Why some transformation disagreements cannot be resolved with another RACI, steering committee, or escalation path?

In the previous Insight on Decision Integrity, I argued that correct data does not necessarily produce a correct decision.

Between organizational reality and executive action sit several layers of mediation:

Reality → Measurement → Data → Selection → Interpretation → Narrative → Decision

At every transition, distortion can enter methodologically, cognitively, accidentally, or through incentives. But that discussion leaves us with a more uncomfortable situation. Sometimes the data is sufficiently reliable. The interpretation is reasonable. The risks are understood.

Several people in the room may even privately agree about what is happening. And yet the organization still struggles to act.

At that point, the problem may no longer be analytical. It may be structural, psychological, or political.

The important question therefore changes from:

“Do we understand what is happening?”

to:

“What prevents the organization from acting on what it already understands?”

That is where organizational friction deserves closer examination.

Not Every Governance Problem Is a Governance Problem

Organizations frequently respond to persistent disagreement by strengthening governance. Another committee is created. Decision rights are documented. A RACI is revised. Escalation paths are clarified. Additional reporting is introduced.

Sometimes that is exactly what is required.

Governance matters. Effective transformation governance requires clear sponsorship, explicit decision rights, accountability, and forums designed to make decisions rather than merely report status. Gartner, for example, explicitly recommends establishing decision rights and focusing transformation governance on decision-making rather than status reporting.

But stronger governance cannot resolve every form of disagreement.

Sometimes the governance architecture is reasonably clear and the disagreement persists because the participants are responding to different consequences.

Consider a transformation that centralizes a capability previously controlled by regional business units. At enterprise level, the logic may be compelling:

• greater standardization;
• lower duplication;
• stronger controls;
• better technology leverage;
• more consistent customer experience.

But for a regional leader, the same transformation may mean:

• less autonomy;
• reduced budget;
• loss of direct resources;
• greater dependency on another function;
• slower exception handling;
• or diminished influence over an area for which they remain commercially accountable.

Both parties can support “the transformation.” They are simply experiencing different transformations.

That distinction matters.

Organizations Do Not Have Incentives. People and Roles Do.

Corporate language often anthropomorphizes the organization.

We say:

“The business wants…”

“Technology believes…”

“The organization has decided…”

These abstractions are useful, but they can conceal something important. Organizations act through individuals occupying roles.

And those roles carry:

accountability, authority, budget, exposure, incentives, status, relationships, history, and consequences.

Transformation changes more than processes and technology. It redistributes some combination of these things. That means resistance is not necessarily irrational. It can be entirely rational from the perspective of the person or function experiencing the change.

Research on transformation similarly emphasizes that incentives, decision rights, operating models, governance, and accountability have to reinforce the intended change rather than operate independently. McKinsey has found materially stronger reported transformation outcomes where incentives were explicitly aligned with transformation goals, while BCG likewise argues that incentives need to reinforce the decision-making, behavior, and accountability required by the transformation.

The implication is important:

Before diagnosing resistance as behavioral, examine whether the transformation is asking rational actors to behave against the incentives created by the organization itself.

Four Different Disagreements Often Get Called “Governance”

When I encounter persistent cross-functional disagreement, I find it useful to distinguish at least four possibilities. They require different interventions.

1. Information Disagreement

The parties do not possess the same facts. One team knows something the other does not.

For example:

Technology understands that a legacy dependency makes the proposed timeline unrealistic. The business understands that a regulatory deadline makes delay commercially unacceptable. Neither side initially possesses the complete picture.

This is fundamentally an information problem. The intervention is better information exchange.

More accurate data, greater transparency, or a shared fact base may resolve much of the disagreement.

2. Interpretation Disagreement

The parties possess broadly the same information but draw different conclusions from it.

For example:

Both teams agree that delivery confidence is 70%.

One leader interprets that as:

“We have sufficient confidence to proceed.”

Another interprets exactly the same evidence as:

“A 30% downside risk is unacceptable for this dependency.”

This is not an information problem. It is an interpretation problem.

The appropriate intervention may involve clarifying risk appetite, assumptions, thresholds, or competing causal models. This connects directly to Decision Integrity.

3. Structural Disagreement

The organization itself has created overlapping or ambiguous authority.

For example:

• one function owns the platform;
• another owns the business process;
• a third owns the risk;
• a fourth owns the budget;
• and nobody possesses sufficient end-to-end authority to resolve the trade-off.

Now the disagreement is structural.

This is where governance redesign, explicit decision rights, and stronger operating-model definition genuinely matter.

Transformation research repeatedly emphasizes the importance of clear decision rights. McKinsey’s work on operating-model transformations, for example, identifies governance, mandates, roles, metrics, and ownership of critical decisions as foundational design questions.

4. Incentive Disagreement

This is the most interesting category. Everyone may understand the facts. Everyone may understand the decision rights. Everyone may understand the enterprise objective. But the consequences of the proposed decision are materially different for each participant.

