the founder of Hunch, who interviewed him onstage in Austin, told The New York Times that Pinterest "defies the mold that you have to build something for New York and San Francisco and then spread it out from there." Then he added, with a candor investors almost never permit themselves: "Everyone's envious. I'm envious. I wish I was an investor. I wish I'd created the site."
Silbermann had already explained why, two years earlier, in a speech at NYU Stern that was more radical than anyone in the room seems to have noticed.
— Ben Silbermann, speech at NYU Stern, 2010If you walk around Brooklyn and ask people how they express themselves. Everyone's a musician or an artist or a filmmaker. But most of us aren't that interesting. Most of us are just consumers of that. And when we collect things and when we share those collections with people that's how we show who we are in the world.
The Observer, reprinting this, suggested he might want to "leave out the less interesting part going forward." It was the most honest thing anyone in tech had said about its audience in years, and it was wrapped, characteristically, in self-deprecation. Most of us. He included himself. He meant it.
The Hockey Stick and the Lawyer
In March 2011, Pinterest released an iPhone app that brought in more downloads than anyone expected. Then the line on the chart went nearly vertical, and the man who had been faking crowds in the Apple Store found himself with a real one.
The milestones arrived in a rush. In May 2011, a $10 million Series A led by Jeremy Levine and Sarah Tavel of Bessemer Venture Partners — Silbermann's first significant institutional backing, raised only after he could show consistent month-over-month growth. In August, Time named Pinterest one of the fifty best websites of the year. On October 7, after introductions from Kevin Hartz and Jeremy Stoppelman, Andreessen Horowitz led a $27 million round that valued the company at $200 million. In December, Hitwise ranked it among the ten largest social networks in the country, with 11 million visits a week. In January 2012, comScore counted 11.7 million unique U.S. visitors, making Pinterest the fastest site ever to cross ten million. It won the Crunchie for Best New Startup of 2011. By March it was, per Alexa, the sixteenth most-visited website in America — ahead of CNN, ahead of ESPN.
Silbermann and a handful of programmers had run the whole operation out of an apartment until mid-2011. When they began recruiting in earnest, the pitch was the adventure, not the certainty, and the closer was the Palo Alto apartment's patio, where they hosted barbecues for engineers who liked that the company hadn't "made it" yet. He hired, he said, "builders" with deep skills in odd corners — robotics, cartooning — and people he simply believed were great before he had any idea what job they'd do. In October 2011 he flew home for a startup conference in Des Moines called Thinc Iowa. "I'm excited to come home," he wrote to the organizer.
Fame, when it came, came in the form of the users themselves. Ann Romney joined in February 2012. Michelle Obama joined in June. Oprah, Yoko Ono, Ryan Seacrest, and, Vanity Fair noted archly, Mark Zuckerberg — "perhaps to keep an eye on a possible threat to his reign." The press came too, and Silbermann mostly turned it away. "We've been too busy to hire a press person yet," he told the Startup Grind audience in February 2012, almost apologetically. He would admit years later that when The New York Times reached out in the early days he'd brushed them off with an "I'm busy," and that this was not, in retrospect, the best strategy. He was an introvert who had been handed a megaphone and preferred to keep it in a drawer.
Growth also brought the first real crisis, and it arrived from the direction a collector should have anticipated: the question of whose things were being collected.
In February 2012, a photographer and lawyer in Atlanta named Kirsten Kowalski published a blog post explaining that, after reading Pinterest's terms closely, she had deleted all her pins — because, by her reading, nearly every pin was a copyright infringement, and the terms placed the liability on the user. The post went viral. Within weeks, The Wall Street Journal was asking whether Pinterest was "the next Napster." A Scientific American blogger, the illustrator Kalliopi Monoyios, quoted the terms of service word by terrifying word: users granted Cold Brew Labs a "worldwide, irrevocable, perpetual, non-exclusive, transferable, royalty-free license, with the right to sublicense, to use, copy, adapt, modify, distribute, license, sell" their content. Anything you pinned, she wrote, belonged to them. "Completely. Wholly. Forever and for always."
