The Five Dysfunctions of a Team, With Examples

Most managers can name the symptom before they can name the cause. The weekly meeting ends with everyone nodding and nothing moving. A deadline slips and nobody says whose piece was late. Two senior people disagree in DMs and agree in the room. Patrick Lencioni gave these patterns a vocabulary in 2002, and more than two decades later the five dysfunctions of a team is still a go-to framework when a group of capable people keeps producing mediocre results.

The Five Layers of Lencioni's Pyramid

The Five Dysfunctions of a Team: A Leadership Fable was first published in 2002 by Jossey-Bass, a Wiley imprint, and is still in print in a 20th anniversary edition. The Table Group, Lencioni's consulting firm, says the book has sold over three million copies. It is written as a fable: a new CEO takes over a talented executive team that cannot work together, and the story walks through how she rebuilds it.

Lencioni arranges the five dysfunctions as a pyramid, each layer resting on the one below:

1. Absence of trust. People will not admit mistakes, weaknesses, or the need for help.
2. Fear of conflict. Without trust, the team avoids open debate about ideas and settles for surface harmony.
3. Lack of commitment. Without debate, people never fully buy in, so decisions stay ambiguous.
4. Avoidance of accountability. Without clear commitments, peers hesitate to call each other out on performance or behavior.
5. Inattention to results. Without accountability, individual status, ego, or department goals crowd out the team's collective outcome.

Fix the bottom and the upper layers get easier. Bolt a new OKR process onto a team with no trust, and it will not hold.

1. Absence of Trust

Lencioni means vulnerability-based trust: whether people can say things like "I got this wrong," "I don't know how to do this," or "I need help" without paying a social price for it.

What it looks like on a normal team: Status updates stay green until the week a project is suddenly red. Engineers polish work privately for days before showing it. In retrospectives, every problem belongs to another team or to "process."

At scale: After an 18-month investigation that reviewed roughly 600,000 pages of records, the U.S. House Transportation Committee's final report on the Boeing 737 MAX described a "culture of concealment." The committee found Boeing withheld critical information from the FAA, customers, and pilots, including internal test data. In one simulator test, a Boeing test pilot took more than 10 seconds to respond to an uncommanded activation of the MCAS flight-control system, a result the pilot called "catastrophic," while federal guidelines assume pilots respond in about 4 seconds. An organization where bad news travels slowly, or does not travel at all, has the bottom layer of Lencioni's pyramid missing.

What the research says: This layer has deep scientific backing, under a different name. Amy Edmondson's 1999 study of 51 work teams found that team psychological safety (a shared belief that the team is safe for interpersonal risk-taking) was associated with learning behavior, and learning behavior in turn was associated with performance. A later meta-analysis by Frazier and colleagues pooled 136 independent samples covering more than 22,000 individuals and nearly 5,000 groups and linked psychological safety to performance, learning, and engagement. Google's internal Project Aristotle research also named psychological safety the most important of five team dynamics. For a closer look at how this erodes on distributed teams, see seven signs your remote team has trust issues.

2. Fear of Conflict

Teams without trust avoid disagreement. Meetings feel pleasant, and the arguments move to hallways and side channels.

What it looks like on a normal team: The loudest or most senior voice sets the direction and everyone else goes along. Objections surface as questions ("Have we thought about...?") that get answered politely and dropped. After a meeting, three people message each other to say the plan will not work. Weeks later, the plan does not work.

At scale: After the Space Shuttle Columbia broke apart in 2003, the Columbia Accident Investigation Board concluded that "the NASA organizational culture had as much to do with this accident as the foam" (CAIB Report, Vol. I; Congressional Research Service synopsis). The Board found organizational barriers that kept engineers' differences of opinion from being heard, and a decision culture that put the burden of proof on engineers raising a concern: they had to show the foam strike threatened the flight. When raising a concern costs more than staying quiet, most people stay quiet.

