Startup Boards: Directors, Observers and Governance That Works

Most founders think about the board twice: when an investor asks for a seat, and when the board does something they did not expect.

Between those two moments sits everything that actually matters — who has votes, what those people legally owe, and whether the meetings produce decisions or just consume a day of preparation every quarter.

This guide covers how board composition evolves, what directors actually owe and to whom, the difference between directors and observers, how to run meetings that are useful, the committees you eventually need, and the moments when governance stops being administrative and starts being decisive.

1. How Board Composition Evolves

At incorporation the board is usually just the founders. It changes at each financing, and the pattern is fairly consistent.

Seed

Frequently no change, or one seat to the lead investor if the round is large. Many seed investors deliberately take an observer seat rather than a directorship — they get information without the fiduciary burden.

Series A

The structure that shapes everything afterwards. The classic arrangement is 2-1-2:

• Two common directors, usually the founders, elected by common stockholders
• One preferred director, elected by the Series A holders
• Two independent directors, appointed by mutual agreement of common and preferred

Note what this means: founders do not control the board. Nor do investors. The independents decide contested questions. That is the point of the structure, and it is why who fills those seats matters more than almost any other governance decision you will make.

Some Series A rounds settle on 3-2 with founders holding a majority. Whether you can achieve that depends entirely on your leverage — which is the same dynamic that governs liquidation preferences and every other term.

Series B and beyond

Each new lead typically wants a seat, and boards drift toward seven or nine members. Larger boards are slower and less candid. A common and sensible move is to consolidate — asking earlier investors to step back to observer status when a new lead joins.

2. What Directors Actually Owe

This is the part most misunderstood, and it becomes acute at exactly the wrong moment.

A director owes fiduciary duties — principally the duty of care (to be informed and act deliberately) and the duty of loyalty (to act in the interests of the company and its shareholders, not their own).

Critically: those duties run to the company and its stockholders generally, not to the investor who appointed them. A partner sitting on your board as the Series B director does not represent the Series B. In practice this creates genuine tension, because the same person is also a fund partner with obligations to their LPs.

The tension becomes concrete in specific situations:

• A sale where preferred and common receive very different outcomes. A $40 million exit that returns preference in full and almost nothing to common is excellent for one class and worthless for the other. Directors must consider the interests of stockholders as a whole.
• A down round or recapitalization led by existing investors, where insiders are on both sides of the transaction.
• Approaching insolvency, where directors' considerations shift toward creditors.

The standard protections in these moments are process-based: independent director approval, recusal of conflicted directors, a special committee, and sometimes an outside fairness opinion. They are not bureaucracy — they are what makes the decision defensible afterwards.

Directors are also personally exposed, which is why D&O insurance is a closing condition in almost every institutional round.

3. Directors Versus Observers

• Directors vote, owe fiduciary duties, are named in filings, are personally exposed, and are covered by D&O.
• Observers attend, receive materials, contribute to discussion, and have no vote and no fiduciary duty. Their rights come from contract, not corporate law.

Observers are convenient for everyone — the investor gets visibility, the founder avoids expanding the board. The risk is accumulation. Four observers plus five directors is a room of nine, in which candour disappears and management stops raising problems.

Manage it deliberately:

• Cap observer rights in your financing documents rather than granting them freely
• Tie them to a minimum ownership threshold, and make them lapse if ownership falls
• Reserve the right to exclude observers from specific sessions — particularly anything involving their potential conflicts
• Hold an executive session at every meeting — directors only, no observers, no management beyond the CEO. This is where the honest conversation happens.

4. Choosing Independent Directors

The independents are the most valuable and most neglected seats on the board.

What to look for:

• Operating experience at your next stage. Someone who has run the function you are about to build.
• Genuine independence. Not a friend of the lead investor, not an advisor being rewarded, not a founder's former colleague who will always agree.
• Willingness to disagree. The whole value of the seat is someone who will say the uncomfortable thing.
• Time. A well-known name who attends two meetings a year is worth less than an available operator.

Compensation: independents are typically granted equity — commonly a fraction of a percent, vesting over two to four years, sometimes with an acceleration provision on a change of control. Cash compensation is rare at early stage. Any grant must be board-approved and struck at the current 409A valuation.

Leaving the seats empty is common and is a mistake. An empty independent seat leaves a two-two board that deadlocks, or hands effective control to whichever side has a tiebreak. Fill them deliberately.

How to actually find one

The usual route — asking your lead investor for a name — produces a candidate who is, by construction, less independent than the seat requires. Better sources: founders one or two stages ahead of you who have recently been through the problem you are about to hit; operators you have met at industry events and already respect; and executives who have run the specific function you are about to build for the first time.

Run a real process. Meet three or four candidates, ask each what they would want to change about how the company is run, and give the strongest one a trial — invite them to observe two meetings before you formalise anything. A director who turns out to be wrong is genuinely difficult to remove, and two meetings of observation costs nothing.

5. Running Meetings That Are Actually Useful

The default failure is a meeting where management presents for ninety minutes, the board asks a few questions, and nothing is decided.

Before

• Send the package 48 to 72 hours ahead — financials, metrics, and a written narrative. Non-negotiable.
• State the decisions you want made. Put them on the first page. A board asked to decide something engages differently from a board being informed.
• Pre-brief individually on anything contentious. Surprising your board in the room is how founders lose support.