The response is a tidy case study in Silbermann's crisis temperament, which is to say: no counterattack, no grandstanding, quick quiet repair. He called Kowalski himself to talk about making the site more compliant. On February 20, Pinterest released a "nopin" HTML tag that let any website opt out of being pinned; Flickr implemented it four days later. In March the company published a statement asserting DMCA safe-harbor protection. On March 23 it rewrote the terms to eliminate the right to sell user content, effective April 6. "Selling content was never our intention," the company's blog said. By May 1, pins from Flickr, Behance, YouTube, and Vimeo carried automatic attribution, and Flickr — which had been the first to install the opt-out — had added a Pin button to its own share menu. At SXSW he said he spent a lot of his time reaching out to people who were worried. "We have a social responsibility to understand what's going on in the community," he told the room, "and feed it into every product decision."
On April 6, 2012 — the very day the new terms took effect — Paul Sciarra left the company to become an entrepreneur in residence at Andreessen Horowitz. The man who had left venture capital to build something went back to venture capital once it was built. Silbermann, who has spoken often about tending co-founder relationships "like any relationship," has said almost nothing publicly about this one.
On May 17, Rakuten, the Japanese e-commerce giant, led a $100 million round at a $1.5 billion valuation. In July the company — some thirty employees — moved from Palo Alto to San Francisco, a decision that made headlines no thirty-person office move should. On August 10, Pinterest dropped its invitation requirement and opened to everyone. In September, Silbermann hired an engineering chief from Amazon, Jon Jenkins, eight years in Seattle. In October the company launched business accounts. A business-plan-competition runner-up was now, two and a half years after launch, a $1.5 billion company with no revenue and no business model — "guessing Silbermann's plans to make money," wrote Vanity Fair, "has become a favorite Silicon Valley parlor game."
He told reporters he was focused on making the product better. He also told MIT Technology Review exactly where the money was, in one phrase, though almost nobody heard it as a business plan: Pinterest was valuable for "helping people to discover things that they didn't know they wanted."
When Carole Cadwalladr of The Observer went to meet Silbermann at Pinterest's brick-and-steel office in downtown San Francisco in 2014 — down the road from Groupon, around the corner from Airbnb, stir-fry smell drifting up through the triple-height space, a copy of Feminist Ryan Gosling on the coffee table — she found the office exactly as she'd expected and the man not at all. "If he were a character in a movie," she wrote, "it would star Jimmy Stewart." She had trouble picking him out on his own company's team page. She called him the Clark Kent of the San Francisco startup scene. His mother, she noted, was one of the site's most popular pinners. And she delivered a verdict on his interview answers that is both devastating and, in its way, admiring: "modest, measured, a bit boring."
Boring was the point. Over the following years, as its peers drowned in harassment campaigns, disinformation, and outrage, Silbermann methodically dismantled the idea that Pinterest was a social network at all.
"When we talk to people about Pinterest we often describe it as not a social network," he told The Guardian in 2016, at the company's new London headquarters, in a building originally built for a manufacturer of car speedometers. "Social networks are about communicating with other people. Pinterest is really about planning and getting ideas for your own personal life." He liked to quote the early users who had put it best: "With social networks, it's them time. With Pinterest, it's me time." To a student audience at the Economic Club of Chicago he framed it temporally: Facebook was for showcasing your past; Pinterest was for planning your future. To CNN's Laurie Segall, in 2019, he went further and said the company's goal was to get you offline — off the site and into your kitchen, your garden, your wedding.
The company called itself "the world's catalog of ideas." The word catalog is worth noticing. It was Tote's word. It had survived.
Product decisions followed the philosophy. In 2016 the "Pin It" button was renamed "Save." In 2017 the company removed the Like button entirely, on the grounds that it was redundant with boards; each user's existing likes were quietly gathered into a board called, simply, "Likes." The heart, the universal currency of social approval, was retired into a drawer. Engineering effort went into seeing rather than socializing: visual search in 2015, then in February 2017 a trio of tools — Lens, Instant Ideas, and Shop the Look — that let you point a phone camera at a lamp or a shoe and find its cousins. Silbermann told CNN he believed the camera would be the next keyboard. It was the old 2011 insight, sharpened: search, he'd said back then, "is not a great way to discover things," because "Amazon or Google does an awesome job of showing you what you're looking for if you know what it is." Pinterest was for when you didn't.
