HUPO 2026, Singapore. Workshop edition.
Most scientists who think about starting a company start with the idea, or the thing they built.
That was the theme that kept surfacing when Lindsay Pino and I ran our entrepreneurship workshop on day zero of HUPO in Singapore. It's the fourth time we've done this now, and every room is different. This one, about thirty people deep and mostly students, founders and bench scientists, kept circling back to one uncomfortable idea: the quality of your technology is not the thing that decides whether you have a business.
We run these under the Chatham House Rule so the panel can be candid, and we write up the learnings afterwards because the questions we get from this community far outnumber the people who can fit in one room. If you want the earlier conversations, they're here: Money, IP, and the people you let in from ASMS, and An entrepreneurial mindset doesn't require starting a company from HUPO Toronto.
Here is what came out of the Singapore room.
L to R: Paula Burton, Richard Shen, Vera Ignjatović, Yen Choo, Lindsay Pino.
Richard put the scientist's trap more plainly than anyone.
"As a scientist I was trained to think that if you have a really cool tool, and it's much better than everything out there, you should commercialize it. But a better tool doesn't mean everybody wants it." - Richard
His test is market size before anything else. If your better mousetrap serves five labs in the world, no amount of brilliance justifies the tens of millions it takes to build and sell it. If, on the other hand, the product serves a market measured in the billions, the math changes entirely. Identify the problem, and the size of the market, before you fall in love with your solution to it.
Yen said the same thing from the failure side.
"Most companies fail because they haven't got the right fit of technology to the market." - Yen
His advice was unglamorous and powerful: go out and talk to a lot of stakeholders, the people you think might actually be your customers, and find out whether there's a market there at all. If there isn't, understand why, and adjust.
Vera comes at it as someone who lives inside clinical workflows. Her instinct is to map the process before proposing a product.
"What is the product, and where does it fit within existing processes? What could it replace, what are the existing costs, and what would the cost be with your product in the workflow? Does it save time, does it need less people to run it?" - Vera
Between the three of them, that's the whole discipline. A tool is a feature. A problem, sized and understood and mapped against what people already do or need to do, is the start of a company.
If your technology is a platform with a dozen possible uses, the next trap is trying to chase after all of those uses. The panel was firm: pick where you attack first, and treat that choice as make-or-break.
"When you have a platform technology with a bunch of applications, the first thing to figure out is your beachhead market. What are you going to attack first? Spend a lot of time on that, because it can make or break your company. Even large pharma get this wrong." - Yen
Richard framed the opportunity in a different way, through what he called the zero-billion-dollar market, borrowing Jensen Huang's language. Investors love to hear about hundred-billion-dollar markets, but the real openings are often the markets that don't exist yet because nobody's made them affordable. He reached for his own world to show it: DNA sequencing went from roughly $250 million a genome to a couple of hundred dollars, and every drop in price unlocked uses that were unthinkable at the old cost. The big incumbents guard the markets they already dominate and miss the ones forming underneath them. That gap is where a startup gets its foothold.
The bridge between the two, the small proven market and the big one, is the part Richard called the valley of death. You can win a handful of early customers and still struggle to reach the next level of adoption. Knowing that chasm is coming, and planning for it, is its own skill.
This is the question every proteomics room asks, because the equipment costs a small fortune and expertise to setup. The panel's answers were refreshingly practical.
Yen's first move is to spend other people's capacity before your own. Translational grants and industry-collaboration schemes get you going without dilution. And a good incubator can hand you an entire instrument fleet.
"The incubator I run gives access to all the equipment in the university, including the Singapore Phenome Centre. In the early days of my own company in the UK, we had a shaker and a pipette, and that was difficult." - Yen
Richard mapped the capital options past that point: angel groups (make sure they have an actual fund to draw from rather than chasing individuals one at a time), and disease-focused foundations, like Curebound in San Diego, that will put $250,000 to $500,000 into a startup over two years if the idea lands in an area others have ignored. I shared our own first step at Mass Dynamics, a $50,000 cheque from an angel that felt like winning the lottery at the time, and what it actually means to give away equity that early.