One function gains control. Another loses it.

One executive gets the upside if the transformation succeeds. Another carries operational risk if it fails.

One team receives efficiency targets. Another must absorb the additional workload required to produce those efficiencies.

One vendor benefits from committing. Another group carries the consequences of an unrealistic commitment.

Now the disagreement is not principally informational or structural.

It is an incentive disagreement.

Another governance meeting may simply give the disagreement another venue in which to reproduce itself.

Formal Authority and Effective Authority Are Not the Same

This leads to a distinction I consider especially important in operating-model transformation:

Formal authority is who the operating model says decides.

Effective authority is who the organization actually waits for.

The two are not always the same.

Imagine an operating-model redesign formally transferring a decision from a global functional leader to a newly established regional leader.

• The organization chart changes immediately.
• The RACI changes immediately.
• The governance document changes immediately.
• Human behavior may not.
• Employees may continue seeking informal confirmation from the previous leader.
• The previous leader may continue intervening.
• Peers may continue treating the former authority as the real decision-maker.
• The new leader may possess formal accountability while remaining psychologically dependent on someone else for legitimacy.
• The organization has changed structurally without yet changing socially.

This creates an important diagnostic:

When a difficult decision arises, who do people actually wait for before acting?

That individual or forum may possess more effective authority than the governance model recognizes. And when formal authority and effective authority diverge for long enough, accountability becomes unstable.

You can therefore change accountability in PowerPoint much faster than you can change perceived authority inside an organization.

The Geography Problem: Integration Is Not Replication

This becomes particularly visible when organizations establish or integrate a new geographical location.

• On paper, integration can appear complete.
• Reporting structures exist.
• Governance forums operate.
• Processes have been replicated.
• Roles have been assigned.
• Technology is available.
• The location is officially part of the global operating model.

But organizational integration occurs at several levels simultaneously.

• There is process integration.
• There is decision integration.
• There is relationship integration.
• There is trust integration.
• There is also identity integration.

A central organization may interpret local deviations as unwillingness to conform.

The local organization may interpret central standardization as insufficient sensitivity to operating reality.

Headquarters may believe authority has been delegated.

The local team may believe every consequential decision still requires implicit approval from headquarters.

Both sides can technically comply with the same operating model while living inside very different psychological interpretations of it.

This is why geographical integration cannot be reduced to replicating roles, processes, and governance.

You are integrating systems of authority and trust.

That takes longer.

The Socially Expensive Truth

There is another question I believe senior leaders should ask more often:

What is difficult to say in this room even though several people probably believe it?

This question can reveal more about organizational health than another status report. The answer might be:

“The timeline is no longer credible.”

Or:

“The operating model has not actually been accepted by this function.”

Or:

“We have assigned accountability without sufficient authority.”

Or:

“The vendor is saying yes because the commercial consequences of saying no are too high.”

Or:

“The initiative is still alive primarily because acknowledging failure would be organizationally expensive.”

The problem in these situations is not necessarily an absence of knowledge.

It is the social cost of expressing knowledge.

Research on psychological safety is relevant here. Psychological safety does not mean absence of challenge or accountability; it concerns whether individuals believe they can raise concerns, questions, or dissent without interpersonal punishment. Recent HBR discussions of the research similarly emphasize that senior-team decision quality deteriorates when people conclude that candour carries unacceptable consequences.

For transformation leaders, this produces an important distinction:

Unknown truth

versus

Known but unspoken truth.

They require very different interventions. Better reporting may help the first. It does almost nothing for the second.

When the Room Manufactures Certainty

One situation deserves particular attention because it occurs frequently in large delivery environments: the pressure placed on vendors or delivery teams to make commitments under uncertainty.

Imagine a program review.

• A critical delivery date is under pressure.

• The executive wants certainty.

• The program leader wants a commitment.

• The vendor knows several technical uncertainties remain unresolved.

• The vendor also understands that saying: “We cannot responsibly commit yet”, may be interpreted as lack of confidence, insufficient ownership, weak partnership, or commercial resistance.

The questioning continues.

• “Can you do it?”
• “We’ll need to validate…”
• “But can you commit?”
• “There are dependencies…”
• “We understand that. Can you commit?”

Eventually:

“Yes.”

The meeting has produced a date.

But has it produced information?

Not necessarily.

Sometimes the commitment is no longer a prediction.

It is:

the socially acceptable answer produced by the room.

This phenomenon creates what I think of as manufactured certainty. The organization begins with genuine uncertainty. Governance applies pressure. The uncertainty is converted into a precise date.

• That date enters the plan.
• The plan enters the dashboard.
• The dashboard gives the date institutional legitimacy.

And several weeks later, leadership asks:

“Why did the vendor miss its commitment?”

