When an Entire Board Started Doing Shadow Work

Boards can ask harder questions of management than they are usually capable of asking of themselves. This is an ordinary structural feature of governance, and it is usually invisible — until the day it is not.

What I am about to describe is a case in which a board decided to stop running its own dynamics and started examining them. I am being deliberately generic about the organisation. The pattern is considerably more specific.

The conditions that made it possible

The board had arrived at a moment most boards arrive at eventually: the management team was being held to a developmental standard the board itself was quietly not meeting. The CEO had been, for eighteen months, receiving direct feedback from the Chair about the need to develop psychological safety in the executive team, address unspoken dynamics, work on conflicts that were surfacing in patterns rather than incidents.

The CEO, who had been doing this work seriously, eventually raised the obvious observation in a private conversation with the Chair. He noted that the board itself ran several of the dynamics it had been asking him to address. It deferred difficult conversations. It had individual members who never engaged in committee discussions but who would conduct extensive private communication afterwards. It had one director whose presence reliably constricted the range of what the board felt able to discuss. The board’s own operating quality had, for years, been something nobody had proposed examining.

The Chair, to her significant credit, did not dismiss the observation. She took it seriously enough to raise it at the next board meeting — not as a critique of individuals, but as a proposal. If we are going to hold management accountable for the quality of their team dynamics, we have to be willing to examine our own. Would the board agree to commission an external facilitator to conduct structured developmental work over the following six months?

The vote was not unanimous. Two directors abstained. One voted against. Five voted in favour. It proceeded.

What the work involved

Six facilitated sessions over six months, each approximately three hours long. Between sessions, individual conversations with each director. The work was structured around four specific phases: individual mapping, collective dynamics naming, specific behavioural commitments, and integration.

In the first phase, each director completed a private written exercise mapping their own characteristic contribution to board dynamics. What they did under pressure. What they avoided. What they knew they were projecting onto specific other directors. What they suspected others were projecting onto them. The exercise was kept confidential — I read the responses, but the individual material was not shared across the board.

In the second phase, I worked with the board as a whole on the dynamics pattern that had emerged from the private mapping. Not naming individuals, but naming the patterns: the particular way disagreement was handled, the specific conversations that reliably did not happen in the room, the implicit rules about who could challenge whom and about what subjects. The board recognised its own dynamics in the description. It was visibly uncomfortable. It was also, for the first time in several years, genuinely engaged.

In the third phase, the board made specific behavioural commitments. Not aspirational. Specific: the next time X happens, we will do Y instead of our current pattern. These were tracked across the subsequent sessions. The tracking mattered, because without it the commitments would have dissolved into good intentions.

In the fourth phase, we integrated what had shifted. Which dynamics had actually changed. Which had not. Which had mutated into different forms. The board developed, for the first time, a vocabulary for its own operating quality that was available to members in the moment rather than only in retrospect.

What specifically shifted

Three changes were measurable over the following year.

The pattern of conversations happening outside the room decreased substantially. Not because the board passed a rule against it. Because the conversations that had previously required side-channels were now being had in the room. The board members who had been running political agendas through private communications with the Chair found that the dynamic had, simply, become visible. Once visible, it became costly in a way it had not been before.

The board began asking different questions of management. Specifically, it began asking questions about the dynamics and interior conditions of the executive team, rather than only about outcomes. This shift was produced by the board having developed its own vocabulary for these dimensions. A board that cannot see its own dynamics cannot ask management about theirs with any credibility. A board that has done the work can.

The director whose presence had constricted the room became, over the course of six months, the director whose contribution unlocked it. He had, it turned out, been carrying a specific pattern from an earlier board role that had been activated by a particular feature of this one. The facilitated work surfaced it. He integrated it. The board discovered that what had read for two years as a problematic presence was actually a misinterpretation of the specific conditions that had been producing the behaviour.

What made the engagement possible

Four specific conditions, in my assessment, distinguished this engagement from ones that would have failed.

The Chair was willing to model the work. She did the private written exercise herself, identified her own patterns, and named them in the room. Without that, the other directors would have experienced the engagement as something imposed on them by someone who had declined to do the work. With it, the vulnerability became permissible for everyone.

The brief was structural, not individual. I was engaged to work with the board as a unit, not to give specific directors feedback. This mattered. The directors could engage knowing that the work would not be used to manage them individually. Political self-protection was not activated.

There was a genuine decision to spend the political capital required. Board dynamics work is not free. It requires directors to be available, to be vulnerable, and to examine things that several of them would strongly prefer not to examine. The Chair and the Lead Independent Director were willing to spend capital on the engagement, and the rest of the board followed.

The work was not framed as therapy. This mattered significantly. The language used was the language of governance quality, operational excellence, and board effectiveness. The practices were depth practices. The vocabulary was commercial. Without that translation, several directors would have refused the engagement regardless of who proposed it.

The wider implication

Most boards are running dynamics they would not tolerate in the organisations they govern. This is not because directors are less capable or less self-aware than the executives they oversee. It is because the structural conditions of board operation — infrequent meetings, political complexity, the careful management of collegiality, the diffusion of individual accountability — make the work of examining the room’s own dynamics considerably harder than the equivalent work with a senior executive team.

Harder does not mean impossible. The engagement I have described was, for the board involved, the most significant governance investment of the preceding decade. It was also considerably cheaper than the consequences of the dynamics they had been running invisibly for years.

The question, for most boards I am aware of, is not whether the dynamics are present. It is whether there is enough willingness at the top of the board to propose examining them. In many cases there is not. In a small number there is, and those boards go on to do developmental work that changes the organisations they govern — not by instructing management, but by modelling what they have asked management to do.