And then, quietly, the company began to do something no one in social media was doing: deciding, on purpose, what its users would not see.
In late 2018 Pinterest began blocking unproven and disproven cancer treatments from its recommendations. Then its staff noticed that the majority of vaccine-related images on the platform were anti-vaccine. In January 2019 the company simply stopped returning search results for vaccines. "We couldn't ensure that we were giving people great information," Silbermann said later, to Kara Swisher at the National Retail Federation, "and we made the decision then, that as a starting point, we would just not serve up content." Later, only verified sources like the CDC and NIH were permitted to answer. For searches associated with depression and anxiety, the company worked with clinical psychologists to build what it called Compassionate Search, designed, in his words, to normalize the emotions but not the behavior of self-harm. In June 2019 it banned the anti-abortion group Live Action for medical misinformation. In 2021 it became the first major platform to ban weight-loss ads altogether.
His explanation, offered to Swisher in January 2020, is the closest thing to a manifesto he has ever spoken aloud. "I think back then it was just sort of a foregone conclusion that if you build technology platforms, good things will automatically happen," he said. "The lesson that everyone's learned over the last few years is that if you want positive things to come out of internet technology, they have to be deliberately engineered that way." And then a sentence that recontextualizes the whole company: "A lot of the internet wasn't built with the needs of especially women in mind."
It was, depending on your vantage, either the most responsible content policy in Silicon Valley or a paternalistic retreat from the chaos that makes other platforms grow. It was both. That is the Pinterest problem in a sentence.
The Trouble with Being Pinterestly
In late 2017, Silbermann sent an email to the entire company urging employees to make decisions faster — not to wait for consensus before acting, to take more ownership. The people who received it heard a different message, which a former executive later distilled for CNBC with brutal economy.
— Former Pinterest executive, to CNBC, April 2019The knock on Pinterest is everyone was so nice. No one was willing to offend anyone.
Employees called themselves "Pinployees." They used the word "pinterestly" to describe the warm, respectful way colleagues were supposed to treat each other. In practice, according to interviews with twenty former employees, executives, and contractors, it meant decision-making by committee, endless rounds of meetings, politicking in place of direct advocacy, and a reluctance to commit to strategic direction that cost the company on e-commerce and video. Senior leaders were leaving: Jon Alferness, who ran ads and commerce; Li Fan, the head of engineering. A leaked 2015 Andreessen Horowitz presentation had projected 2018 revenue of $2.8 billion; actual 2018 revenue was $756 million — up 60 percent, impressive by any standard but the one it had been held to. In October 2016 the company announced 150 million monthly users, a milestone it had privately aimed to hit by the end of 2015.
The patience that had saved Pinterest in 2010 was now being read as indecision. In 2018 a reporter spoke to more than a dozen people who had worked with or for the company, and found frustration in every direction. Matt Novak of All Blue Capital, which had bought shares on the secondary market, said his firm was trying to sell. "If they don't keep up," he said, "they very quickly become prehistoric." Silbermann's answer was a sentence that could serve as his epitaph, depending on who writes it: "There's a natural rate at which you can scale a company that's healthy."
He was not oblivious. At a Village Global dinner around that time he talked about keeping a decision journal — writing down each significant call and his reasoning at the moment, so that later he could check his judgment against the outcome rather than against a memory "updated along the way." He talked about asking, of any decision, "If I joined the company tomorrow and heard about this decision and it was stripped of all context, what would it symbolize?" He reminded his team, he said, that "sometimes a good decision is a timely decision." He confessed that he had spent too long trying to make Pinterest resemble his hero companies, and that he wished he had "earlier celebrated the things that made our company weird." This is a man who read a different business book every weekend, according to a Lightspeed partner who hosted him in 2012 — the conscientious son, still studying for an exam nobody else could see.
He was also, by temperament, a product person who had to learn to stop being one. "I'm still involved in product," he told Startup School in 2016, "but my job isn't to pixel critic the service. It's hopefully to ask the right questions and get the teams to come up with better ideas." He had come to believe, he said at Village Global, that he should have focused on "the who as much as the what in every decision" — that the founder who solves problems on a whiteboard eventually has to become a founder who chooses the people who solve them. He called his lowest moment not a funding crisis but a night when engineers warned him that a database sharding operation could fail and destroy all the data. The collector's nightmare: the drawer, emptied.
