Few organizational interventions create a stronger sense of action than a reorganization. New reporting relationships are debated, leadership roles are reconsidered, functions move, layers are added or removed, responsibilities are redistributed, and eventually a new organization chart emerges that visibly represents change.
The activity itself can create confidence that the organization is addressing its problems. Leaders can see the proposed future state. Employees can compare the old structure with the new one. Responsibilities appear to move closer to customers, functions appear more integrated, spans look more balanced, and organizational boundaries seem more logical.
Yet many reorganizations begin failing before any of those changes are implemented. The failure often starts when leaders move from recognizing an organizational problem to designing a structural solution without establishing sufficient evidence that structure is actually the primary cause of the problem.
Slow decisions may be interpreted as evidence of too many layers. Weak accountability may be attributed to reporting relationships. Poor collaboration may be diagnosed as organizational silos. Leadership overload may trigger changes in spans of control. Customer frustration may lead leaders to consolidate functions around customer segments. Rising costs may produce delayering. A new strategy may automatically trigger a new organization chart.
Any of those structural interventions could be appropriate. The problem is that the symptoms themselves do not establish that conclusion.
Slow decisions can result from unclear decision authority, fragmented information, unnecessary governance, competing priorities, risk aversion, insufficient leadership capability, or an operating culture that requires consensus before action. Poor collaboration can result from conflicting incentives, incompatible processes, weak cross-functional relationships, unclear priorities, or goals that encourage functions to optimize independently. Leadership overload may reflect excessive spans, but it may also indicate that authority has remained too centralized, managers lack sufficient capability, or senior leaders are routinely compensating for weaknesses elsewhere in the organization.
These conditions can look structural from the executive level because their consequences often become visible through hierarchy. The underlying cause may exist somewhere else entirely.
This creates one of the central risks of organizational redesign: leaders can accurately recognize that the organization is not working and still redesign the wrong part of it.
Research in organizational design has long demonstrated that organizational effectiveness depends on alignment among multiple dimensions rather than structure in isolation. Contingency theory challenged the idea that a universally superior organizational structure exists. Lawrence and Lorsch demonstrated the importance of differentiation and integration under different environmental conditions. Galbraith emphasized the relationship among strategy, structure, processes, rewards, and people, while Nadler and Tushman's congruence perspective reinforced the importance of alignment among organizational components.
The practical implication is significant. Organizational performance emerges from a system. Structure influences that system, but it does not independently determine how the system performs.
A reorganization should therefore begin with a different question than the one many leadership teams instinctively ask. Instead of asking what the new organization should look like, leaders should first ask what must become meaningfully different about how the organization operates, and what evidence indicates that structure is preventing that from happening today?
That question changes the purpose of organizational redesign. The organization chart is no longer the starting point. It becomes one potential response to a diagnosed organizational condition.
A reorganization is not a solution simply because an organization has changed its structure. It is a hypothesis that a particular set of design changes will produce particular improvements in organizational performance.
That hypothesis should be established before the boxes move.
Reorganizations Usually Begin With Real Problems
Most reorganizations are not arbitrary. Leaders usually have legitimate reasons for considering change. The business may have outgrown a structure designed for a smaller organization. An acquisition may have created overlapping capabilities and unclear ownership. A new strategy may require greater integration across products or markets. Customer expectations may have changed. Costs may have increased faster than value. Decision making may have slowed. Leaders may be spending too much time coordinating work that should occur naturally within the operating model.
These are valid signals that the organization's current design deserves examination.
The problem begins when a signal is treated as a diagnosis.
Consider an organization in which executives repeatedly complain that decisions take too long. Leadership may reasonably suspect that the organization has too many layers or that authority sits too high in the hierarchy. A delayering effort follows, management positions are removed, spans increase, and decision authority is expected to move closer to the work.
Six months later, decisions are still slow.
Suppose the original problem was not hierarchy. Suppose decision authority was never clearly defined, several functions retained informal veto power, governance forums still required consensus, leaders remained reluctant to make decisions without executive approval, and the information required to decide remained fragmented across systems.
Removing a layer would change none of those conditions.
The organization correctly identified the symptom and incorrectly identified the cause.
This distinction between organizational signals and organizational causes is foundational to effective redesign. Signals tell leaders where to investigate. They do not tell leaders what intervention to implement. Yet organizations frequently collapse those two steps. A visible problem is identified, an intuitive explanation is accepted, and design begins.
That is how a reorganization can fail before it begins.
Why Structure Becomes the Default Answer
If organizations contain many interacting systems, why does structure so often become the first target?
