Most teams do not fail because of bad strategy or lack of talent. They fail because of the invisible dynamics playing out beneath the surface, the ones nobody wants to name out loud. Patrick Lencioni identified these patterns over two decades ago, and yet organizations everywhere are still wrestling with the exact same problems today.
Lencioni's model, introduced in his landmark book "The Five Dysfunctions of a Team," gave leaders a clear and practical framework for diagnosing why teams underperform. But knowing the framework and actually breaking free from it are two very different things. The dysfunctions have a stubborn way of creeping back in, even after workshops, offsites, and leadership training designed to eliminate them.
So why does this keep happening? In this analysis, we will take a hard look at each of the five dysfunctions, explore the psychological and organizational forces that keep them alive, and explain what leaders need to understand if they genuinely want to build teams that perform at a higher level. If your team keeps hitting the same walls, this is worth reading carefully.
The 20-Year Problem: A Framework Everyone Knows and Teams Still Struggle With
Patrick Lencioni's The Five Dysfunctions of a Team has achieved something rare in management literature: it became a standard. Published in 2002, the framework's five-level pyramid, moving from absence of trust through fear of conflict, lack of commitment, avoidance of accountability, and inattention to results, has shaped leadership workshops, board retreats, and HR programmes across industries for more than two decades. The Five Behaviors® programme, the official commercial product built on the model, reports that 89% of learners say it improved their team's effectiveness. That is a striking number. It is also an incomplete one.
The more revealing data point is behavioural. The official Lencioni YouTube video published in November 2024 had already accumulated over 450,000 views within months of release. Practitioners at every level of seniority are still actively searching for answers the framework introduced but has not fully closed. If the diagnosis were genuinely resolved, the search traffic would fade. It has not.
This raises the central question this article examines: if the diagnosis is accurate, widely adopted, and commercially validated, why do the same dysfunctions reliably resurface in the same organisations after every intervention? The answer proposed here is not that Lencioni is wrong. The pyramid's sequential logic reflects real organisational dynamics, and complex systems critiques of the model confirm that its intuitions about trust as a foundation are well-grounded. The problem is categorical, not empirical. Behavioural interventions applied to structural problems require perpetual re-application. They treat symptoms with precision while the underlying architecture continues producing those symptoms unchanged.
Lencioni locates the cure in changing how people behave. The structural argument locates the cure in changing how the organisation is designed. That distinction, understated in two decades of practitioner literature, is where this analysis begins.
What Lencioni Gets Right: A Steelman of the Five Dysfunctions Pyramid
The most intellectually honest place to start with Lencioni is not where the model falls short, but where it genuinely delivers. A steelman demands that courtesy before any critique earns its keep.
The pyramid's central contribution is architectural rather than merely descriptive. By sequencing the five dysfunctions in a deliberately causal order, Lencioni redirects diagnostic attention from surface symptoms toward structural roots. When a team consistently misses its commitments, the instinct is to tighten accountability mechanisms; the pyramid instructs you to look two levels lower, at the trust deficit that is preventing honest conflict in the first place. This re-direction, from symptom to cause, is the model's most practically durable insight and the reason it has retained relevance across more than two decades of team development work.
Lencioni's own analogy to Maslow's Hierarchy of Needs is instructive here, and it is worth holding precisely. Just as Maslow argued that physiological safety must be secured before belonging becomes psychologically accessible, Lencioni argues that vulnerability-based trust must be established before productive conflict becomes possible. You cannot skip levels in either hierarchy. A team attempting to hold each other accountable while operating on a foundation of defensive politeness is, in Maslovian terms, trying to reach self-actualisation without first addressing safety. The hierarchy logic is not merely rhetorical; it carries real diagnostic weight.
Crucially, the foundational claim did not remain untested. Google's Project Aristotle (2016) independently identified psychological safety as the single strongest predictor of high-performing teams, arriving at Lencioni's foundational variable from a data-driven direction rather than a practitioner-intuition one. Amy Edmondson's earlier research across 51 work teams had reached compatible conclusions as far back as 1999. The convergence of a practitioner fable written in 2002 with independent corporate and academic research conducted across different methodologies is a meaningful, non-trivial form of validation.