Then, in June 2020, the niceness itself came under indictment.
Ifeoma Ozoma had joined Pinterest from Facebook in July 2018 as a public policy and social impact manager, drawn, she said, by its reputation as one of the friendlier places in tech. "I thought, 'This is a company that believes in diversity, or at least says it does,'" she said. She led the vaccine misinformation work that the company would later showcase as evidence of its conscience. She pushed the platform to stop promoting weddings at former slave plantations, which it did in December 2019 after a campaign by Color of Change. And then, in her performance review, she was dinged for failing to present the arguments in favor of continuing. "I should have come up with the pros for promoting slave plantations," she said. "Like, to this day, I think about that, and the fact that it was put in writing — in a performance review that then impacted my pay." She and her colleague Aerica Shimizu Banks, also a Black woman in the same role, went public on Twitter alleging underpayment, racist comments, and retaliation. Ozoma said a white male engineer had shared her personal information with a site known for doxxing.
The company's first statement was a reflex of institutional self-defense: "We're confident both employees were treated fairly." Within a week it had reversed. Silbermann sent an email to all staff, obtained by Bloomberg, containing a sentence nobody had ever heard from him about his own company: "What I've learned over the past few weeks is that parts of our culture are broken." And: "I'm truly sorry for letting you down." He pledged to add a person of color to the board by year's end. In August, dozens of employees staged a virtual walkout. In December, Pinterest agreed to pay its former chief operating officer more than $20 million to settle a gender discrimination lawsuit — among the allegations reported in the press was that Silbermann had sidelined women in leadership. In November 2021, it settled a shareholder suit over the Ozoma and Banks allegations, committing $50 million to diversity work and releasing former employees from nondisclosure agreements. The total, by Fortune's count, came to more than $70 million.
There is a terrible irony in the juxtaposition, and it should not be resolved too quickly. A company built for women, whose founder said the internet had failed women, whose culture prized gentleness so much that executives complained no one would offend anyone, was accused by the women and people of color inside it of being something else entirely. Niceness, it turned out, was not the same as fairness. A culture can be so allergic to conflict that the person raising a legitimate grievance becomes the problem. The quality that made Pinterest the internet's last gentle place was the same quality that let its people go unheard.
Finding, Not Buying
Pinterest went public on April 18, 2019, at $19 a share, and closed its first day of trading at $24.40. It had once been privately valued at $12 billion and had sought a lower valuation in its IPO; the market restored the higher number in an afternoon. That fall, Silbermann flew to Washington to accept the Center for Excellence in Education's Lieberman Award at the Russell Senate Office Building. It came with a $10,000 stipend. He gave it back to the organization that had sent him to MIT at sixteen.
The pandemic was very good to Pinterest. People stuck at home redecorated, baked, planned weddings they could not yet hold. Advertising revenue for 2020 reached $1.7 billion, up 48 percent. In 2021 the company earned $2.58 billion in revenue and $316 million in net income. Monthly users peaked near 478 million. In October 2021, Bloomberg reported that PayPal was exploring a purchase at around $70 a share; within a week, after its own shareholders balked, PayPal walked away.
And yet the question that had shadowed the company since Tote never went away. The IPO prospectus promised to make it "easier for Pinners to go from inspiration to action — in particular, we want to make Pinterest more shoppable." A 2019 Cowen survey found that 47 percent of Pinterest's U.S. visitors used the site to find or shop for products, versus 15 percent or less on Facebook, Instagram, Snapchat, and Twitter. The intent was there. The cash register wasn't. In 2015 the company had launched Buyable Pins — and taken no cut of the purchases. In the first quarter of 2022, Pinterest earned $5 per user in the United States and Canada; Facebook earned $48.29. The Information reported that Silbermann's deliberate style had contributed to two e-commerce acquisitions falling through, including a potential $500 million purchase of Verishop. A native checkout option arrived only in early 2022.
The search giant that had once hired him, meanwhile, delivered a blow from the outside. After a Google algorithm change in November 2021, Pinterest reported that U.S. monthly users arriving from the web had dropped around 30 percent year over year. The man who had built a discovery engine because search was bad at discovery found that much of his audience had been discovering him through search all along.