During

• Do not present the deck. Everyone read it. Ten minutes on what changed, then discussion.
• Spend the majority of the time on two or three real questions — the ones you genuinely do not know the answer to.
• Handle formal approvals efficiently — option grants, minutes, consents. They matter legally and take five minutes.
• Hold an executive session, every time, so it is never a signal that something is wrong.

After

• Minutes, promptly. They are a legal record and a diligence item. Approve them at the next meeting and keep them with your consents.
• Circulate decisions and owners within 48 hours.
• Sign the consents. Unsigned consents are among the most common findings in diligence.

Quarterly is the standard cadence, with monthly written updates in between — see our guide to investor updates.

6. Committees, and When You Actually Need Them

Early-stage boards do not need committees, and founders who create them at Series A are adding process without function. They become necessary as the board grows and as specific decisions start requiring independent judgement.

The compensation committee is usually the first, typically at Series B. Its real purpose is not policy — it is that decisions about executive pay and equity should not be made by the executives receiving them. Once you have a management team with meaningful grants, having independents own that process removes an obvious conflict and makes the outcomes defensible.

The audit committee follows when you have audited financials, material revenue and a finance function of real size. Its function is a direct line between the auditor and independent directors that does not pass through the CFO. Below that scale, the full board reviewing the audit is sufficient.

A special or transaction committee is formed for a specific purpose and dissolved afterwards — most commonly to evaluate an acquisition offer or an insider-led financing where several directors are conflicted. This is the committee that matters most, because it is the mechanism by which a decision made under conflict becomes defensible. Composition is the whole point: only genuinely disinterested directors, with their own counsel if the stakes justify it. Our guide to selling your company covers where this arises.

The general principle: form a committee when a decision needs to be made by people who are not conflicted, not because a governance checklist says a company your size should have one.

7. Protective Provisions: Control Without Seats

Board seats are only half the control picture. Protective provisions in your charter give preferred holders veto rights over specified actions regardless of board composition — typically including selling the company, issuing senior securities, changing the size of the board, amending the charter, taking on debt above a threshold, and paying dividends.

This means an investor with no board seat can still block a sale. Read these carefully at the term sheet stage; they are as consequential as the seats and get a fraction of the attention.

Frequently Asked Questions

Can we refuse to give an investor a board seat?

At seed, frequently yes — many seed investors are content with observer rights. At Series A a lead will nearly always require a seat, and refusing generally means finding another lead. What is negotiable is the overall composition, and specifically whether independents are appointed by mutual agreement.

What happens if the board deadlocks?

Nothing gets approved, which in practice paralyses the company. This is precisely why odd-numbered boards and filled independent seats matter. Some charters include tiebreak mechanisms, but a board that regularly deadlocks has a composition problem no mechanism fixes.

Can a founder be removed by the board?

Yes. The board appoints and removes officers, including the CEO. Founder employment is a board decision; founder share ownership is separate and governed by vesting — see our guide to founder vesting and acceleration. Removal from the board itself depends on which class elects that seat.

Do we need a board before we raise?

You have one from incorporation, even if it is just you. What most pre-seed companies benefit from is an informal advisory board — no fiduciary duties, no votes, purely advice, usually compensated with small equity grants. It is a genuinely useful structure and carries almost none of the complexity.

How do we manage a board member who is not helping?

If they are an independent, the seat can be changed — that is the point of an appointment by mutual agreement, and a conversation with your lead is the route. If they are a preferred director, you are largely dependent on the firm reassigning the partner, which does happen and is worth asking about. Either way, address it early; a disengaged director is a compounding problem.

What is the difference between a board and an advisory board?

Everything that matters legally. A board of directors has statutory authority, votes, appoints officers and owes fiduciary duties. An advisory board has none of these — it is a contractual arrangement with people who give advice, usually for a small equity grant vesting over one to two years. Advisors cannot bind the company and carry no liability, which is why the arrangement is easy to set up and easy to end. Founders sometimes describe advisors as “our board” to investors, which reads badly; be precise about which you have.

Should a founder be chair?

At early stage, usually yes and it is largely a formality — the CEO chairs the meeting because there is nobody else. As the board grows and particularly if the CEO role separates from the founder, an independent chair becomes valuable, because someone other than the person being evaluated should run the executive session and manage the CEO relationship. The transition is worth making deliberately rather than at a moment of crisis, when it will read as a loss of confidence.

How much time does a board seat actually take?

For an engaged independent, roughly one to two days per quarter — reading the package properly, the meeting itself, a call or two with the CEO between meetings, and occasional help on a specific problem. Candidates who tell you it will take less are describing a seat that adds nothing. Be explicit about the expectation before they accept, because the most common failure is not a bad director but an absent one who agreed to something they did not have capacity for.

What board materials do acquirers and investors actually review?

Minutes and consents, principally, and they read them carefully. What they are checking is that every share issuance, option grant, financing and material contract was properly approved, that the 409A was referenced where it needed to be, and that decisions made under conflict show the right process. Gaps here are among the most common and most time-consuming diligence findings, and they are almost always fixable in advance and almost never fixable under deadline.

The Bottom Line

Board composition is decided in the Series A term sheet and lives with you for years. Pay as much attention to it as to valuation, fill the independent seats deliberately, and read the protective provisions alongside the seat allocation.

Then run meetings that ask for decisions rather than delivering reports, and keep the minutes and consents signed.

Global Capital Network connects founders with investors and with the operators who make good independent directors. Get in touch or see our upcoming events.

This article is general information, not legal advice. Fiduciary duties are jurisdiction-specific and fact-dependent. Work with qualified corporate counsel.