The Five Dysfunctions of a Team is a leadership framework created by Patrick Lencioni, introduced in his 2002 book of the same name. It describes the five root causes that prevent teams from functioning well, arranged as a pyramid where each dysfunction feeds the one above it.
This is a summary of the framework, an explanation of the pyramid model, and a practical look at how teams actually fix these dysfunctions inside a working operating system.
Five Dysfunctions of a Team summary
Lencioni's model identifies five dysfunctions that compound. A team struggling with the first dysfunction will almost always struggle with the second, and so on up the stack. The five, from the base of the pyramid to the top:
- Absence of trust
- Fear of conflict
- Lack of commitment
- Avoidance of accountability
- Inattention to results
The insight that made the book stick is not that these dysfunctions exist. Every manager has seen them. The insight is the order. Trust is not just one of five problems. It is the foundation that every other problem sits on. You cannot fix accountability without fixing conflict first. You cannot fix conflict without fixing trust first.
This matters because most teams try to fix the wrong layer. They put in OKRs to fix inattention to results, and the OKRs fail because nobody is holding each other accountable. They add accountability mechanisms, and those fail because the team will not have honest conflict. They build conflict norms, and those fail because the team does not actually trust each other. The pyramid is the diagnosis and the sequence.
Patrick Lencioni and why this framework held up
Patrick Lencioni wrote Five Dysfunctions of a Team as a business fable, which meant most business readers initially dismissed it as a soft book. It has now sold over five million copies and is regularly cited alongside more data-heavy frameworks in serious organizational research.
It held up because the model is grounded in something real: teams are made of people, and people behave predictably in groups. The five dysfunctions are not cultural opinions. They are patterns that appear across industries, company sizes, and geographies because they are driven by human psychology, not by management style.
Lencioni's other frameworks (the ideal team player model, the working genius model) extend the same underlying logic. But Five Dysfunctions is the one that got adopted at scale because it gave leaders a sequenced map, not just a list.
The framework is often run alongside EOS, the Entrepreneurial Operating System created by Gino Wickman. EOS gives a team its operating cadence (weekly Level 10 Meetings, Rocks, Scorecards, Accountability Chart). Lencioni's framework gives a team its relationship health diagnostic. They serve different purposes and work well together, which is why many EOS-running companies keep both in the room.
Absence of trust: the foundation
Lencioni defines trust here very specifically. He is not talking about trust in the sense of "I trust you to do your job." He is talking about vulnerability-based trust: the willingness to be honest about mistakes, weaknesses, and gaps without fear of it being used against you.
A team without this kind of trust operates with a protective layer on at all times. Nobody says what they actually think in the meeting. Problems get discovered late because people do not surface them early. Managers find out about failures from customers, not from their team.
This is the most common dysfunction in high-performing-on-paper teams. The team has good people, strong skills, and decent results. But they are performing at 60% of their potential because every interaction has a small tax on it: the performance of confidence, the suppression of doubt, the management of how things look.
Fixing it requires repeated demonstrations that vulnerability is safe. One exercise Lencioni recommends is the personal histories exercise (each team member shares brief personal background in a structured setting). Another is the personality profile review, where the team maps behavioral styles and discusses them openly. The goal is to make the team human to each other before asking them to perform together.
At Sneeze It, we run this with the human seats on our operating chart. When Bogdan joined as COO, the trust-building preceded the accountability conversations. In an org running AI agents alongside humans, the same principle applies to the agent seats, just differently: the humans who own agent accountability need to trust each other before they can surface when an agent is underperforming. The dysfunction lives in the people, not the tools.
The pyramid model: how each dysfunction builds on the last
The pyramid structure is the part of Lencioni's framework that is most often misunderstood and most often skipped.
People treat the five dysfunctions as five separate problems to solve in parallel. Lencioni's point is that they are sequential. You do not get to fix layer two until you have done real work on layer one.
Here is what the stack looks like in practice:
Layer 1: Absence of trust makes honest conversation impossible. People optimize for looking good.
Layer 2: Fear of conflict is the direct result. When trust is low, people avoid conflict because conflict is dangerous. Meetings become performance. The real issues live outside the room. This is not peace. This is artificial harmony, and it is expensive.
Layer 3: Lack of commitment follows from fear of conflict. People will not commit to decisions they were not allowed to weigh in on honestly. Commitment is not the same as consensus. Lencioni is explicit on this: a team can disagree and then commit, but only if the disagreement was aired. Fake consensus produces fake commitment.
Layer 4: Avoidance of accountability comes from lack of commitment. When people did not truly commit, they will not hold each other to the commitment. Peer accountability disappears. The weight falls on the leader to be the sole source of accountability, which does not scale.
