In the Dropbox Folder

The summer Dylan Field spent at Flipboard in 2011 was the one that never left him. He was there as a technical product manager, taking a semester off from Brown, watching one of Silicon Valley's most design-obsessed companies build its product. And what he kept watching, with the specific attention of someone who couldn't quite let it go, was how broken the actual process of designing was.

The team was in Fireworks almost every single day, collaborating through a Dropbox folder. There were attempts to do a blog where they could post work in progress, but honestly all the collaboration was a mess, and that was with a very design-forward team. Flipboard cared about design more than almost anyone. And still, the way designers worked with each other, the way they shared files, the way they told each other which version was current, resembled something from ten years earlier. A Google Doc was a link. A design file was an attachment in an email, or a Dropbox notification, or a conversation that started with "wait, which one are you looking at?"

Field filed this away. He went back to Brown. And then he met Evan Wallace.

Wallace was a teaching assistant in the CS department, a year ahead of Field, and he had been doing something unusual in his spare time. He had built a WebGL demo of a swimming pool, water refracting light across the bottom tiles in real time, fully interactive in a browser window, rendered with a quality that had no business existing on the web. WIRED wrote it up. Field had been blown away by it. He saw the experiments with WebGL, a technology that let you render GPU-level graphics inside a browser, and was convinced that the sky was the limit for what could be done in that space.

The conversation between them started with what was possible technically. It kept coming back to what Field had seen at Flipboard. Evan tried to convince Dylan that they should go build Photoshop in the browser. Dylan's first reaction was that that was impossible and a terrible idea. From there, they kept talking more and more. Would this be possible? Would they do it? They started looking really deeply into WebGL and convinced themselves that, yes, it was possible.

What tipped it was Field's time as an intern. The technical possibility and the observed pain were two halves of the same idea: design collaboration was as broken as document collaboration had been before Google Docs, and the browser had just become capable enough to fix it. The design team at Flipboard would spend hours on version control and files, yet still struggle with collaboration. Designers would go off in their own corner for a few days and emerge with new designs, instead of bringing everyone along. It was like dealing with Microsoft Word files in the era of Google Docs.

Field applied for the Thiel Fellowship. His parents, who had poured everything into his education, were not happy about it. His father, a respiratory therapist, had given Field this straightforward message: "Pretty much everything we earned went to education." Field got the fellowship anyway, dropped out, moved to San Francisco, and Wallace followed. They started full-time in August 2012 with $100,000 between them and no product, only the conviction that design tools were about to become collaborative, and that the window to build it was open right now.

What they were actually building

For the first year, they didn't entirely know. The founding thesis was clear but the product was not. They tried drone software first, then pivoted to computational photography, then photo editing, then a meme generator. It wasn't until June 2013 that they really said, okay, let's focus on interface design. And even then, it was interface design "plus other things," and it wasn't clear what those other things were. There was a point where they had to sit down and define it. They got in a room together, went through everything they thought they were building, and crossed things off one by one until what remained was what they were actually doing.

That focus was not obvious at the time. "Photoshop in the browser" was a description broad enough to mean almost anything. What they were choosing, specifically, was interface design: the work of designing the screens of software products. A market that in 2013 looked narrow. IBM had one designer for every 72 engineers. The idea that design would become a cross-functional activity, that product managers and engineers and executives would all end up working inside design files regularly, was not a mainstream view. The TAM, priced in 2013 terms against the number of professional UI designers in the world, was not impressive.

Index Ventures partner Danny Rimer heard them pitch and was uncertain about what the product would be. But listening to Field talk, he knew he was someone who would pour his heart and soul into bringing it to life. For Rimer, that was enough. He wrote a check. The seed round closed at $3.8 million. Figma had a little money and a very large technical problem.

The physics problem underneath

Building a design tool in the browser in 2012 was not an engineering challenge. It was closer to a physics problem. Professional design work requires constant, real-time rendering of complex visuals: anti-aliased curves, transparency effects, blending modes, blur, masking, all of it responding instantly to a cursor moving across the screen. The web's existing graphics primitives in 2012 were nowhere near capable of this. HTML Canvas was too slow. SVG was too limited. The gap between what browsers could draw and what professional designers needed was enormous.

Evan Wallace had spent years understanding exactly this gap, which is why he was one of the few people who knew it could be closed. WebGL was a browser graphics API originally designed for 3D applications. Betting on it for complex 2D work was unusual. It wasn't widely used for this at the time. But the team saw its potential to power a smooth, infinite canvas in the browser.

