Skin in the Game: If You Don't Share the Risk, You're Only Advising

There's a simple question that tells you more about an innovation effort than any roadmap or pitch deck: what happens to the partner if the venture fails?

With a classic consultancy, nothing happens. The fee is paid, the contract fulfilled, the team at the next client. With an agency it's the same: the hours are billed, the product delivered, the risk sits entirely with the client. The partner delivered what was agreed and carries no share in the outcome.

This asymmetry isn't morally wrong. It's structural. But it explains much of why so much innovation work doesn't survive first contact with reality.

If you don't share the risk, you're only giving advice.

The principal-agent problem in one sentence

Whoever is paid for hours is optimised for hours. Whoever is paid for outcomes is optimised for outcomes. This isn't a question of character but of incentive. And incentives are stronger than intentions.

A classic time-and-materials model rewards activity, not success. The longer the project runs, the more is earned. The incentive to reach a viable business quickly, even uncomfortably, is weak at best. Since Michael Jensen and William Meckling in 1976, economists have described this as the principal-agent problem: whoever doesn't bear the consequences of their decisions decides differently from whoever does. The client wants an outcome, the contractor a fully booked contract. Those two goals aren't the same.

Crucially, this holds for the most capable and most honest consultants too. It's not an accusation against people. It's a property of the construction. Good people in a bad incentive structure produce bad results, and they do so reliably.

What the data says

You don't have to take this on faith. You can measure it, and economists have done so in places you'd hardly expect.

Steven Levitt and Chad Syverson analysed ten years of home sales in Illinois. When realtors sell their own homes, they get 3.7 percent more on average and take 9.5 days longer than with their clients' homes. On a client's home they only receive a small share of the extra price but bear the full extra work. So they sell faster.

Medicine is no different. Erica Johnson and Marit Rehavi showed that physician mothers are around ten percent less likely to have a C-section than other mothers. And when Medicare shifted its payment rates in 1997, the number of treatments rose where they paid better: two percent higher fees, around three percent more care, according to Jeffrey Clemens and Joshua Gottlieb.

Same person, same knowledge, a different incentive. And already a different outcome.

On the other side, research shows what happens when the stakeholder is close to the work. Shai Bernstein, Xavier Giroud and Richard Townsend used an elegant natural experiment: when a new direct flight cuts the travel time between a lead investor and its portfolio company, patents, citations and the likelihood of an IPO or acquisition rise afterwards. More proximity, better outcomes. Measured causally, not just claimed.

Where ventures actually die

That fits with how ventures fail. CB Insights analysed 431 shutdowns of VC-backed companies since 2023. In 70 percent, the money had run out in the end. Underneath are the real causes: poor product-market fit in 43 percent, bad timing in 29, unsustainable unit economics in 19.

Rarely the technology, almost always the commercial side. No market. No product-market fit. No viable unit economics. Exhausted capital. These are exactly the things a departed consultant never had to answer for.

The pattern behind it

Look more closely and you'll find this pattern almost everywhere technology fails to become a company. The consultant earns on hours, not on the company. The investor wants a late, safe entry; the builder carries the early risk. The corporate wants control; the venture needs freedom. The inventor wants recognition; the market wants a product.

Whether it's the strategy that falls apart on its way into execution, the technology that stays in the drawer, or the research result that never sees a cap table: there's almost always a place where someone is rewarded for something other than the success of the whole. More on that in Innovation Theater, Stranded IP and Between the Lab and the Cap Table.

That's the uncomfortable insight: the missing system isn't primarily a shortage of money, talent or technology. It's a shortage of aligned incentives.

The solution isn't trust. It's structure.

You could try to solve the problem with good intentions: partners who mean well, contracts that urge collaboration. But good intentions aren't a business model, and exhortations don't change incentives. What aligns incentives is a single thing: a stake in the outcome.

When the partner only wins if the venture wins, everything changes, without the person having to change. The question in their head is no longer "how do I fill the contract?" but "how does this become a company?" Same person, same skill, a different compass.

Sharing the risk isn't a legal concept here. It means economic risk: equity in the venture, your own capital in play, and work that runs without margin and only pays off if the business carries. That's less comfortable than a consulting fee. But that discomfort is precisely the point.

A guardrail: a label guarantees nothing

Now the objection, and it's a fair one. Venture studios and venture builders have for years advertised figures suggesting their ventures do far better than classic startups. The best-known source is a 2020 white paper by the Global Startup Studio Network: of studio ventures that raise a seed round, 72 percent reach Series A, against 42 percent of traditional seed-funded startups.

