Kevin Ryan on Spotting 10-Year Trends

Kevin Ryan's whiteboard has twenty items on it, and none of them are companies. They are trends: gig work, automation in restaurants, healthy eating, the slow bleed of unstructured data into every database on earth. The exercise, which Ryan has described in interviews and repeats often, is deceptively simple. List twenty things you believe will still be true in ten years. Then ask what has to exist to serve that future but doesn't exist yet (03:57). Most founders skip that second step. Ryan built a career on not skipping it.

The ten-year window is not arbitrary. Ryan, who ran DoubleClick before founding the venture firm AlleyCorp in 2007, argues that anything worth building simply takes that long, so chasing a trend that's already fashionable means arriving after the money has been made.

"If it's something that is hypey today and is gonna be done in two years or started five years ago, you're too late." — Kevin Ryan, [04:43]

He has been right often enough to matter. Internet advertising became DoubleClick, which sold for $1.1 billion (02:06). Unstructured data, information that doesn't fit neatly into rows and columns, became MongoDB, a database company that took a decade to go public. Psychedelic medicine became Transcend Therapeutics. Nuclear energy became a stake in a company called Valar Atomics. But Ryan is just as candid about the trend he saw and didn't act on. In 2003, his team at DoubleClick noticed that the cost of serving video online was falling while ad rates were rising, and the two lines were about to cross. They didn't build the company. Two years later, someone else did, and called it YouTube. Ryan now values the outcome, Google's acquisition of YouTube, at roughly $300 billion (13:59). The lesson he draws isn't just about picking the right trend. It's about pushing one step further and asking who benefits when a trend plays out, what he calls second-order thinking: if the picks-and-shovels sellers around a boom are doing well, what does that imply about behavior a layer downstream.

The Discipline of Building Narrow

AlleyCorp doesn't run like a typical incubator. In the early years, Ryan and his cofounder Dwight Merriman, a technical partner Ryan credits as the brains behind both DoubleClick's and MongoDB's engineering, would each put in around $500,000, act as founders themselves, and spend about a year actually building and launching a product before raising outside venture money (28:24). That structure forced a discipline that shows up again and again in Ryan's case studies. Business Insider started with three journalists covering only New York tech, not national news, not Wall Street, just one narrow beat, and only expanded once that beat was working (30:59). Gilt Groupe, the flash-sale retailer Ryan cofounded, started with one sale a week of women's clothing before adding men's, then kids, then travel, then home.

"It's better to do one thing really well than a bad job on everything." — Kevin Ryan, [31:52]

Recruiting into that model required convincing talented people of exactly two things: that the idea was good, and that AlleyCorp could genuinely help them execute it (32:57). Ryan says he never plans an exit or lines up investors in advance. He describes recognizing a viable idea the way some people describe falling in love: it becomes what he calls a "business crush," an idea he can't stop elaborating on in his head, and if that feeling survives two weeks, he commits (39:21). MongoDB tested the patience behind that commitment. The company generated no revenue for three and a half years, and Ryan compares an unproven database to a pacemaker: nobody wants to be the first patient, so early customers have to be won for free before paying ones arrive. It took roughly a decade to reach a public offering; the company is worth about $30 billion today.

That same instinct for durable trends over hot ones now points AlleyCorp toward what Ryan calls deep tech, meaning capital-intensive, hard-science ventures like energy and robotics rather than software alone. Three years ago, before nuclear power was fashionable in Washington, AlleyCorp backed a small modular reactor company called Valar Atomics at a $20 million valuation. It has since raised money from the venture firm Sequoia at a $6 billion valuation (42:11). Ryan credits part of that swing to a policy shift: the Trump administration, he says, has moved both nuclear energy and psychedelic medicine forward faster than he expected, even as he disagrees with it on plenty else. AlleyCorp's first venture fund is now tracking a roughly 60% internal rate of return, a measure of how quickly an investment compounds over time (45:00), and its new $335 million fund, announced in 2026, keeps the same early-stage focus rather than scaling into a larger, fee-driven asset manager (03:26).

A Softer Molecule, A Different Kind of Company

Ryan's most personal bet sits furthest from his background in advertising and databases. At 54, having never tried a psychedelic himself, he read journalist Michael Pollan's 2017 book "How to Change Your Mind" and says it reordered his sense of what was possible in mental health treatment (51:51). He began funding research at Yale's Center for Psychedelic Research, then in 2021 concluded that the next phase needed to be a company, not a nonprofit, because getting a single compound through the full Food and Drug Administration approval process costs roughly $200 million to $250 million, more than donor funding alone could sustain (53:46).

Working with Yale researcher Ben Komenda, Ryan's team landed on methylone, a molecule with a gentler profile than MDMA, the drug best known for its role in PTSD trials. In Transcend Therapeutics' trials, patients could take methylone weekly, while MDMA is generally limited to about four uses a year before it loses effect and depletes serotonin more severely (55:33). Methylone also produces less of a post-session low, and its shorter duration of action matters for scaling any treatment into ordinary healthcare settings, where longer sessions mean higher staffing costs (56:15). The company built its case around post-traumatic stress disorder in part for political reasons: nobody, Ryan notes, wants to be seen denying help to veterans (59:56). Its actual trial population skewed 60% women, since sexual assault, not combat, is the more common cause of PTSD.

Transcend was structured as a public benefit corporation, a legal status that binds a company to a stated social mission alongside profit, with a twist Ryan says is rare: original shareholders committed to give 10% of their financial gains to a foundation supporting psychedelic causes, a pledge now worth about $20 million in planned giving over the next several months (60:17). Ryan says he has been disappointed watching wealthy people whose own lives were changed by these medicines decline to support the field that made the treatment possible.

Asked what message he'd put on a billboard seen by billions, Ryan didn't pick a business slogan. He pointed to income inequality and immigration policy, arguing that restricting high-skilled workers, such as PhDs in artificial intelligence, from coming to the United States pushes the next generation of founders and jobs elsewhere (78:16). He also warned that AI will displace jobs faster than the country is prepared to help people retrain, and that unaddressed inequality pushes desperate families toward political extremes (76:01). It's the same forecasting habit that filled his whiteboard, aimed for once not at a market, but at the country that made his career possible.