On June 28, 2022, Pinterest announced that Silbermann would step down as CEO and become executive chairman, a newly created role. His successor was Bill Ready, Google's president of commerce and payments — a payments executive, the precise kind of person Tote had needed in 2008. Silbermann was generous to the point of self-erasure. Ready, he told The Wall Street Journal, was "actually going to be a better CEO than I am for this next chapter." Fortune's eulogy was gentle and merciless at once: it would be "grossly unfair" to call his tenure a disappointment — 433 million users, $2.6 billion in revenue, a $13 billion market cap — and yet it was hard to shake "that what-might-have-been feeling."
The next evening, Silbermann and Ready sat for an interview with Jim Cramer on Mad Money. Cramer called Silbermann a hero, called the decline from 478 million users to 431 million something only Wall Street would regard as suboptimal, and asked whether Wall Street had made his job tiresome. He also, with the cheerful confidence of television, told Silbermann that he'd come from Nebraska. Silbermann did not correct him. He answered, as he always did, about consumers — "the thing I wake up doing every day" — and about the acquisition of a fashion startup called THE YES, and about the resources the company now had to invest. When Cramer brought up the Google algorithm change, sitting beside the man who might have been involved in it, Silbermann said, mildly, that they hadn't been talking much about Ready's last job.
Afterward the company would partner with Amazon to bring outside ads onto its pages, and revenue would climb past $4.2 billion in 2025; in January 2026 it would cut about 15 percent of its workforce as it struggled for ad dollars against deeper-pocketed platforms. But the moment that lingers from the interview came near the beginning, before any of the strategy talk, when Cramer explained that he had been texting that day with his daughter, a baker who went to Parsons, and asked her what she thought of Pinterest.
She loved her mood boards, she told him. For decorating her house. For planning her wedding. For finding inspiration for next year.
And had she ever bought anything on Pinterest?
"Hell no."
A stranger, somewhere he'd never been, using something he made, and finding it kind of useful.
Part IIThe Playbook
What follows is not a hagiography of Silbermann's methods but a distillation of them — including the ones that worked too well. His career offers an unusually clean natural experiment: nearly every trait that built Pinterest also constrained it. The principles below are drawn from that record, and several carry their own warning labels.
The Silbermann Playbook
- 1.Follow the behavior, not the business model.
- 2.Ship when one thing is worthy of a stranger's time.
- 3.Find more people like your first lovers — not the users you wish you had.
- 4.Make early users feel like correspondents.
- 5.Match your users' effort with your own craft.
- 6.Engineer the outcome you want; don't expect it.
- 7.Define yourself by what you refuse to be.
- 8.Repair crises with tools, not statements.
- 9.Put a clock on consensus.
- 10.Keep a decision journal and judge decisions as precedents.
- 11.Know which chapter you're the right author for.
Principle 1
Follow the behavior, not the business model
Tote was designed to sell things, and its users declined to buy them. What they did instead — grabbing items, tagging them, showing them to friends later — was not in the plan, and Silbermann's decisive act was to treat that unplanned behavior as the real product. Brian Cohen's memory of him is not of a great pitcher but of a watcher: "He recognized that these women were sharing their tastes." The pivot from catalog to collection is one of the cleanest examples in consumer tech of a founder letting user behavior overrule his own thesis.
The cost is equally instructive. Silbermann followed the behavior so faithfully that the original monetization logic — commerce — was set aside for a decade. Buyable Pins launched in 2015 without taking a cut; native checkout didn't arrive until 2022. Following behavior is a superb way to find product-market fit and an incomplete way to find a business. The behavior tells you what people love; it does not automatically tell you what they'll pay for, or how the company will capture the value it creates.
Tactic: When users repurpose your product, build for the repurposing — but write down, on day one, the specific mechanism by which that new behavior will eventually produce revenue, and revisit it every quarter.
Principle 2
Ship when one thing is worthy of a stranger's time
Silbermann rejected the reflexive minimum-viable-product gospel not out of perfectionism but out of a diagnosis of why Tote failed: "We weren't great at one thing." His replacement standard was precise — ship when you have "one thing that is worthy of someone's time." For Pinterest that one thing was aesthetic: the collections had to look good enough that people would want to show them to friends. If the product's single great quality isn't there, he argued, you can't even learn from feedback, because people will just tell you what you already knew — that it's not good.