A more uncomfortable question might be:

“What conditions existed when we obtained that commitment?”

Commitment, Forecast, Aspiration, and Negotiating Position

These terms are often used too loosely. They should be separated.

A forecast

“Based on what we currently know, this is our best estimate of what will happen.” A forecast contains uncertainty.

A commitment

“We are accepting accountability for delivering this outcome.” That is stronger.

An aspiration

“This is the outcome we want the organization to stretch toward.” Useful, but different again.

A negotiating position

“This is the outcome I am asking another party to accept.” That is not evidence about feasibility.

Problems begin when organizations convert one into another without acknowledging the transition.

• An aspiration becomes a target.
• The target becomes a commitment.
• The commitment becomes a forecast.
• The forecast enters executive reporting as though it were an empirical fact.

When the distinction disappears, certainty increases linguistically while the underlying evidence remains unchanged.

This is another Decision Integrity problem, but now driven by organizational incentives.

Politics Is Not Necessarily Dysfunction

The word politics deserves some care.

Organizational politics is often discussed as though it refers exclusively to manipulation, empire-building, or bad behaviour.

That definition is too narrow.

Whenever resources are scarce, priorities compete, authority is distributed, and outcomes affect different groups differently, political behaviour will exist.

That is not necessarily evidence of organizational pathology.

It is partly a consequence of organizing human beings around competing responsibilities.

The objective should therefore not be:

“Remove politics from the organization.” That is unrealistic.

A more useful objective is:

Make material incentives, trade-offs, and power dependencies sufficiently visible that they can be governed rather than disguised.

This is where sophisticated leadership differs from procedural governance.

Procedural governance asks:

Who owns the decision?

Sophisticated governance also asks:

What will this decision do to the people expected to support it?

Five Questions Before Calling It a Governance Problem

When disagreement persists, I would encourage leaders to work through five questions in sequence.

1. Are we working from the same information?

If not, resolve the information gap first. Do not diagnose politics where simple information asymmetry explains the disagreement.

2. Are we interpreting the same information differently?

If so, expose assumptions. Ask what risk thresholds, causal models, or prior beliefs create the difference.

3. Are decision rights genuinely clear?

Not merely documented.

Ask:

Who can actually make this decision without seeking informal permission elsewhere?

If the answer is unclear, the problem may genuinely be governance.

4. Do the stakeholders experience materially different consequences?

Ask what each participant gains, loses, inherits, controls, or becomes accountable for if the proposed decision proceeds.

This is the incentive question. And it often explains behaviour that otherwise appears irrational.

5. What truth has become professionally expensive to say?

This is the question I would ask last.

Because once facts, interpretations, structures, and incentives have been examined, silence itself becomes informative.

• What would someone say if there were no reputational consequences?
• What would the vendor say if the contract were not at risk?
• What would the delivery lead say if the original business case did not need defending?
• What would the regional leader say if disagreement with headquarters were not interpreted as lack of alignment?

Sometimes the most important information is not missing. It is being withheld by the environment.

Governance Should Surface Friction, Not Merely Contain It

This produces a different view of governance.

Good governance should not eliminate disagreement. In complex transformation, disagreement is inevitable. Its purpose should be to make disagreement diagnosable.

• Is this a fact problem?
• An interpretation problem?
• A structural problem?
• An incentive problem?
• An authority problem?
• A capacity problem?

Only once the disagreement is correctly classified can the organization select an appropriate intervention.

Otherwise, organizations frequently respond to every form of friction with the same instruments:

• more escalation;
• more meetings;
• more reporting;
• more governance;
• more documentation.

And eventually governance becomes another layer through which the underlying disagreement travels.

Returning to TRACE

This brings the first three Insights together.

TRACE asks:

Does the transformation narrative remain connected to organizational reality?

Decision Integrity asks:

Can we trust the path through which reality becomes an executive conclusion?

And Organizational Friction asks:

Even when reality is understood, what prevents the organization from acting on it?

Taken together, the sequence is:

1) ORGANIZATIONAL REALITY
  ↓
Can we see it?
  ↓
2) TRACE
  ↓
Can we interpret it with integrity?
  ↓
3) DECISION INTEGRITY
  ↓
Can the organization act on what it knows?
  ↓
4) ORGANIZATIONAL FRICTION

• Technology remains essential throughout.
• Processes remain essential.
• Governance remains essential.

But organizations ultimately transform through people occupying roles inside systems of incentives, authority, relationships, identity, and consequence.

That is why I remain cautious when organizational difficulties are described simply as the “people side” of transformation.

People are not a peripheral dimension surrounding the transformation. They are the mechanism through which every other dimension becomes operational.

And sometimes the most useful question in the room is not:

“What does the governance model say?”

It is:

“What is actually preventing an intelligent person in this system from doing what the governance model says they should do?”

The answer to that question often tells us where the real transformation work begins.