What the research says: Here the science is more complicated than the book. A meta-analysis by de Wit, Greer, and Jehn covering 116 studies and 8,880 groups found that relationship conflict and process conflict had consistently negative associations with group outcomes, while task conflict (disagreement about the work itself) had no consistent positive or negative association with performance overall. For managers, the research backs disagreement that stays focused on the task and away from personalities, and it offers no evidence that simply adding more conflict helps. If unaddressed friction is already costing your team, the math in what workplace conflict costs and how to prevent it is sobering.

3. Lack of Commitment

Lencioni's point about commitment is that people can commit to a decision they disagreed with, as long as they were heard. Teams that skip the debate get fake agreement, and fake agreement produces half-hearted execution.

What it looks like (composite scenario): Consider a product team that leaves a planning meeting having "decided" to prioritize an enterprise feature. Nobody objected. Over the next sprint, the designer keeps iterating on the consumer onboarding flow she thinks is more important, one engineer starts the enterprise work, another keeps refactoring, and the product manager is surprised at the demo. Everyone left the room with a slightly different version of the decision, and nobody felt obligated to the version they had never argued for.

What the research says: Procedural justice research lines up well with this layer. In an experiment with intact management teams, Korsgaard, Schweiger, and Sapienza found that when leaders visibly considered members' input, members rated the process as fairer, and that perceived fairness increased commitment to the decision, attachment to the group, and trust in the leader, even for people whose view did not win. A broader meta-analysis of 183 organizational justice studies by Colquitt and colleagues found procedural justice made its own distinct contribution to outcomes such as organizational commitment.

4. Avoidance of Accountability

Lencioni's version of accountability is peer-to-peer: teammates willing to say "that's late" or "that's below our standard" to each other. When only the boss holds people accountable, the boss becomes the bottleneck, and everyone else learns that uncomfortable conversations are someone else's job.

What it looks like (composite scenario): Consider an engineering team where one developer's pull requests routinely skip tests. Three teammates have noticed. Each of them patches the missing tests during code review, or works around them, and complains to the manager in one-on-ones. The manager, hearing it three times, raises it in a performance conversation months later, and the developer is blindsided because no peer ever said a word. Standards on the team are now whatever the manager has time to enforce.

What the research says: In a study of 38 self-managed teams, Erez, LePine, and Elms found teams using peer evaluation had higher workload sharing, voice, cooperation, performance, and satisfaction than teams evaluated only by an outside authority. Teams that rotated leadership also showed more voice, cooperation, and performance. Karau and Williams' meta-analysis of 78 social loafing studies identified evaluation potential (whether an individual's contribution can be seen and judged) as one of the strongest moderators of how hard people work in groups. For practical ways to build this without becoming the team's enforcer, see how to build team accountability without micromanaging.

5. Inattention to Results

The top of the pyramid is the one the rest exists to serve. Teams fall into it when individual goals, team-of-origin loyalty, or personal status compete with the collective outcome, and win.

What it looks like on a normal team: The sales lead hits quota while her new customers churn in three months. Executives defend their own budget lines and nobody defends the company number. A team celebrates shipping on time without checking whether the release moved the metric it was built for.

At scale: The same House report on the 737 MAX names pressure to update the 737 design quickly and at low cost as a root cause of the failures it investigated. Schedule and cost targets won out over the collective result passengers depended on: a safe aircraft.

What the research says: This is the layer with the clearest quantitative support. A meta-analysis by Kleingeld, van Mierlo, and Arends found that group goals improved group performance with an overall effect size of d = 0.56, rising to d = 0.80 for specific, difficult goals compared with nonspecific goals. Individual goals inside groups cut both ways: goals designed to maximize individual performance ("egocentric" goals) hurt group performance (d = −1.75), while individual goals framed around each person's contribution to the group ("groupcentric" goals) helped it (d = +1.20). How you write individual goals can push a team toward or away from this dysfunction. Google's re:Work summary makes a related point about composition: how a team works together counts for more than who is on it, and teams with stronger team culture were rated as effective twice as often by executives.