This reframes the launch question from how little can we build to what is the one thing we must be excellent at, which is a far more useful constraint. Evan Sharp's fifty functional layouts were not gold-plating; they were the company concentrating all of its limited resources on its single point of differentiation. Everything else — engineering robustness, business model, press — was allowed to be mediocre for a while.
Tactic: Before launch, name the one quality a stranger would notice in thirty seconds, and refuse to ship until that single quality — not the feature list — is genuinely excellent.
Principle 3
Find more people like your first lovers — not the users you wish you had
Silicon Valley wisdom said a social product should win over technologists first and diffuse outward. Pinterest's first enthusiasts were Silbermann's friends from Des Moines, planning homes and dinners. A weaker founder would have retooled for the audience he wanted. Silbermann asked instead: "Maybe I can just find more people like me" — and went looking for where those people gathered, which turned out to be design and lifestyle blogs. "If we had been really dogmatic about wanting cool Silicon Valley people to like it," he said, "we probably wouldn't have made the service we made."
The lesson is about the asymmetry of love. Three thousand lukewarm signups are worth less than a handful of people who genuinely use the product for their actual lives. Chris Dixon's admission — "Everyone's envious. I'm envious." — is what it sounds like when an industry realizes it was looking at the wrong map. The mainstream audience other startups struggle to reach after conquering the Valley was, for Pinterest, the starting point.
Tactic: Identify your ten most devoted users, map the communities they already belong to offline and online, and spend your next month of marketing exclusively inside those communities.
Principle 4
Make early users feel like correspondents
His cell phone number in every support email. Personal letters to the first 5,000 users. Meetings in person. Coffee-shop sessions where he asked strangers what they expected to see before they pressed a button and whether they saw it afterward. Silbermann's early user care was not a growth tactic so much as an act of correspondence — the posture of a man writing back to everyone who wrote to him. "Since it took us so long to get those users," he said, "we cared about them so much."
The underappreciated benefit was informational. That predict-then-observe button test is a miniature usability lab requiring no budget. The phone calls when the site went down told him exactly how much people depended on it. Founders who outsource support early also outsource their best stream of unfiltered information about what the product means to people.
P
The Early-User Toolkit
The low-cost methods Silbermann used before Pinterest had real traction.
| Method | What it produced | Stage |
|---|---|---|
| Cell number in every support email | Real-time signal on outages and dependency | Scan |
| Personal letters to first 5,000 users | Qualitative feedback and loyalty | Scan |
| Coffee-shop "predict, then press" tests | Free usability research on rough edges | Try |
| Blogger events and "Pin It Forward" | Community-to-community distribution | Scale |
| Faking a crowd in the Apple Store | Morale, mostly | Risk |
Tactic: Until you have a thousand active users, personally answer every support message and run one in-person "predict, then press" test every week.
Principle 5
Match your users' effort with your own craft
"You have to show that you put in as much effort as you expect [your users] to put in." This is the most quietly profound thing Silbermann has said about product, and it applies to any product that asks users to create. Pinterest asked people to spend hours curating boards. A clunky interface would have communicated that the company didn't take that labor seriously, and users would have reciprocated by not taking it seriously either.
The principle generalizes beyond aesthetics. Any product built on user contribution — reviews, playlists, documents, code — is engaged in an implicit exchange of effort. Polish is a form of respect, and respect is a form of retention. The grid that half the internet copied was not just a visual innovation; it was a promise that your collection would be displayed with care.
Tactic: For every action you ask users to perform, audit whether the result is presented with at least as much craft as the effort it demanded — and fix the ones where it isn't.
Principle 6
Engineer the outcome you want; don't expect it
"If you want positive things to come out of internet technology, they have to be deliberately engineered that way." Silbermann's content policies — blocking vaccine searches in January 2019 rather than serving misinformation, Compassionate Search for mental-health queries, banning weight-loss ads in 2021 — followed from rejecting the belief that neutral platforms produce good results by default. "If you don't take some responsibility for what people see," he told Kara Swisher, "you're at some level responsible for the downstream consequences of that."
