The chair’s craft: agenda power, dissent and the chair–CEO relationship

What the chair actually does#

The role of the board chair is often described as running the meeting, and that description is why so many chairs run good meetings of boards that do not govern. The difference between a board that governs and a board that attends is almost always the chair. Not because the chair decides — the chair has one vote — but because the chair decides what the board will decide, in what order, with what information and in what conditions of challenge.

That is a craft with four levers: the agenda, dissent, the relationship with the chief executive, and the board itself. Each can be worked deliberately. Most chairs work one or two of them well and leave the others to habit.

• The agenda: what the board will decide, in what order, with which paper and how much time.
• Dissent: whether doubt in the room is drawn out, answered and closed, or left to silence.
• The chief executive: a relationship close enough for early problems and distant enough for a no.
• The board itself: its evaluation, its refreshment and its succession.

The agenda as an instrument of governance#

The agenda is the chair’s first instrument of governance: what is on it, in what order, with how much time and which paper, determines what the board can actually govern. An agenda that opens with the decisions and closes with the reports produces a different board from one that does the reverse. An item given a fixed short slot cannot be governed, only noted. A paper the chair has not read before it is circulated will arrive written to be approved rather than to inform.

Chairs who understand this treat the agenda as the place where governance is designed, and the meeting as the place where it is executed. They decide which items the board must decide, which it must challenge, and which it may note — and they build the papers and the time to match.

Manufacturing dissent#

Dissent is manufactured, not hoped for. Boards are populated by people who are courteous, busy and aware that they will sit together again next quarter; left alone, they agree. The chair’s work is a sequence: draw the doubt out of the director who has it and has not voiced it, protect that doubt while management answers it rather than letting the room move on, and then close the debate without cutting it off — so that the director who dissented feels answered, not overruled.

Dissent is manufactured, not hoped for.

The sequence fails at each step in recognisable ways: the chair who asks whether anyone has concerns and takes the silence as an answer; the chair who lets management respond with a reassurance and moves on; the chair who closes by summarising a consensus the room did not reach. Each is fixable, and each is invisible to the chair until a peer names it.

The chair–CEO relationship#

The chair–CEO relationship is an architecture, not a friendship: close enough that the chief executive brings problems early, distant enough that the board can still say no. Too close, and the chair becomes the chief executive’s advocate in the room, and challenge from other directors starts to feel like disloyalty. Too distant, and management prepares papers to survive the board rather than to inform it, and the chair learns of problems from the paper rather than from the person. The architecture is built in the conversations between meetings — what the chief executive tells the chair first, what the chair does with it, and what both agree the board will see.

The board as the chair’s own work#

The fourth lever is the one most often deferred. The board is the chair’s own responsibility: its evaluation, its refreshment and its succession are the work the chair alone can lead. An honest evaluation asks whether the board governed this year or attended; refreshment asks which skills the board needs that it does not have; succession asks who will chair next, and whether the board has a view on the chief executive’s succession that is its own rather than management’s. None of this is urgent in any given quarter, which is why chairs who do not schedule it never do it.

What to do next#

A chair who wants to test their own craft can do it with four questions about the last meeting: which decision did the agenda make possible, and which did it prevent; which dissent surfaced, and which was left in the room; what did the chief executive tell you before the pack that the board never saw; and when the board last evaluated itself honestly. Chairs, vice-chairs and committee chairs work these levers on their own boards, in case work and chaired simulations, in the Chairing the Board masterclass in The Helm.