How the Layers Stack in One Team

The five dysfunctions rarely show up alone. Picture a single composite team across one quarter. The tech lead privately doubts the migration plan but does not want to look slow, so she says nothing (trust). The architecture review is polite and short (conflict). Everyone leaves with a different mental version of the cutover date (commitment). When the database work slips, no one flags it, because flagging it means admitting their own piece slipped too (accountability). By quarter end, each person can point to a completed task list, and the migration is still not done (results). Diagnosing only the last symptom (a missed deadline) leads to a process fix. Diagnosing the stack points back to the first conversation that never happened.

If you are not sure which layer is weakest on your own team, the free team dysfunction assessment asks you to rate 15 statements across the five layers and takes a few minutes. A related lens is whether your team is cohesive or merely courteous, which the cohesive-or-just-polite hybrid team diagnostic covers.

Criticisms and Limits of the Model

The five dysfunctions framework is useful, and it is also a practitioner model with thin direct evidence. Managers should know where the ground is firm.

• It started as a fable. The book draws on Lencioni's consulting experience. It does not report a study, and peer-reviewed tests of the model itself are sparse and mostly qualitative or case-based.
• The sequence is untested. We found no study testing whether trust must come before healthy conflict, conflict before commitment, and so on. The pyramid order is Lencioni's argument. Each layer maps onto a researched construct, but the stacking is a hypothesis.
• Measured effects are modest. A 2026 preprint by Soares, Lima, and Ferrão, which built a survey measure of the five dysfunctions for engineering capstone teams, reported only a “weak-to-moderate relationship between team dynamics and performance”, with goal alignment and focus on deliverables standing out. It has not yet been peer-reviewed, and the authors argue team dynamics should be weighed alongside technical and contextual factors.
• It may be missing a layer. A 2023 study by Aggarwal in the International Journal of e-Collaboration examined groups that became dysfunctional enough to be reassigned or dismantled and argued the model should add "lack of functional expertise." Excellent dynamics cannot compensate for a team that lacks a critical skill.
• There is a commercial interest. The Table Group sells the official assessment, workbooks, and training. Judge the model on the research, and read the company's marketing claims as marketing.

The framework still earns its place, since each layer points at something the broader research supports. Treat it as a diagnostic vocabulary, and lean on the underlying research when deciding what to change.

How to Diagnose Your Own Team

You have three practical options, and they work best in combination.

1. The official assessment. The Table Group's Online Team Assessment is a 42-question, web-based survey with a color-coded team report, priced at $56.50 per license (one license per team member) at the time of writing. QuestWorks is independent of Patrick Lencioni and The Table Group and has no affiliation with the official assessment.
2. A free self-check. The QuestWorks team dysfunction quiz is organized around the same five layers and gives a quick read on where to look first. It is a conversation starter and makes no claim to be a validated instrument.
3. Observed behavior. Surveys depend on the very trust you are trying to measure. Notice who admits a mistake first, whether objections surface in the meeting or after it, whether peers flag missed commitments, and whether people talk about the team's number or their own.

Turning Diagnosis Into Practice

Knowing which layer is weak is the easy half. Changing it takes repetition: teams build trust, productive conflict, and peer accountability by doing those things many times under some pressure, and a single offsite rarely provides enough reps.

That is the premise behind QuestWorks, a team intelligence platform built around a weekly 25-minute game day. The whole team plays at the same time, seated into groups of 3 to 6 people, working through cinematic, voice-controlled quests that require coordination under time pressure. Leaders get a weekly Team Intelligence Score from 0 to 100 with an 8-week trend, named dimensions such as Role Clarity and Decision Velocity, and a "Do this week" list of concrete actions, drawn from how the team behaves in play. Participation is voluntary and the data is never tied to performance reviews. The tradeoffs: it asks for 25 minutes every week, the game format will not suit every team's culture, and it works best alongside a frank team conversation about the dysfunctions. There is a 30-day free trial covering four game days, no credit card required, and after that it costs $199 per month per team, with 10 seats included.

Whatever you use, start at the bottom of the pyramid. A team that can say "I was wrong" out loud has a much easier path through the other four layers.