Part of the answer is visibility.
Structure is one of the few organizational systems executives can literally see. An organization chart translates a complicated enterprise into a comprehensible visual. Boxes represent roles or functions. Lines represent formal authority. Layers indicate hierarchy. Business units and functions can be grouped, separated, expanded, consolidated, or moved.
This makes structure unusually attractive as an executive intervention. A leadership team can spend an afternoon examining an organization chart and leave with visible evidence that decisions have been made. A function has moved. A layer has disappeared. A new role exists. Reporting relationships have changed. The future organization can be placed on a slide and communicated.
Compare that with redesigning decision authority. Decision authority is distributed across formal roles, governance processes, leadership behavior, organizational norms, information access, risk tolerances, and informal influence. Clarifying it requires understanding how decisions actually occur, not simply where authority appears to reside on paper.
The same is true of organizational capability. A box can be created for a new capability in minutes. Building the capability may require new expertise, processes, technology, leadership expectations, incentives, talent practices, and relationships across existing functions. Information flow, trust, culture, incentives, and cross-functional relationships are equally difficult to represent in a static chart.
These organizational conditions are harder to see and harder to redesign.
Structure, by contrast, is concrete.
TQG describes this tendency as structural bias, the tendency to favor structural interventions because structure is more visible, tangible, and easier to manipulate than many of the organizational systems that determine how work gets done.
Structural bias does not mean leaders are irrational. It means the design environment itself makes certain solutions easier to see than others. The risk is that collaboration problems become restructuring problems, accountability problems become reporting-line problems, speed problems become delayering problems, and customer problems become grouping problems before leaders have demonstrated that structure is the constraining variable.
The organization chart may become increasingly precise while the diagnosis underneath it remains surprisingly weak.
The Visible Organization and the Operating Organization
One of the most useful distinctions leaders can make during organizational redesign is between the visible organization and the operating organization.
The visible organization is what appears on the organization chart. It includes functions, business units, roles, titles, management layers, and formal reporting relationships. These elements matter because they establish formal accountability, grouping, hierarchy, and organizational boundaries.
The operating organization determines how work actually moves through and across those boundaries. It includes decision rights, governance, processes, information flows, leadership behaviors, incentives, capabilities, cross-functional relationships, informal networks, and the norms employees use when formal mechanisms are insufficient.
The two organizations overlap, but they are not identical.
An organization may formally decentralize decision authority while leaders continue escalating decisions upward because the culture punishes mistakes. Two functions may be placed under the same executive and continue operating independently because their incentives remain incompatible. A new customer organization may be created while the processes and technology employees use remain organized around products. Layers may be removed while governance expands to compensate for the loss of coordination.
In each case, the visible organization changes more quickly than the operating organization.
This gap helps explain why reorganizations can produce dramatic changes on paper while employees experience surprisingly little improvement in how work gets done. A new reporting line does not automatically create a new relationship. A new role does not automatically create a new capability. A flatter structure does not automatically produce faster decisions.
The structure matters, but it operates through a broader system.
The distinction is not intended to suggest that the visible organization is superficial or unimportant. Formal structure establishes real authority, accountability, grouping, hierarchy, and organizational boundaries.
The point is that structure operates through a broader organizational system.
Changing the visible organization without understanding the operating organization can therefore produce structural change without operational change.
When Symptoms Become Structural Solutions
Structural bias becomes particularly dangerous when symptoms are translated directly into design choices.
If collaboration is weak, functions are combined. If accountability is unclear, reporting relationships are changed. If leaders feel disconnected from customers, customer-facing units are created. If execution is slow, layers are removed. If innovation is weak, a new innovation function is established. If costs are high, activities are centralized. If the organization feels bureaucratic, authority is decentralized.
Each response can be appropriate.
The concern is that the same category of intervention is being used to address problems whose causes may be fundamentally different.
This creates a false sense of precision. A leadership team may spend weeks debating whether a function should report to the COO or CEO without first determining whether the reporting relationship has anything to do with the performance problem. It may carefully rebalance spans while leaving unclear what managers are actually expected to do. It may consolidate functions to improve coordination while failing to change the decision process that made coordination difficult in the first place.
The more detailed the organization chart becomes, the easier it is to forget that the entire redesign may rest on an untested causal assumption.
That is why the first major artifact of redesign should not be the future-state chart.
It should be a defensible diagnosis of the organizational condition that needs to change.
Reorganizations Can Move Problems Rather Than Solve Them
Structural redesign often changes where organizational tension appears rather than eliminating the tension itself.
