Solo founder or team?
Core idea#
Deeptech companies need more different kinds of work done early than most startups: research, engineering, product, regulation, sales, fundraising. One person can rarely do all of it well. That's the case for co-founders. The case against: the wrong co-founder is one of the most damaging mistakes a startup can make, and splitting equity early with someone who later leaves or underperforms is hard to undo.
So the real question isn't "solo or team?" in the abstract. It's:
- What does this company need in its first two years that I can't do or can't learn fast enough?
- Is there a specific person who covers that, who I trust and have worked with, and who is ready to commit?
- If not, can I cover the gap another way (early hires, advisers, a part-time co-founder, a partner company) without slowing the company down too much?
Solo is a legitimate choice, increasingly common, and sometimes right. But it should be a decision, not a default because you haven't found anyone.
Who it's for: researchers deciding whether to bring in co-founders before incorporating or raising, and founders reconsidering their team shape.
What the data says#
Solo founding is rising#
The share of new US startups on Carta with a solo founder rose from 23.7% in 2019 to 36.3% in the first half of 2025 (Carta Solo Founders Report). About 36% of startups founded in 2025 were solo-led, up from 31% in 2024, and the share has doubled over ten years (Carta Founder Ownership Report 2026).
But teams still raise more#
- Solo-led companies were 30% of startups founded in 2024 but received only 14.7% of the cash raised in priced equity rounds that year (Carta).
- In 2025, startups with two co-founders accounted for 36% of all rounds closed on Carta; solo-founded startups raised just 20% (Carta).
- Among VC-backed companies, solo founders are about 20% (Sierra Ventures, with Carta's Peter Walker).
Deeptech and biotech teams are larger#
In biotech, only 35% of new startups have one or two founders, and 37% have four or five (Carta Founder Ownership Report). Solo founders are under-represented in pharmaceuticals and biotech (Carta and Solo Founders, State of Solo Founding). Scienceheavy companies often have a scientific founder, a technical or engineering founder and a business founder.
Co-founders leave#
More than a quarter of two-founder teams lose a co-founder by year four (Sierra Ventures). That's why founder vesting (playbook 3.06) isn't optional, and why testing a co- founder relationship before committing matters.
What this means#
Solo is increasingly accepted, especially where software and AI tools let one person build a lot. In deeptech, where the work spans lab, factory and market, investors still tend to expect a team, or a clear plan for building one.
Deciding#
Solo can work when#
- You have, or can quickly gain, the commercial skills the next 18 months need (selling, fundraising, hiring).
- The technology can be developed by a small team you hire, rather than needing a peer-level technical partner.
- You have strong advisers and early hires lined up to cover gaps (playbook 1.12).
- An institution or programme provides structure (incubator, fellowship, accelerator).
- You'd rather keep control and accept a harder fundraise.
A co-founder is worth it when they bring#
- A missing capability you can't hire for yet: commercial leadership, regulatory expertise, manufacturing scale-up, or deep engineering that complements your science.
- Commitment equal to yours: full time, on the same timeline, sharing the risk.
- Trust built through shared work, not just friendship or a short conversation.
- Different strengths, shared values: you disagree productively on how, not on what or why.
If someone brings only one of these, they may be better as an early hire or adviser.
Common deeptech team shapes#
| Shape | Who | Works when | Watch out for |
|---|---|---|---|
| Scientist + business founder |
Inventor as CTO or CSO; commercial co-founder as CEO |
The technology needs a seller and fundraiser from day one |
Mismatched commitment; business founder not technical enough to sell |
| Scientist + engineer | Inventor plus someone who builds products |
The main risk is turning lab results into a working product |
No one owns customers and fundraising |
| Three founders: science, engineering, business |
Common in hardware and biotech |
Complex products with long timelines |
Equity split and decision-making get harder |
| Professor + postdoc or student |
PI as scientific founder; junior researcher full time |
University spinouts | Power imbalance; PI's time commitment (playbooks 2.12, 2.13) |
| Solo scientist with early hires |
Founder plus first employees with equity |
The founder can sell and raise |
Investor scepticism; founder overload |
When to decide#
Ideally before you incorporate and before significant equity is issued. Adding a cofounder later is possible, but they'll typically get less equity and may feel less ownership. If you're still searching, say so to investors and show your plan.
