Board oversight of strategy execution works when the board asks, before approval, what the plan depends on inside the organisation and who owns each dependency, and then returns to those answers at each review. Directors rate their boards most effective at strategy and least effective at organisational health and talent, which is where most execution risk sits.
Where is a board's blind spot on execution?
In the organisation between the plan and the numbers. Boards see the strategy and the results. They rarely see the handoffs, decision owners and leadership capacity the strategy relies on, and directors say so: in McKinsey's research they rated their boards least effective at organisational health and talent management, and spent three days a year on it.
The gap is structural, not a failure of diligence. A board sees management's account of the plan and the financial outcome one or two quarters later. Between the two sits the organisation that converts one into the other, and the board has no routine instrument for it. Management knows more about that organisation than the board does, which is the ordinary condition of oversight, and it means the board's influence lies in the questions it asks before the plan is approved and the evidence it insists on afterwards.
McKinsey's 2018 research found that fewer directors than before said their boards had a good understanding of their companies' overall strategy, while respondents wanted more time on strategy and on organisational matters such as structure, culture and talent. The appetite is there. What is usually missing is a set of questions that turn organisational matters into something a board can oversee rather than discuss.
What should the board ask before approving a plan?
Four things. Which handoffs between functions does the plan depend on, and who owns each? Which decisions move, and to whom? Which roles carry the plan, and what happens if one is empty for six months? And what evidence, beyond management's confidence, supports the answers? A plan that cannot answer these is not ready, however good the numbers.
McKinsey's earlier research among directors found that 44% said their boards simply reviewed and approved management's proposed strategies. Review and approval is where the four questions do most good, because that is the last moment the board can change the plan cheaply. Asking them is not a vote of no confidence in management. It is the board doing for the organisation what it already does for the financing: testing the assumptions the plan rests on before committing to them.
What should the board look at after approval?
The same four answers, at each review point, with evidence of whether they still hold. Did the named handoff owners stay in post? Did the decisions that were meant to move actually move? Are the roles that carry the plan filled and stable? Reviewing these alongside the financials gives the board an early reading, before the numbers move.
Russell Reynolds' research among 919 directors found that strategic planning and review and enterprise-wide risk review were the activities on which boards spent the most time, and questions whether that time produces results. One reason is that strategy reviews tend to re-present the plan rather than test the organisation carrying it. A standing item that tracks the handoffs, decisions and roles named at approval changes what the review is for, without adding a day to the calendar.
What evidence should the board ask management for?
Something independent of the plan's narrative. Management's confidence is an input, not evidence. The board can ask for a reading of the organisation against the plan: where the demand on teams and leaders exceeds what they presently support, which interfaces are fragile, and what argues against the reading. The form matters less than the fact that it exists before approval.
Boards already apply this standard elsewhere. No board approves a financing on the treasurer's assurance alone, and no audit committee signs accounts without an independent view. Execution risk is the one material risk to a plan where the board routinely accepts the sponsor's assurance as the evidence. Asking for a separate reading, whether from an internal function, an adviser or an instrument built for it, brings the standard into line with the rest of the board's work.
Whose job is this on the board?
The chair's, in setting the agenda, and the committee that owns people matters, in following up. Remuneration and nomination committees already look at the leaders who carry the plan. Adding the question of whether the organisation around them can carry it is an extension of their remit, and it puts execution risk on the agenda before it becomes a crisis.
The practical version is a page in the board pack, owned by the chair or the committee chair, that lists the handoffs, decisions and roles the current plan depends on, who owns each, and what has changed since the last meeting. It takes management an afternoon to prepare the first time and an hour to update. Its value is in the question it forces at every review: not whether the plan is on track, but whether the organisation that has to deliver it still can.
Where this stops
This note is about the questions a board can ask about the organisation behind a plan. It does not cover the board's legal duties, which differ by jurisdiction and governance code, the design of board committees, or how to evaluate individual executives. The research cited reports what directors say about their own boards and cannot show that asking these questions improves outcomes in a particular company.
How Atlas reads it
Atlas treats a new strategy as one of the six moments that should prompt a reading of the organisation behind the plan, and the board is the natural principal for that reading: owners and boards know less about execution capacity than the management they oversee, and monitoring pays when it costs less than the loss it prevents. A board sponsor is one of the buyers Atlas expects for decisions about leadership, succession and governance.
The reading sets what the plan demands against what the organisation presently supports, and shows which handoffs, decisions and roles carry the most risk. Each judgement carries the evidence for it and what argues against it, so a board can see the basis for a recommendation rather than take it on assurance, and the judgement is bounded to this plan and this horizon.
Sources
- McKinsey & Company, Toward a value-creating board, McKinsey Quarterly (2016)
- McKinsey & Company, A time for boards to act, McKinsey Quarterly (2018)
- McKinsey & Company, The Board Perspective: A collection of McKinsey insights focusing on boards of directors (2017)
- Russell Reynolds Associates, Three Areas Where Boards Spend Their Time But Don't See Results, Harvard Law School Forum on Corporate Governance (2025)
Questions
To test, before approval, what the plan depends on inside the organisation and who owns each dependency, and to return to those answers at each review. The board does not run execution. It makes sure the questions about handoffs, decisions and roles were asked, answered with evidence, and revisited when something material changes.
Which handoffs between functions the plan depends on and who owns each. Which decisions move and to whom. Which roles carry the plan and what happens if one is empty for six months. And what evidence beyond management's confidence supports those answers. A plan that cannot answer them is not ready for approval, whatever the financial case says.
Because they see the plan and the results but not the organisation between them. Directors rate their boards least effective at organisational health and talent management and spend only a few days a year on it, in McKinsey's research. Management knows more about execution capacity than the board does, which is why the board's questions matter.
At every review point the plan already has, and whenever something material changes: a new strategy, new ownership, a new operating model, new pressure, a new dependency or new evidence. The review is short if the answers given at approval are tracked as standing items. It is long and late if they are reconstructed after the numbers have moved.
When the plan is material enough that the board would not accept the sponsor's assurance on its financing, it should not accept it on the organisation either. An independent reading of the organisation against the plan, with the evidence for and against it, gives the board the same standard of evidence it applies to the accounts.