Centralizing a capability may reduce duplication and strengthen standards while increasing distance from the business. Decentralizing it may improve responsiveness while recreating duplication and inconsistency. Increasing managerial spans may reduce layers and cost while reducing coaching capacity. Stronger business units may improve local accountability while making enterprise coordination more difficult. Consolidating decision authority may improve consistency while slowing local action.
These are not necessarily design failures.
They are organizational tradeoffs.
Every structure privileges certain outcomes while making others harder to achieve. There is no organizational design that simultaneously maximizes speed, control, specialization, integration, local responsiveness, enterprise consistency, innovation, efficiency, and accountability.
The relevant design question is therefore not whether the new structure eliminates organizational tension. It is whether the structure places the most important tensions where the organization is best equipped to manage them.
A reorganization that does not make those tradeoffs explicit may simply move friction from one interface to another and declare the first problem solved. Employees often discover the tradeoff before executives do because they experience the new dependencies in their daily work.
This is one reason TQG treats structure as part of a broader operating system rather than as the primary object of design. A structural choice should be assessed not only by the problem it resolves, but by the dependencies, coordination requirements, and new failure modes it introduces.
The Hidden Cost of Reorganization Debt
A poorly targeted reorganization does more than fail to solve the original problem.
It creates new organizational costs.
Some are visible. Organizations spend time on design teams, workforce planning, legal review, job architecture, compensation changes, communications, systems updates, recruiting, severance, and implementation. Leadership attention shifts inward. Managers answer employee questions while simultaneously maintaining business performance. Employees devote cognitive energy to understanding what the changes mean for their roles, careers, relationships, and workloads.
Those costs may be justified when redesign materially improves the organization.
They are much harder to justify when the organization reaches the other side of the transition and discovers that the original performance problem remains.
Repeated structural changes create an additional burden because employees learn from reorganizations. If they experience several redesigns that change leaders, titles, and reporting relationships without improving how work gets done, they begin forming expectations about future change. Announcements that once generated curiosity begin producing skepticism. Managers become reluctant to invest deeply in new arrangements because they expect another redesign to follow. Employees may wait for the structure to change again rather than investing in the current model.
TQG describes this accumulated burden as reorganization debt, the loss of organizational trust, energy, continuity, and change capacity created when repeated structural interventions fail to resolve the conditions that justified them.
Reorganization debt can compound. Informal networks are disrupted. Institutional knowledge can be lost. New leaders inherit teams and operating problems they did not create. Relationships must be rebuilt. The next transformation begins with less credibility than the previous one.
This produces a counterintuitive implication. Repeated reorganizations should not automatically be interpreted as evidence that an organization is adaptive.
An organization that continually redraws its structure without resolving persistent execution problems may instead be demonstrating an inability to diagnose itself.
A Reorganization Is a Hypothesis, Not a Solution
Every reorganization contains an implicit causal argument.
Leadership is effectively saying that if certain organizational conditions are changed, some aspect of organizational performance should improve.
If a company removes management layers, it is hypothesizing that fewer layers will improve something, perhaps decision speed, communication, cost, leadership proximity, or accountability. If the organization centralizes a function, it is hypothesizing that consolidation will create value through consistency, scale, expertise, control, or reduced duplication. If the organization decentralizes, it is hypothesizing that proximity to customers, markets, products, or operating decisions will create more value than the efficiencies of centralization.
The structural change is therefore not the outcome.
It is the intervention.
TQG's perspective is that every significant organizational redesign should make this causal logic explicit before implementation. A useful design hypothesis should connect the organizational condition being changed, the outcome expected to improve, and the evidence indicating why the current condition is constraining performance.
For example, an organization might conclude that decision authority is repeatedly escalating beyond the formally assigned owner, creating avoidable delays. Its design hypothesis may be that clarifying decision rights and removing duplicative governance will reduce decision cycle time without increasing unacceptable risk.
Another organization may find that fragmented ownership across product, sales, and service is generating repeated handoffs and inconsistent customer accountability. Its hypothesis might be that integrating selected customer-facing responsibilities and redesigning cross-functional decision rights will improve customer continuity.
These statements are more useful than saying the organization needs to become flatter, more agile, more customer-centric, or more integrated. They identify what leaders believe is happening, what they intend to change, and what evidence should eventually demonstrate whether that belief was correct.
Once reorganizations are understood this way, the organization chart becomes one possible intervention rather than the objective of the exercise.
The TQG Organizational Design Evidence Chain
The design hypothesis should emerge from a disciplined chain of evidence rather than executive preference alone.