The strategic insight is that Pinterest's safety was not a cost center but a positioning. Advertisers and users alike came to understand it as the internet's gentle place, and that identity was manufactured, choice by choice, through deliberate restrictions. The tradeoff is real — curation can shade into paternalism, and engineered calm grows more slowly than engineered outrage — but the alternative, as Pinterest's peers learned, is having your product's character defined by its worst users.
Tactic: Write down the three outcomes your product must never produce, and design at least one explicit mechanism — not a policy statement — to prevent each.
Principle 7
Define yourself by what you refuse to be
"We're not a social network." Silbermann repeated it for years, and then he made it true: renaming "Pin It" to "Save," removing the Like button in 2017 and folding likes into a board, emphasizing planning over communication — "With social networks, it's them time. With Pinterest, it's me time." Each negative definition clarified a positive one. By refusing to compete on social graph, he avoided a war with Facebook he could not win and claimed a category — visual discovery for one's own future — that no one else was defending.
He later admitted the opposite tendency had haunted him: years spent trying to make Pinterest resemble his hero companies. "I would've earlier celebrated the things that made our company weird," he said, "because those ended up actually being the things that make it special." Refusal is how a company protects its weirdness from the gravitational pull of more famous peers.
≠
Conventional Wisdom vs. Silbermann
Where Pinterest deliberately ran against the Silicon Valley grain.
| Conventional wisdom | Silbermann's approach |
|---|---|
| Win tech early adopters, then the mainstream. | Start with friends in Des Moines and lifestyle bloggers. |
| Ship a minimum viable product and iterate in public. | Ship only when one quality is worthy of a stranger's time. |
| Maximize engagement through social approval. | Remove the Like button; optimize for "me time." |
| Platforms are neutral; serve what users search for. | Refuse to serve vaccine results you can't vouch for. |
| Move fast and break things. | "There's a natural rate at which you can scale a company that's healthy." |
| Court the press; tell your story loudly. | "We've been too busy to hire a press person yet." |
Tactic: Write a one-sentence statement of the category you refuse to compete in, and use it to kill at least one roadmap item that would quietly pull you toward it.
Principle 8
Repair crises with tools, not statements
When Kirsten Kowalski's copyright post went viral in February 2012 and critics compared Pinterest to Napster, Silbermann did not issue a defiant manifesto. He called Kowalski. Within roughly ten weeks the company shipped a "nopin" opt-out tag (February 20), rewrote the terms to remove the right to sell user content (announced March 23, effective April 6), and added automatic attribution for Flickr, Behance, YouTube, and Vimeo (May 1). Flickr, which had been the first site to use the opt-out, ended by adding a Pin button to its own share menu. The adversary became a distribution partner.
The 2020 discrimination crisis offers the contrast. The company's first move was a statement — "We're confident both employees were treated fairly" — and it had to be reversed within a week by a much costlier one: "parts of our culture are broken." The settlements that followed exceeded $70 million. Statements are fast and cheap; tools and structural changes are slow and credible. When the first response is a statement, the second is usually an apology.
Tactic: When a crisis breaks, call the person who raised it before issuing any public statement, and make your first public communication announce a concrete change rather than a defense.
Principle 9
Put a clock on consensus
This is the principle Silbermann learned the hard way, and it is the one most worth stealing from his mistakes. The "pinterestly" culture — warm, respectful, consensus-seeking — produced, by the account of twenty former insiders, decision-making by committee, politicking in place of advocacy, and missteps on commerce and video. The late-2017 companywide email urging faster decisions arrived years after the habits had set. Reported acquisitions, including a potential $500 million purchase of Verishop, fell through amid deliberation.
Niceness is not the problem; unbounded consensus is. A culture that equates disagreement with unkindness will eventually stop surfacing the disagreements it most needs to hear — including, as the Ozoma and Banks allegations suggested, grievances about fairness itself. Silbermann's own later formulation is the remedy: "Sometimes a good decision is a timely decision." The fix is structural: name the decider, set the deadline, and make dissent a duty rather than a breach of manners.
Tactic: For every significant decision, name a single owner and a hard deadline at the outset, and require that at least one dissenting view be written down before the call is made.
