What an Unequal Split Can Price—and an Equal Split Can Hide
Once founders have decided to negotiate rather than accept the equal default, the ownership percentages become evidence about what the team was willing to recognize as a difference. This chapter focuses on how to read that evidence. The key tool is the share premium: a comparison with the equal benchmark that shows what an unequal founding team priced, while also making clear that no percentage is a complete measure of a founder’s value.
The benchmark is a counterfactual, not a norm
Suppose a startup has N founders. The equal benchmark is 1/N of the founder equity for each person: 50% in a two-founder team, about 33.3% in a three-founder team, and 25% in a four-founder team. A founder’s share premium is the actual percentage they receive minus that benchmark.
If two founders divide the equity 60/40, the first founder has a 10-percentage-point premium and the second has a 10-point shortfall. The premium is not a reward that exists independently of the team. It is a relative measure: one founder’s gain is another founder’s loss, and the premiums within a team add up to zero. The measure answers a narrow question—who received more or less than an equal division—not the much larger question of who will create more value over the company’s life.
That narrowness is useful. It gives the team’s first deal a common reference point without pretending that equality is the objectively correct allocation. The researchers can ask what characteristics are associated with a founder moving above or below 1/N, and they can ask whether the characteristics that make a team more likely to split unequally also appear in the eventual premiums. Without a benchmark, a 30% stake could look large or small depending on whether the team had two, three, or five founders. With it, the analysis stays within the team’s own allocation problem.
The data come from the same CompStudy material examined in the earlier chapter: 1,476 founders in 511 private technology and life-sciences ventures. About one-third of the teams split equally. The share-premium analysis focuses on the 340 teams that did not. That restriction matters. The results describe how unequal teams distributed their ownership; they do not directly observe the unchosen allocation that an equal team might have reached after a different negotiation.
This is the first boundary to keep in view. When a study finds that a founder characteristic is associated with a higher premium, it is not saying that every team should convert that characteristic into a fixed number of percentage points. It is showing that, among teams that differentiated ownership, that characteristic often traveled with a larger claim relative to the equal benchmark.
What unequal teams actually price
The study identifies three especially consistent relationships. Founders with prior entrepreneurial experience received higher share premia. The founder identified as the person whose idea began the venture also received a higher premium. Founders who contributed more founding capital received higher premia. These are not merely abstract labels. They represent three ways a founding team may understand future risk and contribution.
Prior entrepreneurial experience can stand for knowledge that is difficult to write into a job description: familiarity with early hiring, fundraising, product uncertainty, or the repeated decisions that keep a young company alive. The data cannot isolate which part of that experience mattered. They show an association between being a serial entrepreneur and receiving more than the equal benchmark in unequal teams.
The idea-person result is similarly more complicated than “ideas are worth a certain share.” An idea may be easy to describe and difficult to execute. The founder who generated it may also bring tacit knowledge, creative ability, customer insight, or the persistence needed to turn an abstract concept into a product. The study’s result is consistent with those forward-looking contributions, but it cannot separate the idea itself from the other skills that may travel with the idea person.
Capital is more visible, but visibility does not make its meaning simple. A founder who invests money absorbs an immediate financial risk and may allow the team to reach a milestone before outside financing. At the same time, the amount of money contributed may correlate with wealth, confidence, or a different ability to bear risk. The finding says that greater founding capital was associated with a higher premium in the observed unequal allocations. It does not establish that cash is always more valuable than unpaid work, technical ability, or access to customers.
The pattern is important because it connects the two sides of the founding decision. At the team level, greater heterogeneity in prior entrepreneurship, idea generation, and capital investment made an equal split less likely. At the individual level, those same characteristics were associated with larger deviations from the equal benchmark when a team did split unequally. The relationship is not a formula, but it is coherent: differences that create a reason to negotiate are often the differences that receive a price once negotiation occurs.
The result for prior years of work experience is a useful counterweight to a simple seniority story. Years of work experience alone did not have a significant effect on the basic share premium model. A founder can have spent many years in employment without receiving a higher founding premium, while a more specific form of entrepreneurial experience does carry an association. What matters in the data is not “more biography” in the abstract, but which experience the team may regard as relevant to building this kind of company.
The allocation path can be shown as a comparison between what is observed and what remains hidden:
An equal split therefore does not contradict the existence of these differences. It tells us that the team did not turn them into different initial percentages, whether because the differences were uncertain, because negotiation was costly, or because equality had value of its own. The equal number is an outcome of the allocation process. It is not a finding that the team had no characteristics that an unequal team might have priced.
Roles make the split less tidy
Founder roles add another layer to the share-premium evidence. In the reported models, holding the CEO position, chairing the board, or serving as CTO was associated with a positive share premium. The CEO and chair effects were especially strong in the specifications that included roles, while the CTO effect was positive but smaller. These results show that a team may price authority and responsibility alongside prior experience, ideas, and money.