What Wallace and early engineer Alec Miller actually built was not a standard web application. The editor was written in C++ and cross-compiled to JavaScript. Internally the code looked like a browser inside a browser: they had their own DOM, their own compositor, their own text layout engine. The renderer was a highly optimized tile-based engine with support for masking, blurring, dithered gradients, blend modes, nested layer opacity, and more, all done on the GPU, fully anti-aliased.

This is what four years of stealth actually meant. Not four years of product discovery. Four years of building the engine that made the product conceivable. And the payoff was unexpected: despite running in the browser, Figma's rendering engine was often faster than the rendering engines in competitor native applications. Sketch, the dominant design tool of the era and a macOS-native app, barely used the GPU at all. The company that was supposed to be fighting uphill against desktop tools had built something technically superior to them.

The choice of C++ compiled to JavaScript also meant that as browser technology improved, Figma could adopt it immediately. When WebAssembly arrived as a standard, Figma's load time dropped by more than 3x. Every subsequent generation of browser improvement compounded on the original investment. The hard early bet kept paying dividends long after Wallace had moved on.

The collaboration nobody wanted

Here is what people forget about Figma's launch: designers hated the central idea.

When Field and Wallace started showing people what Figma could do, the reaction to real-time collaborative editing was not excitement. It was fear. No one was clamoring for a multiplayer design tool. If anything, people hated the idea. Designers worried that live collaborative editing would result in "hovering art directors" and "design by committee" catastrophes. Design, in the cultural self-understanding of the profession in 2016, was not a collaborative activity. It was a solitary one, protected from interference, something you emerged from with a finished thing. The idea of seeing someone else's cursor moving around in your file was viscerally uncomfortable.

One of the comments on Designer News when Figma launched read: "If this is the future of design, I'm changing careers." Another one Field liked was: "A camel is a horse designed by committee."

The team launched in closed beta in December 2015 without multiplayer at all. They knew it was the product's real differentiator, but it wasn't ready. They spent the next nine months building it while the closed beta gathered feedback, then shipped it publicly in September 2016. Ultimately, they had to do it because it just felt wrong not to offer multiplayer as a tool on the web. It eliminates the need to export, sync, or email copies of files, and allows more people to take part in the design process. Just by having the right link, everyone can view the current status of a design project without interrupting the person doing the work.

That last line is the strategic insight the design community took years to absorb. The cursors were not the product. The link was the product. A Figma file lived at a URL. You sent it. The person on the other end opened it in their browser, saw the latest version, and could leave a comment or inspect a component without installing anything, without creating an account if they were just viewing, and without Field's team needing to close a sales deal. Every time a designer shared a Figma file with a developer, or with a product manager, or sent a prototype link to a stakeholder, Figma acquired a new user. The product was its own distribution channel.

Three years of not charging, and why that was a choice

Between 2016 and 2017, Figma had a meaningful and growing user base and essentially no revenue. This was intentional, but it was also becoming a problem.

The logic of staying free was correct: the product spread because it cost nothing to try, and the mechanism that spread it (the shareable link) only worked if the person receiving the link didn't hit a paywall when they opened it. Viewers had to be free for the loop to function. But free also meant that enterprise procurement couldn't adopt it, because procurement requires a vendor that isn't going to die. Jon Friedman, a design executive at Microsoft, visited Figma's San Francisco headquarters to deliver that message directly. Field remembered Friedman saying: "Look, we're all worried you're going to die as a company. We can't spread it inside Microsoft as a company even though we like it, because you're not charging."

The same pressure was coming from inside the company. Index Ventures partner Danny Rimer remembered the board lunch where they urged Field to start charging. An enterprise customer had told him the same thing: "We can't scale Figma internally because no one believes you'll be around in a year."

Field's response to that pressure reveals something important about his understanding of the product. He did not just introduce a paid tier. He made a very specific choice about what to charge for. Editors paid. Viewers did not. Field explained it directly: "One of our intentional strategies was to make it so that viewers were free in Figma. We saw other competitors charge for viewers and also wanted to get more people into our tool, with the hope that over time, they become editors and see the power of Figma for editing."

The viewer-is-free decision is the one that often gets glossed over in the Figma story, but it is probably the most consequential pricing call the company ever made. It kept the link mechanism working even after monetization began. The developer who received a Figma link could still open it, inspect the design, copy the CSS values, and leave a comment, all without paying. This made Figma genuinely useful to engineers in a way no other design tool had been. Engineers became Figma advocates inside companies. They were the ones who went to their managers and said this is what we should standardize on, not the designers, whose opinions were expected, but the engineers, whose opinions carried different weight in engineering-led organizations.

Revenue went from $700K in 2017 to $4M in 2018 to $25M in 2019. The growth rate was extraordinary. The pricing structure had unlocked something the free period had built but not yet captured.

