Board decisions are rarely shaped only in the boardroom. Long before the vote, framing, sequencing, access and the cost of dissent determine whether directors exercise genuine collective judgement or merely legitimise an outcome already formed elsewhere.
How informal power shapes formal governance.
The board meeting begins at 9:00. The CEO presents a major strategic proposal. The financial case is compelling, the risks have been documented and the recommendation appears clear. Directors ask questions. Management responds confidently. The chair invites further discussion. At 10:22, the proposal is approved unanimously.
The minutes will record a disciplined governance process. What they will not record is what happened during the previous two weeks.
The CEO had already spoken privately with three influential directors. The chair had signalled that delaying the proposal could damage market confidence. The CFO had been asked to make the downside scenario more proportionate. One sceptical director had discovered that nobody else appeared willing to challenge the timing. Management had rehearsed the most difficult questions and agreed which concerns should be acknowledged, redirected or contained.
Nothing improper may have occurred. Each conversation can be explained as preparation. Each adjustment can be defended as clarification. Each individual may believe they acted in the organisation’s best interests. Yet by the time the board entered the room, the range of acceptable outcomes had already narrowed. The meeting did not truly make the decision.
It gave the decision institutional legitimacy.
The formal meeting is often the final act
Boards invest heavily in formal governance. They establish committees, circulate papers, record declarations of interest, document discussions and vote according to agreed procedures. These mechanisms matter. But they can create a dangerous illusion: that the quality of governance is determined mainly by what happens during the scheduled meeting.
In reality, many consequential decisions are shaped much earlier. They are shaped when the question is first framed. They are shaped through who receives access to the CEO and when. They are shaped by which risks enter the board paper, which are softened and which never appear. They are shaped through private conversations that indicate where power is already leaning.
Governance does not begin when the chair calls the meeting to order. It begins when someone decides what question the board will be asked. Consider the difference between two questions: Should the board approve this acquisition? And is this acquisition the best use of the company’s capital and leadership capacity at this moment? The first question directs attention towards approval. The second reopens the strategic choice. The wording may appear minor. The governance consequence is not. Whoever frames the question influences the range of answers the board is likely to consider.
The shadow board is not necessarily a conspiracy
Informal conversations are unavoidable in senior leadership. Directors need context. Chairs need to understand emerging concerns. CEOs should not surprise the board with complex proposals that members have had no opportunity to examine. Early discussion can expose weaknesses, improve recommendations and prevent meetings from becoming exercises in confusion.
The problem is not that conversations happen before the meeting. The problem begins when those conversations stop being used to improve the decision and start being used to secure the outcome. A legitimate consultation asks what we are missing. An attempt to control the process asks whether the CEO can count on the director’s support.
The distinction is rarely stated so clearly. It appears through tone, timing and expectation. A director may be asked for advice, while quietly understanding that the CEO is also measuring their loyalty. A chair may test how much opposition exists, not to strengthen the debate but to prevent it from disrupting the meeting. An executive may adjust the presentation after receiving feedback without reconsidering the underlying proposal.
This is how the shadow board forms. It is not a formal committee. It has no charter, recorded membership or published authority. It may include the chair, the founder, a dominant investor, a trusted former director, an external adviser, the CFO, the chief of staff or whichever individuals have privileged access to the leader. Its influence comes not from official decision rights but from proximity, trust, information and the ability to shape interpretation before others enter the discussion.
Power is often most effective when it does not need to declare itself.
How a decision is made before the vote
Several dynamics can quietly move a decision from open consideration to predetermined approval.
The framing advantage. The person who presents a proposal usually controls its starting assumptions. Management decides which problem the proposal is intended to solve, which alternatives deserve comparison and which risks are considered material. By the time the board receives the paper, numerous strategic choices may already have disappeared from view. A board asked to choose between Option A and Option B may never discover that Option C was rejected months earlier by one executive, or that doing nothing for six months was never seriously evaluated. The most powerful influence over a decision is sometimes not the argument made in the room; it is the option that never enters the room.
The sequencing advantage. The first credible interpretation of a complex issue becomes an anchor. A director who receives an early private briefing from the CEO will begin examining subsequent information through that narrative. Other directors may receive the same facts later, but they do not receive them from a neutral starting position. By the time the board paper arrives, influential members may already believe they understand what the decision means, why it is urgent and which objections are likely to be unrealistic. Information has technically been shared. Interpretation has not been distributed equally.