The Evidence Chain deliberately changes where organizational redesign begins.
Leaders first identify the organizational outcome that is not occurring at the level required by the strategy. They then examine the visible signals, such as slow decisions, duplicate work, unclear accountability, poor customer continuity, excessive escalation, or leadership overload.
Those signals trigger investigation rather than intervention.
Evidence is gathered from operating data, decision patterns, workflows, organizational metrics, stakeholder perspectives, role and governance analysis, and other relevant sources. Patterns are examined until leaders can distinguish recurring organizational conditions from isolated frustrations.
Only then should root causes be established.
The next step is to define the operating requirements the future organization must satisfy. If slow decisions are the problem, perhaps certain decisions need to occur closer to the customer, with fewer approvals and clearer risk thresholds. If coordination is the problem, the future model may require stronger integration across a particular workflow without necessarily placing every participating function under the same leader.
From those requirements, leaders develop a design hypothesis and select the intervention. Structure may be part of the answer, but so may decision rights, governance, process redesign, leadership expectations, capability building, incentives, technology, or several changes working together.
Finally, the organization measures whether the outcomes that initiated the redesign actually improve.
If they do not, the appropriate response is not automatically another reorganization.
It is to revisit the hypothesis.
Structure Should Follow Operating Requirements
The Evidence Chain leads to one of the most important implications of this article: structure should follow operating requirements, not precede them.
Before changing reporting relationships, leaders should understand the performance outcomes the organization needs, the capabilities required to produce them, how work must flow, which decisions are critical, where authority should reside, what integration is necessary, and which interfaces create the greatest friction.
Only then can leaders make an informed decision about what structure best supports those requirements.
This reverses a common redesign process. Instead of beginning with the organization chart and asking how work should fit inside it, leaders begin with the work and determine what organizational arrangement is required to support it.
A management span is not effective because it matches a benchmark. It is effective when it allows the manager to provide the integration, judgment, coaching, development, and accountability the work requires.
A reporting relationship is not effective because it creates a clean chart. It is effective when it places authority and coordination where they improve organizational performance.
A layer is not unnecessary simply because another organization operates with fewer. It becomes unnecessary when the value it creates does not justify the cost, decision distance, or coordination burden it introduces.
Design requirements should therefore come before design preferences.
Leadership teams often enter redesign with strong views about centralization, decentralization, spans, layers, or functional ownership. Those experiences can provide valuable judgment, but they can also become solutions in search of problems.
Operating requirements create a common standard against which alternatives can be evaluated. The conversation becomes less about which structural philosophy a leader prefers and more about which configuration best supports the work the organization needs to perform.
Sometimes that analysis will confirm that significant structural change is necessary.
Sometimes the best organizational redesign will involve surprisingly little movement on the organization chart.
That is not a failure of design.
It is evidence that diagnosis came before intervention.
How to Know Whether the Redesign Worked
If a reorganization begins as a hypothesis, implementation cannot be the end of the process.
The organization needs evidence that the hypothesis was correct.
Many redesigns are evaluated primarily through implementation milestones. Leaders track whether new roles were filled, employees moved to the correct teams, systems reflected new reporting relationships, communications were completed, and transition activities occurred on schedule.
Those measures are necessary because poor implementation can undermine a sound design.
They do not establish whether the organization became more effective.
The outcome used to justify a reorganization should also help determine whether the reorganization worked.
If decision speed was the problem, decision cycle time should improve. If leadership overload drove the redesign, executive escalation and coordination burden should decline. If customer fragmentation was the concern, customer continuity and cross-functional coordination should improve. If duplication justified consolidation, duplicated work and cost should decline without creating unacceptable losses in responsiveness or capability.
This requires baseline evidence wherever practical. Without a baseline, leaders may be able to describe the new organization but struggle to demonstrate whether it is actually better than the old one.
Measurement should also examine the tradeoffs created by the new design. A delayering effort may reduce management cost while increasing escalation and reducing coaching capacity. Centralization may improve consistency while weakening responsiveness. Decentralization may accelerate local decisions while increasing duplication.
The purpose is not to create an enormous scorecard around every reorganization. It is to test whether the intervention produced the organizational improvement it was designed to create while monitoring the most consequential costs introduced by the design.
A flatter organization is not automatically a faster organization. A centralized function is not automatically a more efficient one. A completed reorganization is not automatically an effective one.
Implementation proves that the change happened. Outcomes provide evidence about whether it mattered.
That distinction should be established before the first reporting line changes.
