Solo Founder vs Co-Founder Team: What the Data Says About Who Raises Venture Capital

Solo founder versus co-founder team is the oldest argument in early-stage investing, and as of October 2026 the data finally allows a more precise answer than ‘VCs prefer teams’. In this article the Kurums Finance team sets out what share of venture-backed companies are solo-founded, how that share has moved, what the peer-reviewed research says about solo-founder outcomes, which famous companies were started alone and which by pairs, why investors still hesitate, when going solo works, how solo founders de-risk their raise, and how the 2026 ‘one-person unicorn’ debate is changing partner attitudes. It belongs to the Venture Capital finance hub on kurums.com and pairs with our data piece on what venture-backed founders have in common and the practical guide to the founding team investors want.

What share of venture-backed startups are solo-founded?

Roughly one in six to one in five. Carta’s Founder Ownership Report (45,000+ startups incorporated 2015–2024) found 35% of 2024 startups were solo-founded but only 17% of those that raised venture capital in their first year; YC put its own solo-founder share at nearly one in five in 2026.

The Carta figure is the most useful because it measures the same cohort at two stages. Solo founders have more than doubled as a share of new incorporations over the decade, driven by lower costs of building software, no-code and AI tooling, and a cultural shift away from the ‘you must have a co-founder’ orthodoxy. But at the point of the first venture round the solo share falls by half, which is the clearest quantitative statement of investor preference available. The gap is a selection effect in both directions: some solo founders never intend to raise, and some investors will not fund one.

At the unicorn end, PrivateCircle Research’s 2024 study of 113 Indian unicorns found 22% of those founded in the last decade had a single founder, with fintech the solo-friendly sector (CRED, Slice, Go Digit and Acko were all started by one person). Ali Tamaseb’s Super Founders data for US unicorns founded 2005–2018 found roughly one in five had a solo founder, and well-known solo-founded companies include Amazon, Dell, eBay, Oracle, Tumblr, Craigslist, Plenty of Fish, Mailchimp and, at the AI end of the spectrum, several of the 2024–2026 application-layer breakouts. Teams of two or three remain the norm, but ‘solo founders do not build big companies’ is simply false.

The direction of travel is toward smaller teams. YC’s 2026 commentary tied the rise of solo founders to two trends: AI makes it possible for one person to build and ship software that previously needed three, and the current batch mix includes more hardware and ‘atoms’ companies where a single deep-domain founder plus capital can be the right structure. Carta’s data shows teams of three, four and five founders have all become less common since 2015.

What does the research say about solo founders versus teams?

The best-known study, by Jason Greenberg (NYU Stern) and Ethan Mollick (Wharton), found solo-founded for-profit ventures were about 2.5 times as likely to survive as team-founded ones and generated more revenue, the opposite of what investor behaviour implies.

Greenberg and Mollick tracked ventures that launched Kickstarter campaigns between 2009 and 2015 and were later formally established, following them for several years; the sample collectively raised about $151 million in crowdfunding and generated about $358 million in revenue. Solo founders were 55% less likely to dissolve than teams of three. For nonprofits the result reversed, with teams about twice as likely to survive, which the authors read as evidence that the value of a team depends on the task: coordination costs are real, and in a small commercial venture a single decision-maker can move faster than a committee. The authors explicitly suggested that investors who avoid solo-founded projects ‘may want to reconsider’.

The study has limits that matter for a venture reader. Kickstarter ventures are small, mostly consumer, and not selected for venture-scale growth; survival and revenue are not the same as the 50x outcomes a VC fund needs. On the other side of the ledger, PrivateCircle found co-founded Indian unicorns reported average revenue about 32% higher than solo-founded ones (Rs 2,909 crore vs Rs 2,196 crore), though six solo-led unicorns had already listed against seven team-led ones, hardly a rout. First Round Capital’s 10 Year Project found solo founders in its portfolio underperformed teams of two or more, but that is a dataset of companies that passed a team-preferring filter in the first place.

The honest synthesis: there is no robust evidence that solo founders produce worse outcomes once funded, some evidence that they do better at the small end, and clear evidence that they are funded less often. That combination is what economists call a market inefficiency, and a handful of funds have built theses on it.

