Ask the members of a leadership team to write down who is on it. Most chief executives assume that is a trivial exercise. Writing in Harvard Business Review in 2009, the Harvard psychologist J. Richard Hackman reported that across the more than 120 top teams his group analyzed, fewer than 10 percent of members agreed about who was on the team, even though almost every team believed its boundary was obvious.
That finding is a useful calibration. Dysfunction at this level is rarely loud. There is no shouting, no walkout, no formal complaint sitting with the chief people officer. There is a group of capable, senior, well paid adults being polite to each other while the organization below them waits for a decision.
The chief executive is usually the last person to see it, for a structural reason rather than a personal one. Behavior changes when they walk in. What they observe is the version of the team that exists in their presence, and the signs below are almost all about what happens in the other rooms.
The short version
Leadership team dysfunction shows up as decisions that do not hold, conversations that happen twice, and a layer below that has learned to route around the team. The eight signs below are behavioral and observable. Most teams display three or four. The cause is usually a mix of an unresolved structural question, a conflict nobody has named, and a chief executive who rewards agreement without meaning to.
1. Decisions get made twice
The decision is taken in the meeting. The real decision is taken afterwards, in a corridor, a car park or a private message between two people who did not say what they thought in the room. Within a fortnight the original decision has quietly changed shape and nobody can point to when.
The tell is not that people disagree later. It is that the disagreement was available in the room and did not appear. Ask how many times in the last quarter a decision was revisited outside the forum that made it. If the answer is more than once or twice, the meeting is ceremonial.
2. The meeting is a broadcast, not a decision forum
Each function reports. Nobody asks another function a difficult question. The agenda is a list of nouns rather than decisions, and the chief executive speaks for more of the session than any two other members combined.
Status reporting is not worthless, but it does not need seven expensive people in a room for three hours. When the agenda has no verbs in it, the team has stopped being a team and become a reporting line that meets.
3. Nobody contradicts the chief executive in front of the others
Disagreement moves to one on ones. The chief executive hears every objection, but privately and serially, which means each objection arrives without the counterargument from the colleague who sees it differently. The chief executive then adjudicates alone and the team never has the argument.
Patrick Lencioni named this pattern in The Five Dysfunctions of a Team in 2002, describing artificial harmony as the consequence of an absence of trust, with a lack of real commitment following behind it. It is a practitioner model built from consulting work rather than a statistical study, and it is worth using as a vocabulary rather than as evidence. The description holds up regardless.
4. Two functions have a running dispute nobody will name
Sales and operations, product and engineering, clinical and commercial. Everyone knows. It has been going on for eighteen months. It gets referred to obliquely, as a process issue or a resourcing question, and it never appears on an agenda under the names of the two people involved.
The cost is not the friction, it is the delay. In research published in Harvard Business Review in 2015, covering nearly 8,000 managers across more than 250 companies, Donald Sull and colleagues found that conflicts between functions are handled badly about two times out of three: resolved after significant delay 38 percent of the time, resolved quickly but poorly 14 percent, or simply left to fester 12 percent. If this is your team, start with how to work through conflict between two senior executives rather than a team session.
5. Everyone is optimizing a different scorecard
Each member can describe their own targets precisely and nobody can describe a shared one. The incentive plan rewards functional performance, so functional performance is what you get, and the person who sacrifices their number for the company's number learns not to do it again.
This looks like a values problem and is almost always a design problem. Before diagnosing the people, read the compensation plan. It is usually a more honest statement of strategy than the strategy document.
6. The layer below has already worked out who is on which side
Directors know which executive to approach for which answer. They pre socialize proposals with the sympathetic member. They warn each other about which meetings are actually decisive. When a leadership team is split, the organization below becomes expert at navigating the split, and that expertise is expensive.
This is the sign with the widest blast radius. Gallup's 2015 State of the American Manager research found that managers account for at least 70 percent of the variance in employee engagement scores across business units, which means whatever the executive team is modeling does not stay at the executive team.
7. The team discusses people more than it discusses the work
A disproportionate share of airtime goes to individuals who are not present: their reliability, their attitude, their suitability. Some of this is legitimate talent work. Most of it, at volume, is a substitute for confronting a structural problem the team cannot solve, because a person is easier to discuss than an operating model.