The model's enduring relevance is also contextualised by the broader trust environment in which organisations operate. Edelman's Trust Barometer research found that across government leaders, CEOs, journalists, and religious leaders, institutional trust remains critically low, with none of the tracked societal leader categories reaching threshold trust levels. The conditions that make trust-first frameworks necessary have intensified in 2026, not diminished. Teams do not form in a social vacuum; when ambient institutional trust is low, the work of building vulnerability-based trust within a team becomes harder and more urgent simultaneously.
Finally, the model functions as a genuinely valuable diagnostic language system. Organisations that work through the Five Dysfunctions framework honestly consistently surface relational friction that previously went unnamed, colloquially treated as "personality clashes" or "communication issues." Naming a dysfunction precisely, distinguishing avoidance of accountability from absence of trust, is a prerequisite for addressing it with any structural rigour. That linguistic precision remains the model's most transferable gift, regardless of what it cannot do.
The Gap Lencioni Doesn't Close: Behaviour as Symptom, Not Cause
Steelmanning Lencioni, as the previous section does, requires equal precision about where the model's analytical lens stops focusing. The Five Dysfunctions framework is, at its core, a behavioural and prescriptive instrument. It identifies five patterns of conduct, absence of trust, fear of conflict, lack of commitment, avoidance of accountability, inattention to results, and prescribes corrective behaviours at each layer. What it does not do, and was never designed to do, is ask whether the organisational structure itself is producing those patterns. That is not a minor omission. It is the gap through which the same teams fall, repeatedly, despite years of well-intentioned intervention.
When Structure Generates Behaviour
The structural argument runs as follows. When a leader carries a relational load that exceeds what one person can sustain, the teams beneath that leader will exhibit predictable downstream symptoms: low psychological safety, conflict avoidance, diffuse accountability. Those symptoms look, from inside Lencioni's pyramid, like trust deficits and commitment failures. They are treated accordingly, with workshops, vulnerability exercises, and conflict norms. But the generative cause, an overloaded leader who lacks the structural support to model and sustain safety consistently, remains intact. This is not a character flaw in team members that training can fix. It is a rational adaptive response to the structural reality above them.
Amy Edmondson's research on psychological safety is instructive here. Safety does not emerge simply because individuals choose more open behaviour; it forms in response to perceived leader accessibility, consistency, and availability. When a leader's attention is fractured across too many direct reports, too many decisions, and insufficient deputy support, the structural conditions for safety are absent regardless of what the team's trust-building offsite produced last quarter.
The Reversion Cycle
Behavioural interventions applied to structural problems follow a recognisable and commercially inconvenient cycle. A workshop raises awareness and generates genuine short-term improvement. Participants leave with new commitments. Then structural pressures, unchanged reporting lines, unchanged spans of control, unchanged leadership load, reassert themselves. Within months, the dysfunction returns. This explains, more precisely than any critique of Lencioni's content, why the same organisations re-engage with Five Dysfunctions material year after year. The intervention is not failing because the model is wrong. It is cycling because the model is addressing symptoms while the cause persists upstream.
Complementary, Not Competing
Precision matters here. This is not a refutation of Lencioni. The pyramid's sequencing of trust as the foundational layer is structurally consistent with the Cone Model's own finding that 14 of 21 things people need to flourish at work are generated peer-to-peer, not top-down. Both frameworks converge on the same insight: horizontal relational quality is the substrate on which team performance rests. Where they diverge is on the question of what produces and sustains that quality.
The Table Group's resource library offers extensive tools for behavioural change. What it does not offer is a structural answer to why the leader at the apex is isolated, why the deputy role is undefined, or why relational load accumulates at the top of organisations in ways that make sustained psychological safety nearly impossible to maintain. Structural redesign produces durable change because it alters the conditions that generate behaviour, rather than coaching individuals to behave differently within unchanged conditions. Behavioural and structural interventions are complementary; but applied alone, behavioural change is temporary improvement written on a substrate that keeps erasing it.