Layer 5: Inattention to results is the outcome. When nobody is holding anyone accountable, the team starts optimizing for status, ego preservation, and individual goals instead of shared outcomes. The results on the scorecard start to mean less than who looks good in the room.
The reason most operating systems fail to move the needle on results is that they instrument layer five (KPIs, scorecards, dashboards) without addressing layers one through four. The instrumentation is correct. The foundation is broken.
How to fix the Five Dysfunctions
The fixes are not complicated. They are consistent and slow, which is why most teams do not execute them.
Fix trust with structured vulnerability. Schedule a session specifically for personal histories and team profiles. Make it a regular check-in, not a one-time retreat. The team needs repeated proof that honesty is safe, not a single demonstration.
Fix conflict by making it a norm. Lencioni calls the leader the "miner of conflict": someone whose job in the meeting is to surface disagreements that are being smoothed over. The question "does anyone have a concern about this we have not said out loud" does a lot of work. So does acknowledging when you change your mind based on pushback.
Fix commitment by separating commitment from consensus. At the end of every significant decision, say the decision out loud and ask each person to confirm they will support it even if they disagreed. This is not theater. It is a practice that makes commitment real.
Fix accountability by making it peer-driven. A team that trusts each other and commits to decisions can hold each other accountable without the leader being the sole enforcer. Weekly scorecard reviews help. So does naming the gap before the leader does. "My number was off this week because X" is a different meeting than "your number was off, why?"
Fix results by keeping the scoreboard honest. This is where a weekly operating cadence (like EOS Level 10 Meetings, developed by Gino Wickman) connects directly to Lencioni's framework. The meeting exists to look at the numbers together, honestly, and address what is below target. If layers one through four are working, this meeting is functional. If they are not, the meeting is theater.
At Sneeze It, our agent Tally pushes KPI values from local sources to our OTP chart four times a day. Tally's job is to keep the scoreboard honest. But Tally only makes the results conversation work if the human layer underneath it has done the trust and conflict work. The tool does not fix the dysfunction. The discipline does. You can read more about how we connect the scorecard to weekly accountability in Humans and agents on the same scorecard.
The sequence matters in the other direction too. Once you fix trust and conflict and commitment and accountability, fixing results becomes mostly a measurement problem. That is a much better problem to have. We wrote about how that shows up in hybrid teams in Adding an AI agent to your org chart is not configuration. It is hiring.
Frequently asked questions
What are the Five Dysfunctions of a Team? Patrick Lencioni's Five Dysfunctions of a Team are, from base to top of the pyramid: absence of trust, fear of conflict, lack of commitment, avoidance of accountability, and inattention to results. Each dysfunction enables the one above it, which means you fix them in order, starting with trust.
Is the Five Dysfunctions model still relevant? Yes. The framework is over two decades old and remains one of the most cited team health models in business because the underlying patterns are grounded in human psychology, not management trends. The specific interventions may look different in a remote or hybrid team, but the five dysfunctions and their sequence apply regardless of context.
How do you measure which dysfunction a team has? Lencioni's assessment tool maps a team's self-reported responses across all five dimensions. As a simpler diagnostic: look at your last five team meetings. If honest disagreement did not happen, fear of conflict is the likely layer to address. If people agreed in the meeting but did not execute afterward, lack of commitment is the layer. Work backward from the most visible symptom to find the root dysfunction underneath it.
How is the Five Dysfunctions model different from EOS or the Accountability Chart? EOS, the Entrepreneurial Operating System created by Gino Wickman, is an operating system for running a business, with tools like Rocks, Scorecards, and the Accountability Chart. Lencioni's model is a team health diagnostic. They address different problems. EOS tells you how to structure your organization and run your meetings. Lencioni tells you why the people in those meetings may not be performing to potential. Many companies use both, which is why OTP is designed to support teams running EOS-style operating rhythms alongside Lencioni-style health practices.
Can you run the Five Dysfunctions model with a remote team? Yes, with modification. Vulnerability-based trust is harder to build when teams never share physical space, which means the structured exercises (personal histories, behavioral profiles, real-time conflict norming) need to be more deliberate, not optional. The dysfunction stack is identical. The trust-building work requires more investment in a distributed environment.
Run your operating system in OTP
OTP is a chart where humans and agents share scorecards, rocks, and weekly meetings. When the Five Dysfunctions work has been done at the human layer, OTP is where the accountability and results layers live, so the weekly meeting has honest numbers and clear owners every time.
In Claude Desktop or Cursor or any MCP client, add this block:
"otp": {
"command": "npx",
"args": ["-y", "@orgtp/mcp-server"]
}
Restart the client. Then ask: "Use OTP to show me which seats on our chart have scorecards and whether any are below target this week."