Finding and testing a co-founder#
Where to look#
| Source | Good for |
|---|---|
| People you've already worked with(lab colleagues, former industry teammates) |
Trust is already tested |
| Customer discovery interviews | Industry people who keep engaging may want to join |
| Business school and executive programmes (MBA students, alumni) |
Commercial co-founders interested in deeptech |
| Entrepreneurs First and similar programmes(1.10) |
Structured co-founder matching |
| Incubators and fellowships(1.10, 1.11) | Often run matching events |
| Tech transfer offices and spinout programmes |
Some keep lists of experienced CEOs for spinouts (3.04) |
| Founder-matching platforms | Wide reach, but trust must be built from scratch |
Test before you commit#
- Have the hard conversations early: roles, titles, time commitment, salary expectations, equity split principles, what happens if one of you leaves, what you'd do with an acquisition offer, values and red lines.
- Work together on something real for 4 to 8 weeks: customer interviews, a grant application, a pitch, a technical plan. Watch how you handle disagreement and deadlines.
- Meet each other's people: partners, former colleagues, references.
- Agree vesting before any equity is issued, with a one-year cliff (playbook 3.06). Vesting is what makes it safe to try.
Worked example (illustrative)#
Dr. Ana Ruiz is developing a new sensor for detecting leaks in hydrogen pipelines. She's strong technically and has run lab budgets, but has never sold to energy companies.
- Search: she met eight potential co-founders through her customer interviews, a business-school alumni network and an incubator matching event.
- Shortlist: two candidates, both with energy-industry sales experience.
- Trial: each joined her for six weeks of customer interviews and a grant application. One was brilliant in meetings but missed deadlines; the other was quieter but reliable, and brought two utilities into pilot talks.
- Decision: the second became co-founder and CEO, with a 50/50 split after a structured discussion, four-year vesting and a one-year cliff, documented in a founders' agreement before incorporation.
India, the US and the UK#
- India. Co-founders from the same institute (IIT batchmates, lab colleagues) are common and bring trust, but often similar skills; a commercial co-founder with industry relationships is valuable, especially for selling to large companies or government. Under DPIIT rules, promoters generally can't receive ESOPs except within the startup window (playbook 5.01), so co-founders' equity should be founder shares, set up early.
- US. Investors are most familiar with two-founder teams and expect vesting. Solo founders do raise, particularly through accelerators that accept them.
- UK. University spinouts often start with academic founders and add a commercial CEO later, sometimes recruited by the university or investors (playbook 3.04). Decide early whether you want that path or would rather find your own commercial cofounder.
Scripts#
Opening the co-founder conversation#
"I've enjoyed working with you on [project]. I'm planning to start a company around [technology], and I think you'd be the right person to lead [area]. Before either of us commits, could we spend six weeks working on it together, and have the hard conversations about roles, equity and time early?"
Raising the difficult topics#
"Let's each answer these separately and then compare: What role do you want in three years? How many hours can you give from when? What salary do you need, and when? What would make you leave? If we got an acquisition offer in year two, what would you want to do?"
Explaining a solo decision to investors#
"I'm the only founder today, deliberately. I've covered [gap] with [adviser / hire], and our first senior hire will be [role], with meaningful equity. I'd rather build the right team than add a co-founder for the slide."
Common mistakes#
- Choosing a friend because they're available. Friendship isn't the same as complementary skills and commitment.
- Splitting equity before testing the relationship.
- No vesting. If a co-founder leaves early, they keep unearned equity.
- Two people with the same skills. Two scientists rarely cover selling and fundraising. A part-time "co-founder" with full-time equity. Match equity to commitment.
- Waiting indefinitely for the perfect co-founder. Set a deadline; if you haven't found one, build the team through hires and advisers.
- Avoiding hard conversations. They only get harder after money arrives.
Related playbooks#
- 1.09 Founder-market fit
- 1.12 Finding mentors and early advisers
- 2.12 Professor as co-founder, adviser or CSO
- 2.13 Students and postdocs as founders
- 3.04 Bringing in a business CEO
- 3.05 Splitting founder equity
- 3.06 Founder vesting, cliffs and acceleration
Sources#
- Carta, Solo Founders Report 2025; Founder Ownership Report 2026; Founder Ownership Report; Two-founder teams.
- Sierra Ventures, What the data says about pre-seed, with Peter Walker.
- Noam Wasserman, The Founder's Dilemmas (2012), on co-founder choice and equity.
Worked example is illustrative. Last checked November 2026.