Who Actually Generates Trust? The Peer-Generation Mechanism Lencioni Leaves Implicit
Lencioni defines trust with precision: it is vulnerability-based, a willingness among team members to "let down our guard, admit our flaws, and ask for help." That definition is genuinely useful. What the model does not define is the structural mechanism by which that vulnerability becomes possible in the first place. The implicit answer, embedded in virtually every commercial implementation of the framework, is that the leader generates it. Leaders model vulnerability. Leaders go first. Leaders create the conditions. MTD Training's overview of Lencioni's five dysfunctions frames the language carefully, describing the required state as a "comfort level among team members," yet the prescription still flows downward: the leader demonstrates, the team follows. The generative mechanism remains unspecified, and that omission carries real consequences.
The 14-of-21 Structural Argument
The Cone Model, developed by William J. Oxfield in The Lonely Leader, makes that mechanism explicit in a way Lencioni's framework does not. Oxfield identifies 21 things people need to flourish at work and then asks a clarifying question most leadership frameworks avoid: which of these actually require the leader to deliver? The answer is seven: Direction, Vision, Mission, Mentoring, Reward, Advocacy, and Autonomy. The remaining fourteen, including Belonging, Safety, Trust, Accountability, Correction, Teamwork, Friendship, Encouragement, and Care, are generated peer-to-peer. They cannot be delivered top-down because their nature is horizontal. A leader cannot give a team member genuine belonging; peers confer it or withhold it. A leader cannot supply peer accountability; the structural conditions either exist for it or they do not.
This 14-of-21 split reframes how to read Lencioni's pyramid. Dysfunction #1 (Absence of Trust) sits in the peer-generated category. So does Dysfunction #4 (Avoidance of Accountability). A leader who responds to these dysfunctions by deciding to be more vulnerable, or more demanding, is operating on the seven items within their span of influence while leaving the fourteen peer-generated items structurally unaddressed. The Five Behaviors platform's trust module reinforces vulnerability-based trust as the solution without specifying the organisational conditions under which peer vulnerability becomes a rational and safe choice. Those conditions are a structural design question, not a behavioural one.
Hybrid Work Has Raised the Stakes
The post-pandemic workplace has sharpened this gap considerably. In co-located environments, the informal peer infrastructure, corridor conversations, shared lunches, spontaneous whiteboard sessions, generated many of the fourteen peer needs passively, without deliberate design. People built trust because proximity made trust cheap to build. Distributed and hybrid teams do not inherit that infrastructure. When peers are physically separated, the relational load that once ran automatically through proximity now requires intentional structural support. The fourteen peer-generated needs do not disappear in a hybrid model; they become harder to meet and easier to neglect while still appearing on an org chart as if nothing has changed.
The Diagnostic Question the Framework Omits
This leads to a specific and underused diagnostic question. Most teams working with Lencioni's pyramid ask: "How do we build trust on this team?" That is a reasonable question, but it defaults to a leader-led answer. The structurally prior question is: "Does our organisational design create the peer proximity and shared accountability that allows trust to self-generate?" Lencioni's model does not prompt that question. It describes the symptom (absence of trust), names the desired state (vulnerability-based trust), and implies the leader as the mechanism. The Cone Model's contribution is to locate the mechanism precisely: fourteen of the twenty-one things people need to flourish are only available from peers, which means the real unit of organisational design is not the leader-team relationship but the peer-to-peer architecture that either enables or obstructs horizontal trust generation. The question of how to build trust, it turns out, is downstream of the question of whether the structure was built to let trust form at all.
The Dimension the Five Dysfunctions Pyramid Doesn't Address: The Leader at the Apex
Every layer of the Five Dysfunctions pyramid maps a relationship between team members: whether they trust one another, engage in productive conflict, commit to shared decisions, hold each other accountable, and prioritise collective results. The Table Group's own resource library organises all content under those five team-health categories with no category dedicated to the structural condition of the person sitting above the pyramid. This omission is not a minor gap. It is a load-bearing blind spot. The pyramid describes what must exist among the people the leader leads, but it never asks what must exist for the leader themselves. The leader is cast entirely as architect of team conditions, never as someone whose own relational architecture requires deliberate structural design.