Principle 10
Keep a decision journal and judge decisions as precedents
Silbermann's decision journal is the most transferable habit in his playbook. Write down the decision and the reasoning at the moment you make it, because otherwise your memory is "this kind of running dialogue that's being updated along the way." Later you can check whether your instincts were sound against the record, not against a flattering reconstruction.
He paired this with two framing questions. First: "What if this decision is a precedent?" — would you be comfortable if it became the new baseline? Second, the outsider test: if someone joined tomorrow and heard about this decision stripped of all context, "what would it symbolize?" And he widened the circle of communication beyond the person directly affected to "the two rings of people around the person." Together these turn individual calls into a coherent body of case law — the culture, written down.
Tactic: Start a decision journal this week; for every hiring, partnership, or product-spec decision, record the reasoning, the alternatives rejected, and what the decision would symbolize to a newcomer.
"The founder is always responsible for driving clarity on what you're building and why," Silbermann said in 2016. What changes is that "more and more your job becomes creating an organization and finding people that can build that and that will do it better than you." In June 2022 he applied that logic to himself, handing the CEO role to Bill Ready, a commerce and payments executive, and saying plainly that Ready was "actually going to be a better CEO than I am for this next chapter."
It is rare for a founder to diagnose his own limits so publicly, and rarer still to stay on as a constructive executive chairman. The critique — that the commerce gap was visible for years and the handoff came late — is fair. But the move itself reflects a principle every founder should hold in reserve: the qualities that got the company to one stage may be precisely the wrong ones for the next, and recognizing that is an act of stewardship rather than defeat.
T
The Silbermann Arc
Key turning points, from Cold Brew Labs to executive chairman.
2008
Silbermann leaves Google; with Paul Sciarra, founds Cold Brew Labs in August. Tote follows.
2010
Pinterest launches in invite-only beta; four months in, ~3,000 accounts.
2011
Bessemer leads $10M Series A; Andreessen Horowitz leads $27M at a $200M valuation.
2012
Copyright fixes; Sciarra departs; Rakuten leads $100M at $1.5B; site opens to all.
2014
Promoted Pins launch; first revenue; $5B valuation.
2017
Like button removed; Lens launches; valuation ~$12.3B; "decide faster" email.
2019
Vaccine search results blocked in January; IPO on April 18 at $19.
2020
"Parts of our culture are broken"; walkout; discrimination settlements begin.
2022
Bill Ready named CEO; Silbermann becomes executive chairman.
Tactic: Once a year, write a candid one-page memo to your board describing the skills the company's next stage requires and honestly assessing whether you possess 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
— Ben Silbermann, The New York Times, November 2011The things you collect say a lot about who you are. The books on someone's shelf, you just learn a lot about that person by looking at them.
— Ben Silbermann, Y Combinator Startup School, 2012My advice is you should ship when you have one thing that you're proud of, like one thing that is worthy of someone's time.
— Ben Silbermann, Recode Decode with Kara Swisher, January 2020The lesson that everyone's learned over the last few years is that if you want positive things to come out of internet technology, they have to be deliberately engineered that way.
— Ben Silbermann, Village Global dinner, 2018I would've earlier celebrated the things that made our company weird because those ended up actually being the things that make it special and successful.
— Ben Silbermann, email to Pinterest employees, June 2020What I've learned over the past few weeks is that parts of our culture are broken.
Maxims
- Watch what users save, not what they buy. The behavior a product accidentally enables is often more valuable than the one it was designed for.
- One great thing beats ten good ones. Ship when a single quality is worthy of a stranger's time, and not before.
- Early adopters come from everywhere. The first people who love your product may live in Des Moines, not Palo Alto — go find more of them.
- Answer the phone yourself. Personal support in the early days is the cheapest research and the deepest loyalty you will ever buy.
- Match the effort you ask for. If users must labor to create, present their work with at least as much craft as they put in.
- Engineer the good on purpose. Neutral platforms do not produce benign outcomes; deliberate design does.
- Define yourself by refusal. Knowing which category you will not compete in protects the weirdness that makes you special.
- Investors can be wrong. Remember that the deals they passed on are the ones that haunt them at night.
- Niceness needs a deadline. A culture that cannot disagree will not decide, and will not hear the grievances it most needs to.
- Write the decision down. A journal of reasoning, judged as precedent, is the founder's best defense against a memory that flatters.