How Sketch lost and InVision disappeared

Sketch in 2016 was the dominant UI design tool, beloved by the design community, far ahead of Adobe's aging suite in terms of what it had been built to do. It had everything Figma needed to beat except one thing: a file was still a file.

Sketch files lived on your Mac. They moved through Dropbox. Reviewing a Sketch design meant either getting a Sketch license or using InVision, a separate tool that let you turn Sketch exports into clickable prototypes. Developer handoff happened through Zeplin, yet another separate tool. Version control happened through Abstract, yet another. The design workflow in 2016 was five tools for one job.

Figma's competitive strategy was not to outfeature Sketch on design capabilities, though it did over time. The strategy was to absorb the ecosystem. Each time Figma added prototyping, or developer inspect, or design system libraries natively, one more external tool became redundant. The value of a Figma file grew, because every stakeholder could access it from the same URL, and the cost of switching away grew with it.

Figma also made it easy to get in. Sketch import worked from the start, so moving from Sketch to Figma required almost no friction. Sketch did not build Figma import until 2023, eight years later. Aakash Gupta By then the game was over.

Figma went from 10% market share in 2017 to 90% in five years. InVision eventually shut down its main product in 2024. Adobe placed XD into maintenance mode the moment it announced it was acquiring Figma, effectively conceding that competing was no longer viable. The company that had seemed to be fighting uphill against entrenched incumbents had, through a combination of a genuinely better technical architecture and a distribution model that none of its competitors could copy without rebuilding from scratch, ended up owning the market.

The management crisis nobody talks about

Field was twenty years old when he started Figma. He had been an intern before that. The qualities that made him a compelling founder, intensity, very high standards, conviction that everything could be done better, turned out to be difficult management traits when applied to a small team under pressure with no product to show yet.

He said later: "I was just not a very good manager when I started Figma. I was always very optimistic; I thought shipping was right around the corner, so I wasn't setting expectations correctly." The pressure of building for years without launch accumulated. At one point the senior members of his team staged an intervention. "It was like, 'You need to get some help.' Afterward I took a few days away from the office."

What followed was one of the more decisive personnel moves in Figma's early history. Field brought in Sho Kuwamoto, a twenty-year veteran of Macromedia, Adobe, and Medium, who joined in 2015 as director of product. Kuwamoto's role was partly to ship the product and partly to stabilize the team. He described Field's disposition with the precision of someone who had figured out how to work with it: "One of Dylan's superpowers, and his kryptonite at the same time, is he really cares a lot about making sure that everything's going to be successful, so he'll just put everything into it."

Around the same time, Claire Butler joined as the tenth employee and the company's first business hire. She started building community relationships before the product was publicly available, running the thesis that community was not something you added after product-market fit but something that created it. She treated every small feature launch as a major launch, to build credibility with the design community and keep nudging people to try the product. The community strategy she built eventually meant that when Figma was ready to expand into enterprise, 70% of enterprise deals originated from designers who had already adopted the product from the bottom up.

Rimer and Lilly, the two lead investors, had heard from employees and sat Field down. "We both heard it," said Rimer. "We sat down with him and explained the situation. We heard it and we said, 'Look, this is an impasse. You're going to have to adapt and change.' And he heard it and he changed. I think that's such a great character trait of Dylan, is to hear the information, be objective about it, process it, and accept it and act accordingly." Field turned out to be genuinely coachable in a way that many founders are not, especially when the feedback came from people he trusted.

The deal that suspended everything

In September 2022, Adobe announced it would acquire Figma for $20 billion. At the time Figma's annual recurring revenue was around $400 million, implying a fifty-times multiple. It was the kind of price that made the design world pause and update its priors about what a design tool could be worth.

For fourteen months, the company lived in regulatory limbo. The US Department of Justice, the European Commission, and the UK Competition and Markets Authority all opened investigations. Adobe had already placed XD into maintenance mode the moment the deal was announced, which made its competitive intent visible to regulators. That transparency cost them. In December 2023, the companies mutually agreed to terminate the merger, citing no clear path to regulatory approval. Adobe paid a $1 billion breakup fee.

The fourteen months of limbo were not neutral. Recruiting was harder. Some employees who had been expecting acquisition liquidity started looking elsewhere. The independence narrative that had always been part of Figma's appeal, the scrappy challenger taking on Adobe, was complicated by the fact that Adobe was about to be the parent company. The story the company had been telling about itself didn't quite fit.

What happened after the deal collapsed was not what most people expected. Figma poured the breakup fee into product. By 2025 it had expanded from four products to eight, launching over 200 features in a single year. FigJam had already launched in 2021, extending Figma's footprint to the brainstorming and whiteboarding that happened before design work began. Dev Mode launched in 2023, translating design files directly into code-ready specifications for engineers. The platform was expanding in both directions along the workflow, upstream toward ideation and downstream toward shipping.