To be honest, that comes with caveats: it's a self-report by 258 studio startups, and the comparators come from entirely different sources and periods. The best independent analysis so far, Big Venture Studio Research 2024 using PitchBook data, finds no statistically significant edge for studios over pre-seed funds and accelerators, albeit on a small base of 38 studio deals. And researchers warn of the "phantom founder": the studio that keeps its equity but stops building.

That doesn't refute the thesis. It sharpens it. A study by Pankaj Patel and Christopher Chan shows that just knowing which studio a venture comes from explains around 30 percent of the differences in outcomes. More than country, industry or founding year. It isn't the model that makes the difference. It's the people running it, and whether they carry the risk through the hard phase too.

And there's a second point: more equity isn't automatically better. Take too much from a founding team and you have skin in the game, but no founders left who want to carry the venture. Aligned incentives means everyone holds enough to fight for it.

What changes about the collaboration

When the partner carries part of the risk, it doesn't only change their motivation. It changes the conversations. A partner with their own money and equity in the venture talks you out of ideas a consultant would happily have sold you. They say no earlier. And they stay when it gets uncomfortable, because that's the only way they get anything out of it.

That sounds like a soft effect. It's the hardest one there is. Because it moves the moment someone tells the truth forward. And every truth that comes earlier is cheaper.

The zoom-out: Europe doesn't have an intentions problem

Europe has money. Europe has science. Europe has companies with technology and market access. What's often missing is a system in between that reliably turns these ingredients into companies.

Such a system needs three things under one roof: the technology, the ability to build, and the capital that invests in what's built. But those three forces alone aren't enough. They have to be held together by something stronger than an org chart: an economy in which everyone wants the same thing, because everyone shares the same risk.

The principle: share the risk, don't shift it

What would a collaboration look like in which the incentives really point the same way? It would have four properties.

First: earn on the outcome, not on time. The work runs at cost, without margin. The return comes through equity, so only if the venture carries.

Second: your own capital in play. Whoever builds also invests. Not symbolically, but enough that failure really hurts.

Third: build until it's independent. Skin in the game applies to the build phase, not just to the cap table. The goal is a venture that works without the builder, not one that depends on it.

Fourth: equity that leaves founders room. Enough of a stake for the builder to fight. Little enough that the founding team fights too.

Coming full circle

So never ask what someone promises. Ask what happens if it goes wrong. The answer tells you everything.

Incentive structures are more honest than statements of intent, because they keep working when no one is watching.

Skin in the game is the only form of conviction you can't fake.

Sources

• Jensen, Meckling, "Theory of the Firm: Managerial Behavior, Agency Costs and Ownership Structure", Journal of Financial Economics 3(4) (1976): principal-agent theory — doi.org
• Levitt, Syverson, "Market Distortions When Agents Are Better Informed: The Value of Information in Real Estate Transactions", Review of Economics and Statistics 90(4) (2008): own homes 3.7% higher, 9.5 days longer on the market — doi.org
• Johnson, Rehavi, "Physicians Treating Physicians: Information and Incentives in Childbirth", American Economic Journal: Economic Policy 8(1) (2016): physician mothers around 10% less likely to get a C-section — doi.org
• Clemens, Gottlieb, "Do Physicians' Financial Incentives Affect Medical Treatment and Patient Health?", American Economic Review 104(4) (2014): 2% higher fees, around 3% more care — doi.org
• Bernstein, Giroud, Townsend, "The Impact of Venture Capital Monitoring", Journal of Finance 71(4) (2016): lead-investor proximity raises innovation and exit likelihood — doi.org
• CB Insights, "Why Startups Fail" (2026 update): 431 shutdowns, 70% ran out of capital, 43% product-market fit, 29% timing, 19% unit economics — cbinsights.com
• Global Startup Studio Network, "Disrupting the Venture Landscape" (2020): self-report by 258 studio startups, 72% from seed to Series A, comparator 42% (Crunchbase) — web.archive.org
• Malyy, Pog et al., "Big Venture Studio Research 2024" (2024): no significant studio edge in PitchBook data — inniches.com
• Patel, Chan, "The Influence of Differences Between Venture Studios on Differences in Venture Outcomes", Venture Capital 26(3) (2024): around 30% of outcome variance explained by the studio — doi.org
• Moiana, Ghezzi, Rangone, "Venture Studios Beyond the Hype: Key Challenges and a Way Forward", Business Horizons 69(4) (2026): the "phantom founder" risk — doi.org