Forget Everything You Know About Startup Incubators. Harvard Got it Right.

You've heard of incubators and accelerators. They're everywhere now. Some entrepreneurial-minded person somewhere is always launching the next revolutionary startup program. They'll get some venture money, rent a cool office space, maybe hire an ex-startup founder to run the program, and declare victory when three companies hit Series A.

Then, ten years later, most of those companies are dead. The program is forgotten. The space gets repurposed.

Here's the depressing truth: Most startup support infrastructure is designed to look good in press releases, not to actually build successful companies. We've mistaken the infrastructure of entrepreneurship—the ping-pong tables, the networking events, the motivational talks—with the actual work of building something that scales.

But Harvard Innovation Labs is different.

I recently talked with Michael Fenn, who runs the health and life sciences division at Harvard's innovation program. And what struck me most wasn't his title or his impressive résumé (though he's got a PhD in biomedical engineering and a track record of building companies that actually exit successfully). What struck me was how methodically, almost boringly, they've approached the problem of helping founders build successful ventures. There's nothing sexy about it. And that's exactly why it works.

The Network Matters More Than You Think

Here's what most people get wrong about startup incubators: They think the value is in the facility, the mentors, or the capital. They build a space, hire some smart people, and assume magic happens.

Michael's philosophy is radically different. When we talked about what makes the difference between a failed venture and a successful one, he didn't point to the office or the funding. He said, again and again: "It always comes back to the people."

Not in a motivational-poster kind of way. In a very specific, practical way.

When a founder walks in the door with an idea, Michael's first question isn't "What's your business model?" or "What's the market size?" It's "Do you have the right people?" And if they don't, "Can we connect you with them?"

That sounds simple. But it reveals something fundamental about how innovation actually works. A great idea with the wrong team gets nowhere. An okay idea with the right team can become something extraordinary. And the most valuable thing an incubator can provide isn't office space—it's access to the right people at the right time.

Harvard's got that in spades. They have a network of faculty advisors, investors, technical experts, domain specialists, and successful entrepreneurs. These aren't people who were hired to mentor startups. They're people who were already there, doing their work, and Harvard simply figured out how to connect them to founders who needed them. The philosophy is almost boring in its clarity: play connector. Know your network. Understand what each person brings. Then make intelligent introductions. Michael has learned that his job is less about having the right answers and more about knowing who does.

The result? Better companies. Companies that solve real problems rather than pitch investor fantasies.

The Unmet Need Obsession

Most startup founders are in love with their solution. They wake up thinking about their technology. They shower thinking about it. They probably talk about it to their spouse or partner and get that look that says, "I love you, but please stop talking about the product."

Michael's approach flips this completely. When evaluating whether to support a venture, Harvard asks: Does this solve an actual unmet need? Is it a problem that real people are experiencing right now? And critically—will they actually adopt the solution you're building, or will they keep using whatever they're using now because the switching costs are too high?

This sounds obvious. It's not. In life sciences, particularly, you see founders who've invented something brilliant in the lab. It's technically sophisticated. It's elegant. It works. But it solves a problem nobody's really suffering from. Or worse, it solves a problem in an expensive way when cheaper alternatives already exist. The founder has fallen in love with the solution and is now trying to find a problem for it. It never works.

Michael walks founders through a brutal logic: Who is the customer? What is their actual problem? And critically—would they pay you to solve it? How much better is your solution than what they're currently doing?

That last question matters more than you think. Because switching costs are real. People are lazy. They're comfortable with the status quo. Inertia is a powerful force. If your solution is only marginally better, they won't switch, no matter how elegant it is. They'll keep using what they know, even if it's worse, because at least it's familiar.

This is why Harvard focuses on ventures that solve "significant unmet needs." Not marginal needs. Not nice-to-haves. Needs where the current standard of care is failing and people are desperate for something better. The pain is real. The motivation to switch is there.

Then comes the adoption question. A brilliant solution that nobody adopts is worthless. So Michael pushes founders to think about: Who are you trying to convince? What's their daily workflow? How does your solution fit into their life? What has to be true for them to choose you over what they're already doing?

It's unglamorous. It's not the stuff of TED talks. But it's the stuff that separates ventures that scale from ventures that quietly fail, burning through investor capital while their founders pretend everything is fine.

The Commercial Reality Check

Here's where a lot of academic spinoffs die: They're run by brilliant scientists who've never had to think like a businessperson. They don't understand intellectual property strategy. They don't know how to navigate regulatory landscapes. They think a great technology will simply speak for itself and the market will come.

The market never comes. Not on its own. The technology doesn't sell itself. And the road to commercialization is more complicated than they imagined.

Michael ensures every venture has what he calls a "viable commercialization plan." Not a nice-to-have. Not something to figure out later when you have more time. A plan that addresses: How will this get to market? What regulatory pathways exist? What does the competitive landscape look like? Who are the people you need to bring on board to make this happen? What's the path from today to a product someone will actually buy?

And then there's intellectual property. In biotech and medical devices, IP is foundational. You either have strong IP protection, or you're vulnerable to being copied the moment you find product-market fit. Your competitors will eat your lunch.

