The CEO–COO Relationship Isn’t a Division of Labor. It’s a Decision-Making Partnership.

Most descriptions of the CEO–COO relationship begin by outlining the division of work.

The CEO owns vision, strategy, fundraising, the board, and external relationships. The COO owns operations, execution, people, processes, and internal management.

There is nothing inherently wrong with that division. Role clarity matters, and ambiguity over who owns what can create enormous problems.

But after working on both sides of that relationship, I don’t think role clarity is enough.

The CEO–COO relationship is ultimately a decision-making partnership—and it has to be able to hold tension without breaking.

The reason is relatively simple: strategy and execution rarely separate as cleanly in practice as they do on an org chart.

A strategic decision eventually encounters operational reality. When it does, the COO shouldn’t simply be asking, “How do we execute this?”

Sometimes the more important question is, “What would have to be true for this strategy to work?”

Those are very different questions.

Where Does Strategy Actually End?

Take entering a new market.

At the highest level, the strategic decision may be straightforward: we’re going to expand.

But that decision immediately creates another set of questions.

Which market?

In what sequence?

With what resources?

What capabilities do we already have, and which ones would have to be built?

What assumptions are we making about demand, customer acquisition, delivery capacity, regulation, margins, or capital requirements?

And perhaps most importantly: what evidence would tell us to commit more capital—and what evidence would tell us that one of our assumptions was wrong?

Those aren’t simply execution questions.

The answers can materially change the strategy itself.

I’ve seen this firsthand. A market can look compelling from the perspective of TAM, strategic positioning, or customer demand and still be the wrong move because the organization doesn’t yet have the capabilities to absorb it, the economics don’t support it, or another adjacent opportunity offers a much better risk-adjusted path.

That doesn’t mean the strategy was foolish. It means operational evidence changed what we knew.

A strong CEO–COO relationship must be able to absorb that information without turning a challenge to an assumption into a challenge to authority.

That’s harder than it sounds.

The COO Has to Have Permission to Disagree

Every CEO says they want candid input.

The real test comes when the input contradicts something they already believe.

A COO may discover that the organization cannot support the growth rate embedded in the plan without adding significantly more capital. Or that the product isn’t ready for the market leadership wants to enter. Or that the economics of an expansion don’t work as originally modeled. Or that a strategic initiative is consuming resources that could generate substantially more value somewhere else.

At that point, the COO has a choice.

Execute the decision as given.

Or bring the evidence back upstream.

In my experience, the second is part of the job.

That doesn’t mean the COO gets a veto over strategy, and it certainly doesn’t mean every disagreement should become a philosophical debate between two executives. Ultimately, somebody has to decide, and in most organizations that person is the CEO.

But there is an important distinction between decision authority and the permission to challenge the assumptions that inform the decision.

A COO who can only challenge a decision after it has begun to fail has been brought into the process too late.

The CEO doesn’t have to agree with the COO.

But the disagreement has to be discussable.

Trust Is More Than Chemistry

This is where the conversation usually turns to trust.

That’s correct, but I think “trust” can become too vague to be operationally useful.

Trust between a CEO and COO isn’t simply whether they like each other, communicate frequently, or have complementary personalities.

Part of trust is knowing what happens when the two disagree.

Can the COO put uncomfortable data on the table without it being interpreted as resistance or disloyalty?

Can the CEO challenge the COO’s interpretation without the COO treating that as interference?

Can both leave the room after a difficult debate, support the decision that was made, and revisit it later if the evidence changes?

That kind of trust develops over time, but the permission for that relationship to exist needs to be established much earlier.

The research is consistent with this.

Harvard Business Review’s work on the COO role found that there is no single model for what a COO should do. The role changes depending on why the company created it in the first place. Some COOs are primarily executors. Some lead transformations or turnarounds. Some complement the CEO’s capabilities. Some are successors. And some are explicitly brought in as partners to the CEO.

Across those different models, however, one factor consistently emerged as critical: a high level of trust between CEO and COO. Importantly, the research didn’t place that burden entirely on the COO. It also identified obligations on the CEO’s side, including granting real authority and decision rights.

That last point matters.

Trust without authority can easily become consultation.

Decision Rights Need to Be Explicit

I’ve written before about decision rights between boards and management. The same principle applies inside the executive team.

What does the COO actually own?

What remains with the CEO?

Which decisions are genuinely shared?

What financial, strategic, or operational threshold causes something to move from one category to another?

What should the COO decide independently and inform the CEO afterward?

What needs to be escalated before action is taken?

These distinctions can feel unnecessarily formal when the company is small, or the relationship is working well.

Until they don’t.

Ambiguity rarely creates its biggest problem when the business is on plan. It becomes visible when cash is tighter than expected, a major customer is at risk, a key executive isn’t working out, an acquisition or expansion needs to be reconsidered, or the evidence supporting an important strategic assumption changes.

That’s a bad moment to discover that the CEO and COO were carrying different definitions of the COO’s authority.

McKinsey’s recent work on the COO agenda makes this point essentially. It recommends that an incoming COO understand what decisions should be made autonomously, what information the CEO needs, and which issues need to be escalated. More recent research increasingly describes the COO as a strategic bridge—mobilizing strategy, reallocating resources, and serving as a trusted decision partner to the CEO.