But the sharpest point was that you might not need to incorporate at all yet. The company is a vehicle for raising capital, so if you're not at the point of raising, you can often go further first by stress-testing the technology academically, building funded collaborations, and widening your network before you ever register anything. Winding a company down is effort. Not starting one prematurely is free.
Vera came back to intellectual property more than once, because it's the thing that quietly kills deals.
"Where does the IP sit? Who owns the IP? People will not invest until that story is clear. It has to be crystal clear on paper and crystal clear as a story." - Vera
Her warning about collaborations was the one that made the room go quiet. Everything is friendly right up until someone decides to spin out, and suddenly ten people have their hands up claiming they contributed. Get the paperwork right before samples change hands, so there's no ambiguity about who owns what. Investors will walk away from a messy IP story rather than try to untangle it.
We talk about founders being married to each other, and it's true. You'll sign a shareholder agreement that's harder to exit than a lot of marriages. But Lindsay extended it in a way I hadn't heard put so well.
"My co-founder is called Alex and my husband is called Alex, so I literally have to ask myself, is this a home-Alex or a work-Alex question?" - Lindsay
On building the team itself, the panel converged on complementarity and advisors. Vera made the case for advisors over co-founders where you can manage it: a small group of formal or informal advisors plugs your skill gaps, adds credibility when you fundraise, and doesn't dilute your ownership the way another partner would.
The formal-education question split the room in a useful way. Vera did a global executive MBA and valued it, because she needed the language to talk to industry and couldn't speak it as a pure scientist. What works for a giant company doesn't map onto twenty people. You learn that at the coal face.
The through line was self-awareness. Know what you're good at, be honest about what you're not, and either hire for the gap part-time or bring in an advisor. Vera's point was that time is money, so decide deliberately whether a given skill is worth learning yourself or worth buying.
My favourite tactic of the day was Vera's, on getting in front of people who don't owe you their time.
"The advice I always give is to get fifteen minutes of someone's time and have them fire off their main problems at you. Don't even talk about yourself." - Vera
That's the whole workshop in one move. Fifteen minutes of listening to a problem beats an hour of pitching your tool or idea.
We closed by asking each person to bust a myth about entrepreneurship. A few landed hard.
One was that entrepreneurs are a special kind of person. They're not. Getting better at it works like getting better at anything, you do it, a little every day, whatever stage the idea is at.
Another was about going in with your eyes open. As I said in the room, by the time you're building this thing you've also got marriages, "babies" (aka ideas or beliefs that are hard to let go of), all sorts going on, and the workload is real. Nobody should start without knowing that.
And the one I come back to, because it ties to everything we've written before: an entrepreneurial mindset doesn't require starting a company. The spirit of it, finding a real problem and being relentless about solving it, lives in any role, in any organization.
A lovely reframe was that you don't decide to become an entrepreneur and then find a mission, it's the other way around. You believe in the thing so completely that there's no other path. And the reminder that failing isn't shameful, it's fast feedback. Pull the plug when you need to, learn why, go again.
Every one of these sessions, people walk in thinking the hard part is the science and the brilliance of what they've built. They walk out realizing the science is often the easy part. The hard part is choosing the right problem, sizing it honestly, protecting it cleanly, and surrounding yourself with people you can stay married to through all of it.
If you were in the room in Singapore, thank you, it was a good one. If you weren't, I hope some of this travels. To stay across other events like this, please join our 'Entreneurship in Mass Spec and Proteomics' community here.
This post was produced by Paula Burton from the session recording, with assistance from an AI writing tool. Each speaker reviewed their own section. No speaker reviewed the full piece end to end, and all synthesis and takeaways are my own. Any errors or omissions are unintentional. The views expressed belong to the author and not necessarily to the author's employer or any other organisation.