So before Harvard invests significant time and resources in a venture, they assess: Do you have the right patent strategy? Do you need to license technology from someone else? Do you understand what you own and what you don't? What are your IP risks?

Again—this is not sexy. Nobody gets excited about IP strategy. Nobody puts it on their Instagram. But it's the difference between building a defensible company and building something someone else will steal the moment it starts working.

Why Top-Down Support Actually Matters

There's a structural lesson buried in how Harvard organized its innovation program. The labs report directly to Harvard leadership—the people with real power and real budget authority. Not to HR. Not buried in some administrative function that gets cut the moment the budget gets tight.

This matters more than most people realize. Because when things get hard—and they always do—innovation initiatives without executive protection get sacrificed on the altar of quarterly earnings. The budget gets reallocated. The team gets dispersed. The whole thing gets dismantled. Some consultant writes a memo about "organizational realignment" and suddenly your entire innovation program is gone.

But if your innovation program has protection from the top, if the person running the whole organization has made a commitment to it, it survives. It thrives. It becomes embedded in how the organization operates.

Michael was blunt about this: "Don't attempt it from the middle. Go straight to your chairman."

This isn't political maneuvering. It's survival. It's understanding how organizations actually work—that the people at the top control resources, and if they don't care about your thing, your thing will die when money gets tight.

The Uncomfortable Realization About Knowledge

As I was talking with Michael, something he said stuck with me. I asked how they preserve knowledge—how they make sure all the learning that happens in these ventures doesn't just evaporate the moment the founder moves on to the next company or the next challenge.

His answer was surprising: They don't really try.

Instead, they put knowledge into the flow of work. They teach people in the context of real projects. They build networks, not databases. They create people who know how to learn, who know who to ask, and who can adapt quickly when everything changes (which it always does).

And here's the uncomfortable part: Most knowledge isn't worth preserving. What matters is building "exponential learners"—people who get better at getting better. People who can look at a changing landscape and figure out what to do without needing a manual. People who know that what's true today might not be true tomorrow.

This flies in the face of how most organizations approach knowledge management. You try to capture it. Systematize it. Put it in databases and wikis and internal knowledge bases that nobody will ever read because they're either out of date or buried in corporate jargon. You've created the infrastructure of knowledge without the actual knowledge flowing through it.

But in a world that's changing faster than we can plan for, that approach doesn't work. You need people who are comfortable with uncertainty. Who can learn from what's happening around them. Who can teach each other. Who understand that the network of smart people you have access to is worth more than any document.

Why This Actually Matters (Even If You Never Run a University)

You're probably not going to build a corporate academy tomorrow. Most people won't. Most people don't have the resources or the organizational authority. But the principles Harvard has figured out apply to any organization trying to scale innovation—whether you're a startup trying to grow, a corporation trying to stay relevant, or a nonprofit trying to have more impact:

Start with the right people, not the right process. Most organizational problems are actually people problems. Before you implement a new framework or process or methodology, ask: Do we have the right people? If not, can we find them? Can we attract them? Can we give them what they need to do their best work?

Obsess over unmet needs, not clever solutions. Don't fall in love with your technology. Fall in love with the problem it solves. Make sure that problem is real, significant, and something people will actually pay to solve. Make sure they'll switch from what they're currently doing.

Understand the commercial reality. A brilliant innovation that can't get to market is just a lab experiment. Understand the regulatory landscape, the competitive landscape, the intellectual property landscape, the customer adoption dynamics. Boring as hell. Also absolutely essential.

Protect your innovation program with executive support. If you're launching something new inside an organization, make sure it has protection from above. Make sure the people who control resources care about it. Otherwise it'll get killed the moment growth slows and budgets tighten and someone needs to cut costs.

Build networks, not databases. Stop trying to capture and preserve all knowledge. Instead, build networks of people who know how to find information, who know who to ask, and who can teach each other. That's how you actually scale learning across an organization.

The Real Challenge

The reason most startup programs fail isn't that they lack good intentions. It's that they misunderstand what actually creates successful ventures. They think it's about inspiration and access and capital. And those things matter. But they're not the primary thing.

The real work is harder. It's about connecting the right people to the right opportunities. It's about helping founders see the actual market, not the fantasy version they've constructed in their head. It's about making sure they understand the regulations and the IP landscape and the commercialization pathway. It's about asking the uncomfortable questions that force them to confront whether they're solving a real problem or just in love with their technology.

It's boring. It's methodical. It requires real expertise and real networks and real willingness to tell people things they don't want to hear.

Which is why most people don't do it. And which is exactly why Harvard, by doing this unglamorous work consistently, is building something genuinely different. They're building companies that actually work, not just companies that have good pitch decks.

---

Want the Full Story?

If this resonates, listen to Yaniv Corem's full conversation with Michael Fenn on The School of Innovation podcast. They go deep into the life sciences landscape, the emerging technologies that are actually going to matter, and what actually makes the difference between a venture that scales and one that quietly dies while everyone pretends it's just going through a challenging phase.

Because here's the thing about startup support: Everyone offers it. Very few people actually understand what it takes to make it work. Even fewer are willing to do the unglamorous work required to make it happen at scale.

The question is: Are you one of those people willing to put in that work?