Different populations, different success definitions. None of these datasets supports a flat rule that solo founders cannot raise or cannot win.

Which famous companies were started solo, and which by teams?

Amazon, Dell, Oracle, eBay and Mailchimp were effectively solo-founded; Google, Airbnb, Stripe, Microsoft and Apple were founded by pairs or trios. Both lists contain generational outcomes, which is why serious investors talk about team dynamics rather than headcount.

Jeff Bezos started Amazon alone in 1994 with family money and hired engineers rather than recruiting a co-founder; Michael Dell built Dell from a dorm room; Larry Ellison had co-founders at Oracle but was the single driving force from day one. Pierre Omidyar wrote eBay’s first code alone over a long weekend. Ben Chestnut and Dan Kurzius built Mailchimp as a pair but without outside capital for twenty years, an example that cuts both ways. On the team side, Larry Page and Sergey Brin, Brian Chesky, Joe Gebbia and Nathan Blecharczyk, Patrick and John Collison, Bill Gates and Paul Allen, and Steve Jobs and Steve Wozniak are the canonical pairs, and in each case the co-founders had worked, studied or lived together before starting the company.

That last detail is the lesson investors draw: the question is not ‘one or two?’ but ‘how well do the people in the room know each other, and does the combination cover the two or three skills the first eighteen months require?’. A pair that met at a co-founder matching event three months ago carries more risk than a solo founder with a battle-tested first engineer. Our article on roles, equity splits and collaboration patterns that raise goes deeper on what ‘working history’ means to a partner.

Why do investors still hesitate to fund solo founders?

Four reasons recur in partner meetings: key-person risk, workload, blind spots, and the hiring signal. None is irrational; all are manageable.

Key-person risk (‘bus factor’)

A venture fund is underwriting a seven-to-ten-year hold. If the only person who understands the product, the customers and the cap table is unavailable for a month, a solo-founded company can stall in a way a two-founder company does not. Funds also know that burnout is common: the founder is CEO, head of product, head of sales and head of fundraising simultaneously.

Workload during the raise

A fundraise takes three to six months of the CEO’s time, and in a team one co-founder keeps the company growing while the other raises. A solo founder’s metrics often flatten during the round, which investors then read as weak traction. The realistic fundraising timeline explains how to compress that period.

Blind spots and no counterweight

Partners value a co-founder who can tell the CEO they are wrong. Without one, the board becomes the only check, and at seed the board is often just the founder. Investors worry about strategic drift, about a non-technical founder outsourcing the core product, and about a technical founder never learning to sell.

The hiring signal

The unspoken question is ‘if nobody would co-found this with you, why would a strong VP join?’. It is a crude proxy, but it is the one partners use most. The answer has to be visible in the team slide: senior early hires, named advisors, and customers or investors who have already committed.

When does going solo actually work?

Solo works best for second-time founders, for founders with deep domain insight who can hire the rest, for capital-efficient software businesses, and increasingly for AI-native companies where one person plus agents can ship what a team of four shipped in 2020.

Second-time founders are the easiest case. A founder who has already built a team, raised a priced round and sold or scaled a company has demonstrated the skills a co-founder would otherwise vouch for, and investors treat the repeat-founder premium as a substitute for a partner. The Carta data on repeat founders capturing more than half of 2025 Seed and Series A dollars includes a disproportionate number of solo repeat founders raising large seeds on a thesis and a hiring plan.

Deep-domain founders are the second case. A surgeon building a surgical-robotics company, a former central-bank regulator building compliance software, or a semiconductor process engineer building an inspection tool does not need a co-founder to prove founder-market fit; they need a strong first engineering hire and a commercial advisor. Fintech’s over-representation among solo-founded Indian unicorns (40% of the solo group per PrivateCircle) fits this pattern: Kunal Shah at CRED and Varun Dua at Acko came in with years of category-specific insight.