The Structural Consequence of the Apex Position
Leadership loneliness is not a character flaw and it is not a coaching problem. It is the predictable structural output of climbing an organisational hierarchy. Every promotion reduces the pool of genuine peers. By the time a leader reaches the apex, the org chart has quietly eliminated the horizontal relationships that generate the relational conditions every person requires to function well. Research has consistently found that more than half of senior leaders report significant professional isolation, not because they lack social skill, but because the design of the org itself removes the structural context in which peer relationships form naturally. Coaching can address mindset. A board relationship can provide governance. A peer CEO network can offer occasional solidarity. None of these is a structural peer with real delegated authority inside the same organisation, sharing the same operational reality, capable of providing the kind of daily horizontal exchange that every layer below the apex takes for granted.
The Deputy as Structural Cure, Not Succession Instrument
The Cone Model's answer to this deficit is precise and deliberately narrow. At the apex of the cone sit not one leader but two: the leader and a genuine deputy with real delegated authority. This is not a succession plan and it is not a second-in-command appointed for operational coverage during absence. It is a structural peer relationship, the one relationship at the apex capable of restoring the leader's access to the fourteen peer-generated flourishing needs that the Cone Model identifies. Of the twenty-one things people require to flourish at work, only seven fall to the leader to provide: Direction, Vision, Mission, Mentoring, Reward, Advocacy, and Autonomy. The remaining fourteen, including Belonging, Safety, Trust, Friendship, Accountability, Encouragement, and Care, are generated horizontally by peers. Without a genuine deputy, the leader at the apex has no structural mechanism to access any of them. The relational load this creates is not sustainable.
The Cascade the Pyramid Predicts but Cannot See
Practitioners working with the Five Dysfunctions framework consistently confirm that the pyramid's foundation, vulnerability-based trust, depends directly on the leader modelling it first. When a leader openly acknowledges gaps and uncertainty, they create the permission structure for the team to do the same. That modelling capacity, however, is not a fixed personality trait. It is a function of the leader's own relational condition. A leader operating under unsustainable relational load, without peer support, without horizontal challenge, without anyone at their level to offer correction or encouragement, will progressively lose the capacity to model the psychological safety that Lencioni's pyramid requires at its base. The signal of that degradation shows up as team dysfunction: trust erodes, conflict becomes avoidance, commitment slides, accountability disappears. The pyramid's own sequential logic, applied one level higher than the pyramid reaches, predicts exactly this failure mode.
The Missing Top Tier
Nothing in the current Lencioni ecosystem addresses this. The Five Behaviors programme offers both a Team and a Personal development track; neither discusses the formal deputy with real delegated authority as a structural design element. The Table Group's resource library contains no category for the leader's structural peer relationship. This absence is not incidental to the framework. It is the precise point at which a genuinely comprehensive team-health model would need a tier above the pyramid it has built. The deputy with real delegated authority is not a supplement to good leadership practice. It is the structural precondition for sustaining everything the Five Dysfunctions pyramid asks the leader to provide.
What Happens at 100 or 1,000 People? Scalability as the Unanswered Question
The previous sections have examined what Lencioni's framework diagnoses, what it misses structurally, and who actually generates the trust it prescribes. One question remains conspicuously unaddressed in the primary literature: what happens to the five dysfunctions when an organisation scales from a single team into a multi-layered structure of 100, 500, or 1,000 people? The answer is not simply "the dysfunctions get worse." The answer is that the framework was never designed to tell you.
A Model Built for One Room
The Table Group's own resources frame the Five Dysfunctions as addressing the relational dynamics within one cohesive group. Every diagnostic question, every prescribed intervention, every fictional scene in the original book centres on a single executive team. That is not a limitation the framework hides; it is the unit of analysis the framework was built for. The problem arises when practitioners carry a single-team model into a multi-layer organisation and expect it to behave as an architectural blueprint. It was never written as one. Neither The Five Dysfunctions nor Lencioni's subsequent works offer a published framework for how these five failure modes manifest differently at 100 people versus 1,000. This is a gap in the primary literature, not a position Lencioni took and defended.