The access advantage. Not every director has the same relationship with the CEO, chair or executive team. Some can call at any time. Some receive additional context informally. Some are invited into early discussions because their expertise is valued. Others encounter the proposal only through formal papers. The result is a board with different levels of information, confidence and psychological ownership. Those consulted early are more likely to defend the proposal because they have helped shape it. Those who encounter it later may hesitate to challenge what already appears to carry momentum. Participation becomes influence, and exclusion becomes disadvantage, even when every director eventually receives the same document.
The social cost of dissent. Once senior figures appear aligned, disagreement becomes more expensive. The dissenting director is no longer merely questioning the proposal. They may feel that they are challenging the CEO’s judgement, disrupting the chair’s process or delaying an initiative colleagues have already accepted. Nobody needs to silence them directly. The room has already taught them what their disagreement will mean. This is one reason intelligent, experienced and independent directors can participate in weak collective decisions. Each person evaluates not only the merits of the proposal but also the relational and reputational consequences of opposition. A decision can be unanimous and still not be genuinely collective.
Consider, for example, a real scenario I have observed in a mid-sized technology firm. The board was presented with a strategic pivot into a new regional market. The chairman and the chief executive had already exchanged emails and held a private dinner to align their views. When a non-executive director raised a genuine concern about currency volatility and local regulatory hurdles, the response was not a counter-argument but a polite smile and a swift redirection to the pre-agreed talking points. The meeting concluded in forty minutes and the decision was passed unanimously. Six months later, the project failed, costing the company over twenty million pounds. In the post-mortem, the board lamented its lack of rigorous scrutiny, yet not a single member recalled having raised the decisive objections that would have saved them. The warning signs were present; they simply had no air to breathe in the room.
This brings us to the decisive test.
- Can a serious question change the proposal?
- Can new evidence delay the vote?
- Can the board reject an option the CEO has already publicly or emotionally embraced?
- Can a director challenge the assumptions without being treated as disloyal, uninformed or obstructive?
A board may allow every director to speak while retaining no meaningful ability to alter the outcome. That is not open governance. It is consultation after closure. The capacity to reopen a decision distinguishes genuine deliberation from ceremonial challenge. If no evidence, argument or risk can change what will happen, the meeting is no longer deciding; it is documenting approval.
Speaking is not the same as influencing
Boards sometimes defend their culture by pointing out that every director had the opportunity to speak. That is not enough. The real question is whether speaking could still change the outcome. A director may raise a serious concern and receive a detailed answer. Management may thank them for the challenge. The concern may even be recorded in the minutes. But if the answer was prepared before the question was asked, if no new evidence will reopen the recommendation and if every objection is treated as something to be managed rather than examined, the opportunity to speak is largely ceremonial.
Psychological safety at board level is not simply permission to disagree. It is evidence that disagreement retains causal power.
Why CEOs are tempted to control the process
A CEO may not intend to manipulate the board. They may genuinely believe that extensive preparation is responsible leadership. They want to avoid surprises, protect confidence and prevent an important proposal from being derailed by questions that could have been addressed earlier. The temptation is understandable.
A difficult board meeting can feel like a public test of the CEO’s competence. After investing months in a strategy, acquisition or transformation, the leader may no longer experience challenge as additional thinking. They may experience it as resistance to work that has already been completed. The closer the proposal becomes to the CEO’s identity, the more likely they are to manage the board rather than use it. They begin briefing allies, resolving objections privately and refining the narrative until uncertainty has been removed from the presentation. The meeting becomes smoother. The CEO appears prepared. The board appears aligned.
But the organisation may have lost the most valuable contribution the board could make: the ability to see what management can no longer see from inside its own conviction. A wise CEO, by contrast, brings the core dilemma to the board early, while the assumptions are still fluid. They ask for guidance, not merely for approval. This requires vulnerability and trust, but it is the only way to transform a ceremonial ritual into a genuine intellectual partnership. Executives who manage the board too successfully can deprive themselves of the board they need. A strong CEO does not bring the board a fragile idea that must be protected from challenge; they bring a serious proposal that can survive it.
The reality of bad faith
So far, this analysis has operated on the generous assumption that all parties act in good faith. This is a necessary and healthy starting point, but it is not always accurate. There are times when the pre-determination is not a product of cognitive bias or social pressure, but of deliberate manipulation. A director with a hidden conflict of interest, or an executive determined to protect their pet project, may intentionally withhold critical information, skew the debate or engineer the narrative to secure a desired outcome. Acknowledging this darker dimension is not cynical; it is prudent.
