Startup board composition: who sits where, and who decides

A board begins as whoever signed the incorporation papers and turns into a negotiated object the day an investor writes a check. Each seat has an owner and each owner has a reason to want it. Founders want to keep the right to run the company, the investor wants a vote on the decisions that move its money, and the company is better off with someone in the room who is neither.

This page covers composition only: which seats exist, how many, who elects each one, when an independent director joins and what the seats are paid. What the board does and what its members owe the company is on the board of directors page, and the financial package a board reads before each meeting is on board reporting. Both sit in the board section.

Who sits on a startup board: three kinds of seat

A startup board typically has a mix of founders, investors and independent directors². The three are different legal objects, because each is elected by a different group of stockholders. The NVCA model voting agreement shows the pattern: holders of preferred stock, exclusively and as a separate class, are entitled to elect one director; holders of common stock elect two; and common and preferred voting together elect the balance⁷.

Investors get a seat through the stock they buy. Investors own preferred stock, and the right to appoint a director comes with it, so an investor who owns less of the company still gets a seat⁴. Ownership alone does not control the board: more stock does not mean control of the board⁴, because board seats represent governance power and shares represent economic power¹.

Observers have no vote, so they do not matter for board control⁴, though one source notes that an observer still has direct access to the decision-making process³.

How many board members should an early-stage startup have?

The law sets a low floor and leaves the rest to the company. A Delaware board consists of 1 or more members, each a natural person⁸. Seat count is a bylaw question, unless the company’s charter pins the number⁸. Owning stock is not a prerequisite for a seat unless one of those two documents makes it one⁸. A quorum is a majority of the total number of directors, unless the certificate or bylaws require more⁸. On a three-seat board, then, two directors make a quorum; on a five-seat board, three do.

Practice settles on small and odd. Early-stage companies typically start with three board members², and the ideal size lands on an odd number, three to five, to avoid deadlocks in voting². An odd number prevents tied votes, offers wide enough perspectives to make discussions valuable and keeps decision-making relatively simple³.

The range of possible boards is wider than the convention. FTX raised billions of external financing with a single board member, the founder-CEO⁵. That is the research’s example of how flexible private boards are, not a model.

Size also follows bargaining power. Where capital is abundant and entrepreneurs have strong bargaining power, boards tend to be under their control, with founders and executives occupying the majority of seats⁵.

Seed-stage boards

Seed boards are small and founder-heavy. A typical seed board has two founder-elected seats and, at most, one seat for the lead seed investor¹. Seed-stage composition is typically founder-heavy, with perhaps one investor representative². Capping the board at three people is often suggested³. Before any priced round, a pre-seed board can be founders only, with one, two or three founders making up the entire board⁴.

The financing documents write the structure down. A seed round’s documents typically provide that the board shall consist of two common directors, one of whom shall be the CEO, and a preferred director appointed by the majority of preferred shareholders or a specific VC firm that led the round⁴.

Convertible instruments change the answer. Seed investors who use simple agreements for future equity (SAFEs) usually receive no board seat at all¹. How that instrument works is on the SAFE notes page.

Whether to add an independent at this stage has a short answer: most seed companies do not need a formal independent director¹, and informal advisors can provide an outside perspective without the governance overhead of a formal seat¹. Some investors skip the seat altogether and take an observer role at pre-seed⁴.

Series A boards

Series A is where a board gets its third constituency. At the Series A level a small board still makes sense, with either three or five people³. Preferred stock comes into existence with the round, so preferred shareholders, common shareholders and independents each need to be represented with a seat³.

On a three-person board one founder leaves to make room for the preferred seat³. That founder can take an observer seat, which gives access without a vote³.

A vacant independent seat is common. Both sides frequently leave it vacant at closing, and the board then operates with founder-investor parity¹. Another source describes the same two-two board and concludes that the founders effectively still maintain control⁴. The two accounts agree on the split and differ on who benefits from a tie.

The chair is a separate question. The CEO can also be the chairperson, although a separate chairperson can help maintain a separation of board and CEO interests³.

Board composition by stage

Control moves in one direction as rounds accumulate. Before the second VC financing round, entrepreneurs typically control startup boards; by the fourth round, most board seats are held by investors⁵.

The shared-control stage is the one where the independent seat earns its keep, and it is the next section’s subject. The investor-controlled stage is the one founders negotiate against from the first term sheet.

What is an independent director on a board?

An independent director is a director who has no employment relationship with the company and is expected to exercise judgment independently of the company’s investors¹. A second definition adds the label some boards use: an independent director, also called a non-executive director, has no material relationship with the organization². In practice that means no shares held, no employment and no personal or professional ties that could influence judgment or create conflicts of interest².

Public companies have a rule. Under Rule 10A-3, each member of the audit committee must be a member of the board and must otherwise be independent⁹. To count as independent, a member may not accept directly or indirectly any consulting, advisory or other compensatory fee from the issuer or a subsidiary, and may not be an affiliated person of the issuer or a subsidiary⁹. The rule is written for listed issuers⁹. On a startup board the independent seat is not mandated by law⁵; it exists because the financing documents create it.

What the seat is for differs from the public-company model. The independent director’s primary job is to act in the best interests of common stockholders, including founders¹. Research on startup boards finds that independent directors act as mediators, resolving potential conflicts between VCs and entrepreneurs⁵, and an earlier law-review analysis describes the tie-breaking seat as a practice of privately held startups⁶. Allocating the tie-breaking vote to an unbiased third party prevents opportunistic behavior by both founders and investors¹.