What Great Organizations Do Differently
Organizations that redesign well begin with organizational outcomes rather than organization charts. They distinguish signals from causes and gather enough evidence to understand where the existing organizational system is constraining performance.
They examine the operating organization alongside the visible one. Reporting relationships matter, but so do decision rights, governance, workflows, information, leadership behavior, incentives, capabilities, and the interfaces through which work moves across boundaries.
These organizations establish operating requirements before evaluating structural alternatives. They ask what must become easier, faster, clearer, more reliable, or more scalable in the future organization, then use those requirements to test potential designs.
They also make tradeoffs explicit. A structure that improves local responsiveness may reduce enterprise consistency. A centralized capability may create scale while increasing distance from the business. A flatter model may accelerate some decisions while increasing managerial demands. Better design does not pretend those consequences disappear. It determines whether the benefit justifies the tradeoff and how the organization will manage it.
Most importantly, these organizations preserve traceability between the problem and the intervention. Major structural choices can be explained through evidence and organizational requirements rather than preference alone. After implementation, leaders evaluate whether the conditions that justified the redesign actually improved.
This approach may result in fewer reorganizations.
That is not a weakness.
It means structural change is reserved for organizational problems that actually require structural intervention.
1. What organizational condition are we trying to change, and what evidence demonstrates that we understand why it exists?
Broad statements such as "we need greater agility" or "accountability is unclear" identify areas of concern, but they do not establish the cause. Leaders should understand where the problem occurs, what mechanisms contribute to it, and why it matters to performance.
2. What must the organization be capable of doing differently when the redesign is complete?
The answer should describe an operating requirement or organizational outcome rather than a preferred structure. If leaders cannot describe what should become meaningfully different, it will be difficult to determine whether the redesign worked.
3. Why do we believe structural change is necessary to create that capability?
Leaders should examine decision rights, governance, processes, incentives, information, technology, leadership behavior, and capability before assuming reporting relationships are the primary constraint.
4. What tradeoffs will the proposed structure create, and are we prepared to manage them?
Every design solves some problems while creating new dependencies. Those consequences should be understood before employees are asked to operate inside the new model.
5. What evidence will tell us the redesign worked, and what evidence would cause us to reconsider the design?
Success measures should be established before implementation. A credible design process defines both the improvement leadership expects and the findings that would indicate the hypothesis was incomplete or wrong.
Together, these questions move organizational redesign away from whether the future-state chart appears cleaner and toward whether the design is supported by a defensible theory of organizational performance.
Reorganizations are sometimes necessary. Organizations grow beyond their structures. Strategies change. Capabilities need to move. Acquisitions create duplication. New technologies alter work. Customer needs evolve. Management layers become unnecessary. Organizational boundaries that once supported performance can eventually constrain it.
The mistake is not reorganizing.
The mistake is assuming that organizational discomfort is evidence that structure is the problem.
A reorganization changes the visible organization immediately. Whether it changes the operating organization depends on the decisions, governance, processes, capabilities, incentives, relationships, and leadership behaviors surrounding the new structure.
That is why a new organization chart should not be the first artifact of organizational redesign. The first artifact should be a defensible diagnosis of the organizational condition that needs to change and a clear description of what the organization must become capable of doing differently.
Only then should leaders determine which combination of structural and nonstructural interventions is most likely to create that capability.
This leads to the central design principle:
A reorganization is not a solution. It is a hypothesis. Structure should change only when evidence supports the belief that changing it will improve the organization's ability to perform.
The real test is not whether the new reporting relationships were implemented successfully, whether employees transitioned into their new roles, or whether the future-state organization chart looks more logical than the one it replaced.
The test is whether the organizational conditions that constrained performance before the redesign are demonstrably better afterward.
If they are, the redesign has created organizational value.
If they are not, another round of boxes and reporting lines is unlikely to provide the answer until leaders return to the question that should have been asked at the beginning: What is actually preventing this organization from performing the way it needs to perform?
Supporting Research
- Burton, R. M., Obel, B., & HÃ¥konsson, D. D. (2020). Organizational design: A step-by-step approach (4th ed.). Cambridge University Press.
- Galbraith, J. R. (2014). Designing organizations: Strategy, structure, and process at the business unit and enterprise levels (3rd ed.). Jossey-Bass.
- Lawrence, P. R., & Lorsch, J. W. (1967). Organization and environment: Managing differentiation and integration. Harvard Business School Press.
- Nadler, D. A., & Tushman, M. L. (1980). A model for diagnosing organizational behavior. Organizational Dynamics, 9(2), 35–51.