The third case is the one changing fastest. By late 2026 it is routine for a single founder to ship a production SaaS product, run customer support and marketing with AI tooling, and reach meaningful revenue before raising. YC’s own batches now feature solo founders at close to the 20% mark, and the ones that get in typically arrive with revenue rather than a prototype. Investors who insisted on teams because ‘one person cannot build and sell at the same time’ have had to update, because the premise is weakening. See our analysis of where venture capital went in 2024–2026 for the AI-share context.

How can a solo founder de-risk the raise?

By answering the four investor concerns before they are asked: name the people who would run the company if you could not, show senior hires or committed advisors, document how decisions get challenged, and raise on traction rather than on a prototype.

Start with the succession answer. A one-paragraph ‘if I am hit by a bus’ plan naming your first engineer, your most involved angel and your lawyer sounds morbid but closes the key-person objection faster than any argument. Second, make the team slide heavier than the headcount: a fractional CFO, a part-time head of growth with a recognisable background, and two or three advisors who hold small vested equity grants (0.25%–1% on two-year vesting is typical) show that experienced people have already bet on you. Third, build the counterweight: a monthly advisory call with notes, a seed board with one independent, and a written decision log give investors the ‘someone can tell the CEO no’ comfort they would otherwise get from a co-founder.

Fourth, and most important, raise later and on numbers. Because the solo discount is largest at the pre-seed stage where investors are buying only the team, it shrinks fastest with evidence. Ten paying customers, a retention curve and a clear sales motion change the conversation from ‘can one person do this?’ to ‘this person is already doing it’. Open-application routes are disproportionately useful here because they judge the form, not the intro: see our list of VC firms that accept open applications in 2026 and the guide to YC, Techstars and Antler application requirements, noting that Antler and Entrepreneur First were built specifically to match solo founders with co-founders before investing.

What is the 2026 ‘one-person unicorn’ debate about?

Whether AI tooling has made it possible for a single founder to build a billion-dollar company without a large team, and what that means for fund models built on backing teams. As of October 2026 no fully solo, near-zero-headcount unicorn has been confirmed, but several solo-founded AI companies have crossed $100 million valuations with fewer than ten employees.

The debate was framed publicly by OpenAI’s Sam Altman in 2024, who predicted a one-person billion-dollar company, and has been fuelled since by a run of tiny-team AI application businesses reaching tens of millions in annual recurring revenue. The investor response has split. Some seed funds now explicitly welcome solo technical founders and have shortened their diligence to a weekend; YC’s acknowledgement that nearly 20% of its companies are solo-founded is the clearest institutional signal. Other partners argue that the companies in question are ‘small and profitable’, not venture-scale, and that the hard parts of scaling (enterprise sales, compliance, international expansion, management) still require people. Garry Tan’s July 2026 remark that seasoned founders are ageing like fine wine was partly a counter to the idea that the AI era belongs to 22-year-old solo coders.

For a founder deciding in late 2026, the practical reading is that the solo discount is narrowing but has not disappeared, and that it is narrowing fastest for technical founders with revenue in software categories. A non-technical solo founder in a hardware or regulated category should still expect to be asked about the team in every meeting. The barbell effect in 2026 funding is relevant: the seed end of the market has become more experimental about who it backs, while growth-stage funds remain conservative about management depth.

Solo or team: how should you decide?

Decide on the work, not on investor preference. Map the three skills the first eighteen months require, be honest about which you have, and add a co-founder only if a specific person you have worked with before covers the gap and you would choose them even if no investor cared.

A simple test partners use is the ‘three jobs’ question: who builds the product, who sells it, and who runs the company? If one person credibly covers two of the three and can hire the third, solo is defensible. If the founder covers one, the company needs a partner or a very early senior hire. The second test is the working-history test. Co-founders who have shipped something together, even a side project, resolve conflicts faster; co-founders assembled through a matching platform without a trial period are the population Wasserman’s conflict data describes.

Whichever structure you choose, the fundraising mechanics are the same: build the investor list, work the warm-introduction path and the open-application route in parallel, and keep metrics growing during the raise. The structure of your founding team is one input into a decision that partners make on dozens; it is rarely the decisive one once there is evidence on the table.

Frequently Asked Questions

Browse the full Who Gets Funded pillar for the companion pieces on founder backgrounds that raise and the common traits of funded founders.