A role is not the same thing as an individual contribution. The CEO may carry greater responsibility for fundraising and hiring, but the title may also reflect who was most comfortable being the public face of the company. A chair may have formal influence over the board without doing the most day-to-day operating work. A CTO may be central to product execution, or the role may be temporary until the team hires more technical leadership. The observed premium could capture expected workload, decision rights, bargaining power, or several of these at once.
This is why a cap table should not be read as a single ranking of human worth. Ownership, title, control, and work can point in the same direction, but they can also diverge. A founder can hold a large stake and a modest title. Another can hold a smaller stake but control a critical operating function. The share-premium analysis finds that roles are associated with ownership differences; it does not prove that a role determines a person’s total future contribution.
Consider a hypothetical three-founder team. One person brings the initial idea, another supplies most of the early capital, and a third has built a company before. They could agree to one-third each. That agreement would preserve equality but would not tell an outside observer how they privately valued the three differences. Or they could negotiate 40/35/25, in which case the percentages would reveal that the team converted some perceived differences into ownership differences. Even then, the numbers would not tell us whether the person with 40% will work harder, whether the idea will prove useful, or whether the founder with 25% holds a complementary skill without which the company would fail.
The hypothetical is not a recommendation to choose one arrangement. It shows what the data can and cannot support. The study can identify recurring associations in unequal teams. It cannot calculate the correct split for a new team whose future tasks, relationships, and financing path are still unknown.
What an equal split leaves unmeasured
The most important hidden quantity is future contribution. The initial deal is made before the founders know exactly how roles will evolve, which skills will become scarce, or which risks will materialize. A founder who contributes little visible work before incorporation may become essential during a later product or sales crisis. Another who appears indispensable at formation may reduce their involvement. An equal split can be a way of sharing that uncertainty rather than pretending to forecast it precisely.
Complementarity is hidden too. Founders do not create value only by adding separate amounts. One person’s technical work may make another’s customer relationships useful; one founder’s capital may buy time for another’s product development; an idea may become valuable only when paired with a particular implementation skill. A differentiated percentage can recognize a visible difference while missing the value created by the combination. An equal split can likewise conceal differences while preserving cooperation around a joint capability.
The measure also does not capture every contractual protection. The detailed study examines the initial percentage and does not analyze later changes in founder shares. Its vesting indicator does not include the schedules and conditions that would show how ownership is earned over time. A founding percentage can therefore be accompanied by arrangements that change the economic meaning of staying, leaving, or missing a milestone. Without those details, the initial number is only one part of the bargain.
Bargaining power is another unmeasured factor. A founder may receive a premium because they bring a scarce capability, because they negotiated forcefully, or because the other founders needed them to join quickly. The data cannot turn those explanations into separate columns. Nor can they tell us whether a lower share reflects lower expected contribution, a stronger preference for equality among the other founders, or a compromise made to keep the team together.
That uncertainty does not make the evidence useless. It changes the claim we should make from it. A premium is a priced proxy for relative value in a particular team and time. It is not an objective score for talent, effort, moral deservingness, or long-run company impact. Conversely, a zero premium in an equal split is not evidence that all relative values were zero. It means the initial allocation did not express them as different shares.
The distinction is especially important when a founder is deciding whether an equal split “looks fair” from the outside. A reader may see an idea founder, a cash-contributing founder, and an operating founder and assume a particular percentage must follow. The study does not support that leap. It supports a more modest question: which differences are the team consciously pricing, which are being handled through roles or compensation, and which are being left for future performance or later renegotiation?
The practical reading of a share premium
A useful interpretation has three layers. First, identify the benchmark: how many founders are in the team and what would an equal division have produced? Second, identify the evidence of differentiation: ideas, prior startup experience, capital, and roles are associated with premia in the unequal-team data. Third, identify what the comparison leaves out: future work, complementarity, vesting, bargaining power, and the team’s private reasons for accepting or rejecting equality.
This approach avoids two opposite mistakes. The first is to treat an unequal split as proof that the larger shareholder is objectively more valuable. The second is to treat an equal split as proof that no meaningful difference exists. Both mistakes turn an allocation into a scorecard. The research supports neither. It shows that founders sometimes pay for recognizable differences and sometimes choose not to express them in the initial percentages.
The deepest lesson is that an equal split hides a pricing decision by making it invisible. The team may have concluded that the differences were not yet knowable, that the cost of ranking them was too high, or that balanced ownership mattered more than precise differentiation. Whatever the reason, the cap table alone cannot tell us which explanation is true. To understand the financial meaning of that choice, we next need to examine what equal and unequal teams look like when they reach the first outside financing round—and why a valuation relationship there should not be mistaken for causation.