I think “decision partner” is the important phrase.

Because it changes where the COO enters the process.

A Shared Operating Picture Is Part of the Relationship

Decision rights alone aren’t enough either.

The CEO and COO also need to be looking at substantially the same business.

That sounds obvious. It isn’t always.

If the CEO is making decisions based on pipeline and market opportunities while the COO is looking at delivery capacity and contribution margin, both may be completely rational and still reach different conclusions.

If one is optimizing for the next financing round and the other for twelve-month cash requirements, the same investment can look either essential or reckless.

If one believes customer demand has been validated and the other believes a handful of large customers are masking weak repeatability, the disagreement isn’t necessarily about strategy.

It may be about facts.

That’s why I view the operating cadence—KPIs, financial reporting, forecasting, operating reviews, decision gates—not as administrative infrastructure, but as part of the CEO–COO relationship itself.

The objective isn’t more dashboards.

It’s a shared understanding of what is happening in the business, which assumptions are holding, which aren’t, and where the evidence has changed enough that a decision may need to change with it.

A shared operating picture doesn’t guarantee agreement.

It makes disagreement more useful.

The Best CEO–COO Relationships Operate Further Upstream

There is an important qualification here.

Not every company needs the same kind of COO.

A CEO can absolutely hire a COO primarily to execute a strategy that is already well defined. In the right company, at the right stage, with the right CEO, that can work extremely well.

But it’s a different role from the one I’m describing.

The strongest CEO–COO partnerships I’ve experienced operate further upstream.

The COO isn’t there to constrain the CEO’s ambition, and the CEO isn’t there to hand the COO a finished strategy and wait for implementation.

The CEO keeps pushing on what could be possible.

The COO keeps testing what would have to be true to make it possible—and whether the organization can actually absorb what leadership is asking it to do.

That creates tension.

I think some tension is exactly what you want.

The CEO may see an opportunity before the organization is ready for it. The COO may see constraints so clearly that the opportunity looks less attractive than it actually is. One can pull too far toward ambition; the other too far toward operational certainty.

Neither perspective is sufficient on its own.

Sometimes the CEO is right. Sometimes the COO is right. Often, the better decision comes from forcing those two perspectives into the same conversation.

The objective isn’t consensus for its own sake.

It’s a better decision.

The Organization Is Watching

There is one other reason this relationship matters beyond the two people involved.

The rest of the company learns how decisions actually get made.

People notice whether disagreement between the CEO and COO is productive or political. They notice whether the COO genuinely has authority or whether every consequential decision eventually has to be revalidated with the CEO. They notice whether data can change leadership’s position or whether the decision was effectively made before the discussion started.

And they adapt.

If employees learn that the COO’s decision can always be overridden by appealing to the CEO, some will eventually start shopping decisions.

If executives learn that challenging an assumption is interpreted as a lack of commitment, fewer uncomfortable facts will make their way upward.

If the leadership team sees the CEO and COO optimizing against different priorities, functions will begin doing the same.

At that point, what started as a relationship problem becomes an operating-system problem.

And those are much harder to fix.

Don’t Eliminate the Tension. Make It Useful.

I don’t think the answer is a perfectly defined CEO box and COO box on an organizational chart.

Nor is it that the two executives need to agree on everything.

Role clarity matters. Decision rights matter. Trust matters.

But the deeper objective is to create a relationship in which strategic ambition and operational evidence can continuously challenge each other without either one automatically winning.

The CEO should be able to push the organization beyond what it currently believes is possible.

The COO should be able to say, “Here’s what would have to be true for us to do that—and here’s what the evidence currently says.”

Then both have to be willing to change their position when the evidence changes.

That’s not dysfunction.

Done well, it’s part of the operating system.

For CEOs and COOs who’ve worked closely together: what did you learn about making that partnership work that you wish you had understood earlier?

Sources

·       Harvard Business Review — Nate Bennett & Stephen A. Miles, “Second in Command: The Misunderstood Role of the Chief Operating Officer” (May 2006). Based on in-depth conversations with dozens of CEOs and COOs. Identifies seven different COO models and finds CEO–COO trust to be a common success factor; also specifically identifies the CEO’s responsibility to grant authority and decision rights.  ⁠
Harvard Business Review — Second in Command

·       McKinsey & Company — “Delivering the Strategy: The COO Agenda” (November 2024). Particularly useful for your strategy/execution argument. It describes the COO as the operational link to strategy and explicitly asks which decisions the COO should make autonomously, which require escalation, and what information the CEO needs. 
⁠McKinsey — Delivering the Strategy: The COO Agenda

·       McKinsey & Company — “The ‘Bridge Builder’ COO: Delivering Results by Engaging Stakeholders” (July 2026). The newest and probably most useful source for the central thesis. It argues that COOs increasingly control fewer of the levers required to deliver performance and therefore need to act as strategic bridges; it specifically describes serving as the CEO’s trusted decision partner as a source of COO impact. 
⁠McKinsey — The ‘Bridge Builder’ COO