Span of Control Does Not Increase Linearly
The Cone Model's fractal stacking argument introduces a structural insight that reframes this scalability problem entirely. As organisations grow, span of control does not simply widen indefinitely. Instead, it follows an oscillating cycle: roughly 10 direct reports at the first level, narrowing to approximately 5 as an intermediate management layer forms, compressing further toward 2 at senior nodes, before resetting to a full peer circle of approximately 10 at each power of ten. This pattern, grounded in parallels from Roman cohort structure, Jethro's counsel to Moses in Exodus 18, and Dunbar's research on cognitively manageable group sizes, means that relational load on leaders oscillates with altitude rather than accumulating monotonically. The layers where span narrows to 5 or 2 are precisely where leaders are most isolated from peer-produced support. Critically, they are also where the conditions Lencioni describes as dysfunctional become structurally overdetermined, not simply behavioural.
The Missing Variable: Horizontal Trust Infrastructure
Belbin's analysis of the Lencioni model notes that complementary frameworks are regularly grafted onto the Five Dysfunctions because single-leader models leave important questions about team composition and peer dynamics unanswered. The Cone Model names this gap precisely. Of the 21 things people need to flourish at work, 14 are generated peer-to-peer: belonging, safety, correction, teamwork, trust, training, recognition, friendship, encouragement, accountability, celebration, challenge, forgiveness, and care. At scale, the structural design question is whether each layer of the organisation has the peer circle capable of generating those 14 needs, or whether the org chart silently assumes the leader at each node is producing all of them. Most conventional org charts make exactly that second assumption, and most scaling organisations pay for it in dysfunction that no amount of team-level intervention repairs.
Why Local Fixes Don't Propagate Upward
This produces the core scalability failure of applying the Five Dysfunctions framework without a structural lens. A sub-team intervention that successfully reduces dysfunction within one layer leaves untouched the structural conditions generating dysfunction in the layer above it. If the senior leadership group still avoids accountability or lacks genuine trust, those patterns cascade downward and erode whatever the sub-team repair achieved. The fix must be fractal, applied at every layer with the same structural logic, or the gains remain local and temporary. Team-level improvements are genuinely valuable; the problem is mistaking them for organisational health. A model that does not account for how its prescriptions propagate through a hierarchy cannot, by design, answer the question that every scaling organisation eventually faces: why did the fix work in the room and fail in the building?
Ancient Structure, Modern Data: Why the Structural Argument Is Not New
The structural argument for how organisations should be designed did not originate in a Silicon Valley boardroom or a twentieth-century management textbook. It has been made, tested, and validated repeatedly across three millennia, in contexts ranging from the Sinai desert to the Roman frontier to a Google research programme. That convergence is not coincidence. It is evidence.
The earliest documented case appears in Exodus 18, where Jethro observes Moses attempting to adjudicate every dispute for an entire nation alone. His counsel is precise and structural: appoint capable leaders over thousands, hundreds, fifties, and tens, and delegate authority accordingly. The solution Jethro offers is not therapeutic. He does not suggest Moses develop greater patience, improve his active listening, or attend a resilience workshop. He prescribes a nested hierarchy with clear relational accountability at each level, because the span-of-control problem cannot be solved by the leader becoming a better person. It can only be solved by redesigning the structure around the leader.
Roman military architects reached the same conclusion independently, and operationalised it at extraordinary scale. The Roman legion was organised into centuries of approximately eighty men, cohorts of roughly four hundred and eighty, and legions of around five thousand. Each tier constituted a bounded peer group with its own internal relational accountability. Unit cohesion, the property military historians consistently identify as the Roman army's decisive advantage, was not a product of inspirational commanding officers. It was a designed structural output, built into the architecture of the organisation before a single soldier was trained. The conditions for what Lencioni would later call vulnerability-based trust were engineered in from the start, not trained in after the fact.
Neuroscience adds a third layer of evidence. Robin Dunbar's research on cognitive limits to social group size established that the human brain can maintain approximately one hundred and fifty stable social relationships before the relational map becomes cognitively unmanageable. When organisations grow past that threshold without structural redesign, trust does not degrade because people have become dysfunctional. It degrades because the cognitive load of tracking that many relationships exceeds neurological capacity. The implication for organisation design is direct: exceeding Dunbar's number without creating bounded sub-structures does not create a training problem. It creates a structural one.