8. The same decision returns every quarter
Pricing. The org structure. Whether to keep the underperforming region. It gets debated, concluded, and reappears three months later as though the earlier conversation had not happened. A decision that will not stay decided is usually a decision that was never actually made, only announced.
Look at what was recorded. If there is no owner, no date and no statement of what the team agreed to stop doing, the meeting produced a sentiment rather than a commitment. That gap is the subject of our piece on why leadership teams are not aligned on strategy.
Why this is hard to fix even when everyone can see it
The difficulty is not diagnosis. Most executive teams, asked privately, will list the same three problems in the same order. The difficulty is that every individual behavior producing the dysfunction is rational for the person doing it. The executive who stays quiet in the room has usually been burned for not staying quiet. The one who protects their function is being paid to protect their function. Nobody is behaving badly by their own lights, which is why appeals to better behavior do nothing.
The second difficulty is that the chief executive is inside the system they would have to change. They set the tone that made disagreement costly, they designed the incentives that fragmented the scorecard, and they usually cannot see either, because their information arrives through the very one on ones that are causing the problem. Asking them to lead the intervention is asking the most implicated person to be the most objective one.
The third is the one that gets avoided longest. Some leadership team dysfunction is a structural question the team is not permitted to reopen: who owns the customer, what the strategy actually is, whether two roles should be one. And some of it, honestly, is a person. Both of those take quarters rather than weeks, and both are contested, expensive and personally uncomfortable for whoever starts them. That is why teams keep choosing the offsite instead.
How to brief a facilitator without poisoning the room
If you are handing this to someone else, whether an internal partner or an outside practitioner, the brief determines the outcome more than the session design does. Give them incidents, not adjectives. Three specific decisions that stalled, with dates and what happened afterwards, are worth more than a paragraph describing the team as siloed.
Include the structural question nobody will reopen, name the two people whose dispute everybody talks about, and say plainly what you have already tried. Leave out your verdict on individuals. The moment a facilitator arrives holding your private assessment of two members, they are working with your view rather than the team's, and the team will detect it in the first hour.
Then do the thing that makes the brief safe: tell the team the brief exists and roughly what is in it, before the work starts. A facilitator who appears to have been briefed in secret is read as the chief executive's instrument, and everything said in front of them is edited accordingly. Transparency about the brief costs you nothing and buys the only thing the session actually needs.
If several of these signs describe your team and you want an outside read before committing to a program, that diagnosis is where our leadership team effectiveness work starts: interviews, observation of a real meeting, and a written picture of what is actually blocking decisions.
Frequently Asked Questions (FAQs)
What is the difference between a dysfunctional team and one that simply disagrees?
Healthy teams disagree in the room and commit to the outcome once it is decided. Dysfunctional teams agree in the room and dissent outside it. The diagnostic question is not how much conflict there is, it is where the conflict happens. Visible disagreement followed by real commitment is a sign of health, not a problem to manage.
Can a leadership team be fixed without changing anyone?
Often, yes, if the cause is structural, if the incentives can be rebalanced and if the chief executive changes how they handle disagreement. It is not possible when one member's behavior is the primary source and has survived direct feedback more than once. Deciding which situation you are in takes evidence rather than instinct, and getting it wrong in either direction is costly.
How long does it take to turn a leadership team around?
Expect two to three quarters for behavior that holds under pressure. Meeting design and decision rules change within weeks. Trust between two members who have avoided each other for a year rebuilds far more slowly, and only through repeated experience of a different outcome. Anyone promising a two day fix is selling the offsite, not the change.
Is leadership team dysfunction the chief executive's fault?
It is their responsibility, which is a different thing. Most of the patterns above are the accumulated consequence of small choices about how disagreement is handled and how performance is rewarded. Treating it as blame usually makes the chief executive defensive and stops the work. Treating it as a design problem they own tends to get it started.
Should we run a team assessment before doing anything?
An instrument is useful when it gives the team shared language for something they already half know. It is a waste when it substitutes for the conversation. Run interviews first, with every member and a sample of the layer below, then choose an instrument if the interviews show the team lacks a vocabulary rather than lacking courage.
How do we know it is a structure problem rather than a team problem?
Test it by asking whether a different set of people in the same roles, with the same incentives and the same decision rights, would have the same argument. If the answer is yes, you have a design problem and a team session will produce goodwill that decays within a month. Fix the ownership question first, then work on the team.