When Figma went public on July 31, 2025, its stock surged 250% on the first day of trading, giving the company a market value of over $61 billion. The company Adobe had tried to buy for $20 billion was worth three times that as an independent entity. That outcome was not inevitable. It required surviving the limbo, holding the organization together through the uncertainty, and having a clear enough product strategy to sprint when the constraint was removed.

What the story is actually about

Figma is usually described as the company that put design in the browser and added collaboration. That framing is accurate but it misses what was actually hard, and what was actually strategic.

The hard part was building a rendering engine good enough to make professional designers forget they were in a browser. That took four years and required one of the most technically unusual choices in design tool history: writing the editor in C++, compiling it to JavaScript, and using WebGL for rendering at a time when nobody used WebGL for 2D professional software. That decision was made in 2012 based on Evan Wallace's understanding of what was possible, not what was proven.

The strategic part was understanding that the browser choice was not only a technical choice. It was a distribution choice. A Figma file was a URL. A URL can be emailed, slacked, embedded, shared with a client, or handed to an engineer without any ceremony. Every time that link was shared, Figma extended its reach. The product grew by being used, not by being sold.

Charging only editors, not viewers, protected that mechanism after monetization began. It also created the conditions for Figma to become genuinely cross-functional: designers designed, engineers inspected, product managers commented, executives reviewed, all in the same file, all from the same link. That is why the market Figma ended up serving was so much larger than the market it initially targeted. One in five Figma licenses, by one estimate, belongs to an actual designer. The rest belong to everyone else in the building process.

The thing that made all of it possible was a founder who understood the problem because he had watched it happen, who found a cofounder who understood the solution because he had spent years at the edge of what browsers could render, and who was coachable enough, when his management style threatened to break the team, to change. That combination, observation, technical depth, and genuine openness to feedback, is rarer than any single one of those traits on its own.

What builders take from this

The product IS the distribution channel. Figma did not grow because of a sales team or a marketing budget. It grew because every shared link was an acquisition event. The browser architecture was chosen because it was technically necessary, but it turned out to also be the best possible distribution mechanism. When your product's core interaction and its growth mechanism are the same thing, growth becomes structurally durable.

Free users are not a cost center. They are the channel. Every viewer who opened a Figma link without paying was a potential future editor, a future advocate, a future person who brought Figma to their next company. Charging viewers would have sealed the product inside design teams. Making viewers free made the product everyone's problem to champion.

The TAM you start with is not the TAM you end with. Figma entered a market that looked like "professional UI designers." It ended up serving everyone who touches the process of building digital products. That expansion was not planned. It was enabled by the architecture. When you make a product genuinely useful to people who weren't your original audience, and those people can access it freely, the market defines itself.

Technical depth taken seriously early pays compound interest. The WebGL renderer Wallace built in 2012 created a quality ceiling that competitors could not reach without rebuilding from scratch. When WebAssembly arrived years later, Figma adopted it in weeks because the codebase was already compatible. The hard infrastructure investment meant every subsequent improvement was cheaper.

The person who surfaces the problem and the person who can solve it technically are rarely the same. What Field and Wallace had was a complementary pair: one who understood what was broken from watching real teams work, one who understood what was newly possible in the browser. The founding insight required both halves. Neither had the full picture alone.

Coachability is a competitive advantage. Field was twenty years old, intensely high-standards, and not yet a competent manager. When his team told him he needed to change, he changed. When his investors told him he needed to start charging, he figured out how to do it without destroying the mechanism that was driving adoption. The ability to update under pressure, from people you trust, is what separates founders who scale from founders who plateau.

Explore Further

• Dylan Field's IPO founder letter — the founding vision in his own words, written for the moment of going public
• Evan Wallace on building a professional design tool on the web — the technical case for why the browser architecture worked, written by the person who built it
• How Figma's multiplayer technology works — Wallace's account of the real-time collaboration system, including why they rejected Google Docs' approach
• Multiplayer Editing in Figma (2016) — the original launch post for the feature that changed the category
• Five phases of Figma's community-led growth — First Round's account of how Claire Butler built the bottoms-up motion before the product even launched
• Sequoia's profile of Dylan Field — the best long-form character portrait, including the Flipboard summer and the launch reactions
• Index Ventures: thirteen years with Dylan Field — the full-arc account from seed round to IPO, including the board conversations about charging
• Dylan Field in conversation with Elad Gil — a candid 2023 interview covering Figma's origins, the Microsoft moment, and the product-market fit they almost denied themselves