The antidote to such behaviour is transparency and rigorous documentation. Every board should maintain a clear decision log that records not just the final resolution, but the dissenting opinions, the discarded alternatives and the explicit risk assessments made at the time. This practice, which we might call decision memory, ensures that no one can retroactively rewrite history to justify a mistake. It also provides a powerful deterrent against those who might otherwise seek to manipulate the process. The existence of a robust, contemporaneous record changes behaviour, because it makes every participant answerable not only for outcomes but also for the integrity of their reasoning at the time.
The chair must protect the integrity of the decision
The chair’s role is not to ensure that every meeting runs smoothly. It is to ensure that the board’s collective judgement remains real. That requires more than inviting comments after management has presented its preferred conclusion. The chair must understand how the decision arrived in the room. Who participated in framing it? Which directors were consulted early? What concerns were raised privately? Did those concerns change the proposal, or merely the language used to defend it? Has management demonstrated a willingness to reconsider, or only a capacity to answer questions?
A chair who already knows how every director intends to vote should pause. That knowledge may reflect excellent preparation. It may also mean that the decision has been closed before collective deliberation has begun. The chair must be willing to create productive uncertainty, particularly when consensus appears too quickly. That could mean asking management to present the strongest case against its own recommendation. It may require postponing a vote when material information arrives late. It may mean inviting a director who has expressed concern privately to bring the substance of that concern into the collective discussion. Most importantly, the chair must separate support for the CEO from protection of the CEO.
A board can support its chief executive and still expose their assumptions to rigorous challenge. In fact, that is one of the most valuable forms of support a board can provide.
Warning signs that the meeting began too late
A board should be concerned when the paper contains one developed recommendation and several visibly weaker alternatives; when directors refer to important private conversations that others did not receive; when major objections are described as matters of communication or implementation rather than strategy; when management can answer every difficult question immediately but shows no willingness to reconsider; when the chair repeatedly refers to issues as already settled; when urgency is used to prevent examination of assumptions that have existed for months; when directors express significant concerns only after the meeting has ended; when the success of the meeting is measured by the speed or unanimity of approval; when nobody can clearly explain what evidence would cause the board to reject or delay the proposal; and when the board is invited to approve the decision but not to question how the decision was framed.
None of these signs proves misconduct. Together, they suggest that formal governance may be arriving after informal power has already completed the important work.
Questions that reopen a predetermined decision
Before approving a consequential proposal, the chair and directors should ask which parts of this decision remain genuinely open; what would need to be true for us to reject this recommendation; which assumption carries the greatest consequence if it is wrong; who has received information or context that the full board has not; which alternative was removed earliest, and by whom; what is the strongest argument against the preferred proposal; has feedback changed the decision, or only improved its presentation; who benefits if this decision proceeds, and who will absorb the risk; what would we decide if the proposal did not already have an executive sponsor; would each director express the same view if the vote were confidential; are we approving the strategy, or responding to the momentum surrounding it; and if we return to this decision in twelve months, what evidence will show that our reasoning was sound.
These are not questions designed to create delay. They are designed to ensure that speed has not replaced judgement.
The board needs a decision memory
Meeting minutes usually record the resolution, key discussion points and formal outcome. They rarely capture the full reasoning behind the decision. A board needs more than a record of what it approved. It needs a decision memory: what it believed at the time, which assumptions were accepted, which risks were consciously taken, which alternatives were rejected and what future evidence would trigger reconsideration. Without that memory, leaders can quietly rewrite the past.
A successful outcome is presented as proof of excellent judgement, even if luck played a decisive role. A failure is blamed on circumstances nobody could have anticipated, even when the warning signs were discussed privately before approval. A decision memory makes leadership answerable not only for outcomes but also for the quality of its reasoning. It also reveals whether the board truly shaped the decision or merely endorsed the narrative management brought into the room.
Real alignment comes after disagreement
Healthy boards do not seek disagreement for its own sake. They seek sufficient challenge to ensure that alignment has been earned. Real alignment can emerge only after directors have examined competing interpretations, confronted uncomfortable evidence and understood the consequences of the choice. Agreement that exists before this process may be loyalty, fatigue, political calculation or a desire to preserve relationships. It is not necessarily judgement.
The quality of a board cannot be measured by how calm the meeting appears, how quickly the vote is completed or how rarely directors openly disagree. Sometimes a smooth board meeting is evidence of excellent preparation. Sometimes it is evidence that the real meeting happened elsewhere. The distinction lies in whether the formal room still possesses the power to change the decision.
Before the next consequential discussion begins, the chair should ask one question: what, precisely, has not yet been decided? If nobody can answer, the board is not meeting to govern. It is meeting to perform approval. The minutes may state that the decision was taken at 10:22. But every person around the table will know whether the board made that decision, or merely arrived in time to approve what had already been decided in the shadows.
