What does a chief operating officer actually do? The widest seat in the house

What a chief operating officer is#

A chief operating officer is the executive who answers for whether the institution’s strategy is actually delivered: the operating model it runs on, the execution that happens across silos, the cost of running it and its resilience when something stops. Strategy is a promise the institution makes; operations is whether it is kept, and the COO answers for the difference.

That definition is deliberately about outcomes rather than departments. Which functions report to the COO varies from one institution to the next — operations, technology, shared services, sometimes much more — and the reporting lines are the least useful way to understand the seat. What does not vary is the question the board will ask the COO: did what we decided actually happen, at the cost we agreed, and did it hold when it was tested?

Why no two mandates are the same#

No two COO mandates are identical, and that is precisely the seat’s difficulty. The chief financial officer and the chief risk officer inherit mandates that are recognisable from one institution to another; the COO inherits whatever the chief executive has decided not to carry personally. Authority is shared across silos — the business lines own their revenue, technology owns its estate, finance owns the budget — while accountability for the outcome is shared with no one.

The practical consequence is that a new COO’s first task is to write the mandate down as it is actually held, not as the organisation chart implies, and to have the chief executive agree to it. A COO who skips that step spends the tenure discovering the mandate’s edges by walking into them.

The three movements of the seat#

Whatever the reporting lines, the work of the seat falls into three movements, and each carries decisions the COO alone must take.

• **The operating model.** End-to-end or functional, and the cost of choosing wrongly; make, buy or partner, as sourcing decisions the institution lives with for a decade; and where digital changes the model itself rather than merely the process.
• **Execution as a discipline.** The COO’s cadence — the metrics and reviews that actually move behaviour; transformation portfolios that survive contact with the front line; and cost programmes that stick, which is a question of sequencing, morale and knowing the point at which further cuts destroy more than they save.
• **Resilience and the board.** Operational risk and continuity as the COO’s standing brief; incident command when the institution stops; and the COO’s standing in the boardroom, from which the road often leads to the CEO seat.

COO and CEO: what each answers for#

The question of COO versus CEO is usually asked as a question of seniority, and that is the wrong axis. The two seats are distinguished by what each answers for. The chief executive answers for the promise — the strategy, the capital behind it, the story told to the board and the market. The chief operating officer answers for whether the promise is kept: the model, the execution, the cost and the day the institution stops. A chief executive who also tries to hold the second set of answers has no COO, whatever the title on the door; a COO who drifts into the first set has stopped doing the job.

Strategy is a promise the institution makes; operations is whether it is kept, and the COO answers for the difference.

Where the seat goes wrong#

The seat fails in recognisable ways. The first is the undefined mandate already described: authority discovered by collision. The second is the COO who runs today’s operations well and never builds tomorrow’s, so that the model quietly stops fitting the strategy. The third is the reverse — a transformation agenda so consuming that the daily operation degrades beneath it. The fourth is treating resilience as someone else’s brief until the day the institution stops and the board asks who was in command. None of these is a failure of technical operations knowledge; all of them are failures of the seat.

What good looks like, and what to do next#

A COO who holds the seat well can show three things: a written mandate the chief executive has agreed to; an operating model the strategy can actually run on, with the sourcing decisions and their long-term cost stated openly; and a cadence of metrics and reviews that has changed behaviour rather than described it. Underneath those sits a standing brief on operational risk and continuity, rehearsed rather than filed. The COO who can show all of that is the executive the board turns to when the promise is tested — and, often, the one it turns to next. The COO: Operations at Scale programme works exactly those three movements with a cross-industry cohort of chief operating officers and enterprise-remit executives.