Google's Project Aristotle, completed in 2016 after studying hundreds of internal teams, arrived at a finding consistent with all three precedents above. Psychological safety, the team-level condition closest to Lencioni's vulnerability-based trust, was the single strongest predictor of team effectiveness. Critically, the research found that psychological safety was generated primarily by team norms and structural context, not by the behaviour of individual leaders alone. Structure, not personality, set the conditions.
The convergence of Exodus 18, Roman cohort design, Dunbar's constraint, and Project Aristotle across three thousand years and four entirely separate disciplines points to a single uncomfortable conclusion: the structural argument is not a novel management theory. It is a repeatedly rediscovered empirical truth. Which makes the persistence of purely behavioural interventions as the dominant industry response to team dysfunction not a neutral professional preference but a question worth pressing. Structural change requires authority, investment, and organisational will. A workshop is easier to commission. That asymmetry in institutional convenience, rather than any asymmetry in evidence, may explain more about why the training industry remains behavioural than any honest assessment of what actually works.
From Diagnosis to Design: The Structural Assessment as the Next Step After Lencioni
Lencioni's framework earns its place as the entry point. It names the dysfunction, gives the team a shared vocabulary, and creates the psychological safety needed to discuss what everyone has been feeling but could not articulate. That diagnostic clarity is genuine and should not be discarded. The limitation is not that the model is wrong; it is that naming a dysfunction is not the same as locating its structural origin.
The structural question Lencioni cannot answer is: where in the organisational design is this dysfunction being generated? Three distinct failure modes require three distinct structural interventions. A peer-circle problem means horizontal trust infrastructure is absent at a particular layer, so accountability cannot flow laterally and routes back upward instead. An apex problem means the leader is operating without a genuine deputy carrying real delegated authority, guaranteeing structural isolation regardless of how much the team improves relationally. A span problem means relational load at a given layer has exceeded sustainable levels, so even willing people cannot maintain healthy working relationships under current conditions. These are design problems, not character problems, and team training cannot resolve them.
The Cone Model's four free assessments are built precisely to answer these questions. The Full Cone 360° maps authority, accountability, and relational load across the organisation's layers. The Proximity and Lone-Wolf Check identifies structural isolation or over-proximity that presents as interpersonal friction. The Five Leadership Voices assessment surfaces gaps in how strategic, relational, and operational voices are distributed. The Find Your Deputy assessment addresses apex isolation directly, providing a structured method for identifying who in the current organisation can genuinely carry the deputy role with real delegated authority rather than a second-in-command title on paper. All four are available free in English and Dutch at theconemodel.com.
The practical sequence this creates is straightforward: use Lencioni to name the dysfunction, use the structural assessments to locate its origin, then redesign the conditions generating it. That sequence produces durable improvement. Organisations that stop at step one frequently find that relational patterns improve temporarily and then revert, because the structural conditions were never addressed.
Conclusion: Extend the Pyramid, Don't Just Climb It
Lencioni's Five Dysfunctions pyramid remains one of the most accurate and practically valuable team diagnostic tools available. Its trust-first sequencing, causal logic, and shared vocabulary have helped leadership teams name what they were experiencing for over two decades, and that contribution is genuine and durable.
The single gap it leaves open is structural. The pyramid diagnoses behavioural symptoms without asking whether the organisation's design is actively producing them, and it addresses team-level dynamics without asking what the leader at the apex requires to sustain those conditions over time.
Three takeaways follow. First, audit whether your org design gives peers the structural proximity to generate trust horizontally, rather than placing that entire burden on the leader. Second, identify whether a genuine deputy with real delegated authority exists at your apex; not a successor in waiting, but a structural partner who resolves the leader's own isolation. Third, apply the fractal lens before scaling: each power-of-ten growth threshold requires structural redesign, not simply another team training cycle.
Leaders who have worked through Lencioni and are ready to address the structural layer beneath the behaviour can begin with the free assessments at theconemodel.com, including the Full Cone 360°, the Find Your Deputy tool, and the Proximity and Lone-Wolf Check. The pyramid is the right starting point. The cone is what extends it.