Delaware courts apply a fact-specific independence test that goes beyond formal labels¹. When a contested transaction receives approval from a majority of disinterested directors, courts apply the business judgment standard rather than the more demanding entire fairness test¹.

When to add an independent board member

At seed, advisors do the job that a seat would¹. Series A is the standard trigger, the round when independent directors typically become standard¹. After a second round of financing it is fairly common to designate an independent seat, and by Series B the absence of one becomes a recognizable governance gap¹.

The research shows when the independent seat carries a tie-breaking vote, and when it does not. Independent directors are less likely to have a tie-breaking vote when VCs and entrepreneurs have worked together in the past⁵. When VCs have a reputation for being heavy-handed in replacing startup executives, independent directors become a valuable resource for settling disputes⁵.

The person in the seat changes with the stage. Independent directors joining early are less likely to have interacted with either VCs or entrepreneurs in the past⁵, which is what a mediation role requires. In later stages, independent directors tend to have stronger connections with VCs, more executive-level experience and previous service on public boards⁵. Independent directors at this stage also review related-party transactions and serve on audit and compensation committees¹.

For the search itself, the sequence is a gap analysis first. The search should begin by examining who already serves on the board and identifying the gaps¹, and the skills to add are those that fill the gaps in the team’s knowledge².

What are the disadvantages of independent directors?

The structure works only when the person is genuinely independent, and each drawback is a way that condition fails or a cost of meeting it.

• Independence in name only. A nominally independent director who defers to investors on every contested vote entirely defeats the seat’s structural purpose¹.
• Selection by one side. In a split board with equal founder and investor seats, if the investor effectively controls who fills the tie-breaking seat, the board functionally favors investor interests even when the formal structure appears balanced¹.
• Too early. Formalizing an independent seat too early introduces structural rigidity before the company has the operating discipline to benefit from it, and a board seat carries fiduciary duties that attach automatically under Delaware law¹.
• Prestige without presence. A high-profile director who shows up disengaged provides no real governance function¹.
• Fiduciary overhang. One academic analysis of the practice suggests that heightened fiduciary protections could undermine the role of the independent director in startup firms⁶.
• A seat that costs equity and, later, cash. The seat is paid, which the compensation section below sets out¹.

None of these is an argument against an independent seat at the stage where sources treat it as standard. They are the conditions under which the seat stops doing its job.

How the seats are written into the documents

Board composition at venture-backed startups is enforced through binding legal instruments rather than informal agreements¹. Two documents carry it. The certificate of incorporation sets which class elects which seat⁷. The voting agreement then binds stockholders: each stockholder agrees to vote, or cause to be voted, all shares owned or controlled, so the named people are elected⁷.

The NVCA model handles each seat differently. The CEO seat follows the job: if the CEO director ceases to serve as chief executive officer, each stockholder promptly votes to remove the former chief executive from the board⁷. The independent seat is the mutual director, an individual who is not otherwise an affiliate of the company or of any investor, designated by mutual agreement of the other seated members⁷. Mutual approval language in the term sheet is necessary but, on its own, not sufficient¹.

Removal follows the class that elected a director. Under Delaware law any director may be removed, with or without cause, by holders of a majority of the shares entitled to vote at an election of directors⁸, and where a class or series elects a director, the removal vote belongs to that class or series⁸.

Seats are only part of governance. Protective provisions give preferred stockholders veto rights over actions like raising additional capital, selling the company or amending the charter¹, which is why the term sheet needs reading as a whole.

Worked example (our arithmetic): votes, shares and a director grant

Kestrel is a made-up software company. Two founders hold 40% each and one investor holds 20%, the split in the example⁴. After a Series A, the board has either three or five seats.

Worked example (our arithmetic): Kestrel’s board votes and a director grant

Founders with 80% of the shares hold 40% of the votes on the five-seat board, and they win a contested vote only with the independent’s vote. That is the gap between economic power and governance power¹.

On pay, the grant sizes below use a made-up base of 10,000,000 fully diluted shares and percentages from the sources.

The 25,000 shares of the illustrative grant vest over four years, or 48 months: 25,000 ÷ 48 is about 521 shares a month, or 6,250 a year. How option grants are taxed is on the stock options page.

Building the right board for your stage

Founders and investors negotiate composition during the financing round itself², and that makes composition a term as important as price: one investor argues that founders should obsess over board composition as much as they obsess over dilution⁴. The levers that sources describe are specific.

The last row comes with a caveat from the same source: the only way to get such super votes or extra common seats is a company with very strong founder leverage⁴. The balance both sides aim for is a fair distribution of seats, so that investors have sufficient input while founder control is not diluted away². For the rounds themselves, see how venture rounds work.

Compensating board members

Pay differs by seat, and it follows who the director represents. Founders and CEOs are company employees and are not paid for serving on their own boards³. Lead investors may be compensated, but many opt for preferred stock³. Independents often get equity in the form of common shares and are the most likely to receive cash at later funding stages³.

The two ranges overlap rather than conflict. CRV’s figure is for independent directors at early-stage startups, and Rho’s is for board members generally, so both are a range that depends on stage, not a rule^{1, 2}. The trade between equity and cash is the point of the grant: with cash the reward is immediate, but the board member’s investment is not as long-term². The exact premium for a chair varies by company and stage¹.