The Gap Between What Leaders Say and What Leaders Do

Every organization presents a version of itself to the world. There is the strategy deck, the organizational chart, the operating model, the values on the wall, and the carefully prepared leadership message. Then there is the experience people in the organization have.

That organization is governed by different questions:

• Who is actually listened to?
• Which risks are rewarded—and which are quietly punished?
• What happens to the person who challenges a senior leader?
• Who must approve a decision, regardless of what the organizational chart says?
• Which problems can be discussed openly, and which ones must be managed around?

The first version that gets presented to the internal and external customers is the visible operating system: strategy, governance, processes, technology, structure, data, and metrics. The second version describes the invisible operating system: leadership behavior, trust, communication, accountability, decision-making, motivation, and culture.

Both affect performance, but most transformation efforts begin by examining only the visible system.

That is often where the trouble starts.

Microsoft’s 2026 Work Trend Index puts the problem plainly: “In many cases, people are ready. The systems around them are not.” Microsoft also found that organizational factors—including culture and manager support—accounted for twice the reported AI impact of individual effort alone.

The people may possess the capacity. The technology may possess the capability. The strategy may be sound.

The organization can still be unable to convert any of it into performance.

The theory you state and the theory you operate

In the 1970s, organizational theorists Chris Argyris and Donald Schön developed a useful distinction for understanding why intelligent, well-intentioned people regularly behave in ways that contradict their stated beliefs.

They argued that people operate with two theories of action.

The espoused theory is what someone says they believe and how they describe their intended behavior.

“I want my team to challenge me.”

“We reward intelligent risk-taking.”

“My door is always open.”

The second is a theory-in-use: the set of assumptions that can be inferred from what the person consistently does.

A leader may sincerely ask employees to surface problems early. But if that leader becomes visibly frustrated, dismissive, or punitive whenever bad news arrives, employees learn the real rule.

The espoused theory says, “Bring me problems.”

The theory-in-use says, “Bring me problems only when you already have the solution and preferably before anyone else notices.”

Argyris and Schön’s work matters because the difference is not necessarily dishonesty. People are frequently unaware of the gap between their stated intentions and their observable behavior. But people in the organization often learn the differences quickly.

People do not calibrate their behavior against the values deck, they calibrate it against consequences.

Why more effort rarely closes the gap

Argyris and Schön also distinguished between single-loop and double-loop learning.

Single-loop learning attempts to correct a problem without questioning the rules or assumptions that produced it. Examples:

Output is falling, so leadership adds meetings.

A project is delayed, so reporting requirements increase.

Adoption is weak, so the organization schedules more training.

Double-loop learning asks a more uncomfortable question,

What assumptions caused us to design the system this way in the first place?

Why does output depend on chronic overtime?

Why must the same executive approve decisions that were supposedly delegated?

Why are employees being trained on a process that their managers do not consistently follow?

Most organizations are more comfortable repairing activity inside the existing system than questioning whether the system itself is producing the problem. That is why the visible symptoms survives multiple improvement initiatives, the right questions are rarely asked.

Schein: the same gap at the level of culture

MIT organizational psychologist Edgar Schein examined similar territory at the organizational level.

Schein described culture through three layers:

• Artifacts are the visible expressions of culture, how meetings operate, who speaks, what gets celebrated, how offices are designed, and which stories are repeated.
• Espoused beliefs and values are what the organization says it believes.
• Underlying assumptions are the deeply embedded beliefs that actually govern behavior.

The third layer matters most and is the hardest to see.

An organization may sincerely promote innovation while operating under the assumption that mistakes damage careers. It may publicly support work-life balance while promoting the people who remain visibly available at all hours. It may encourage decentralized decision-making while requiring informal approval from the founder before any meaningful action proceeds.

The espoused value does not automatically change the underlying assumption. When the two conflict, employees usually follow the assumption because that is where the consequences live.

Senge: individual gaps become organizational patterns

Peter Senge brought these ideas into a broader systems perspective.

In The Fifth Discipline, Senge argued that organizational outcomes are not merely the sum of individual strengths and weaknesses. They emerge from structures, assumptions, feedback loops, and repeated patterns of behavior.

An individual leader’s theory-in-use does not remain individual.

It shapes meetings and meetings shape decisions.

Decisions shape incentives and incentives shape what information moves upward.

Over time, the organization develops a coherent pattern that nobody explicitly designed but everyone learns to navigate. That is why replacing one manager, rewriting one value, or delivering one training program rarely changes the full system.

The pattern is often larger than any single participant.

The newer research reaches the same conclusion

Robert Kegan and Lisa Lahey later described how a sincere commitment to change can be blocked by an equally powerful but hidden competing commitment. A leader may genuinely want to delegate more while also believing, often without consciously articulating it, that being indispensable is the source of their value.

Amy Edmondson’s research on psychological safety demonstrates what happens when organizational conditions make candor risky. Problems do not disappear, they become less visible.

Different researchers use different language, but the central warning remains consistent.

You cannot understand an organization only by asking what it believes.

You must also pay attention to what happens when living by that belief becomes inconvenient, politically difficult, or personally costly.

Why this matters before changing strategy or technology

A strategy based only on what leadership says about the organization is built on incomplete data. The same is true of a transformation plan, operating-model redesign, Microsoft Copilot rollout, or leadership initiative.

Technology does not enter a neutral environment. It enters an existing system of authority, trust, habits, incentives, and decision rights.Technology can improve that system and it can also accelerate its contradictions.

• If employees are already unclear about who owns decisions, adding faster communication and AI-generated recommendations may produce more activity without producing more clarity.
• If managers do not trust employees to exercise judgment, giving those employees more technological capability will not automatically create empowerment.
• If people have learned that uncomfortable information is punished, a new dashboard will not make the information more discussable.

The first task is not to prescribe another solution. The first task is to understand which organization is actually operating: the one described in the deck or the one revealed through behavior.

That is the foundation of organizational alignment.

The Aureus Advisory Group perspective

Most organizations already possess significant human, technological, and operational capacity.

The challenge is converting that capacity into measurable performance.

That requires alignment between the visible system the organization has designed and the invisible system its people experience every day.

Aureus Advisory Group helps leaders identify the organizational constraints that may be limiting performance, transformation readiness, adoption, and value realization.

Before another solution is layered onto the wrong problem.

Before beginning you next transformation, make sure you are solving for the organization that actually exists, not the one you think or hope exists.

It is important to understand that we are not just another employee engagement solution. You probably already do that, and if you don't you probably should consider it.

Our role is looking at the full picture with the data you already collect, fill in gaps where appropriate. Then we work with you to implement small itterative changes, remeasure to make sure the changes affected the right behaviors and moved you towards alignment.

We do not replace or augment anything you are already doing, we help you maximize its value.

Schedule a Strategic Conversation

Sources and further reading

Chris Argyris and Donald Schön, Theory in Practice: Increasing Professional Effectiveness.

Edgar Schein, Organizational Culture and Leadership.

Peter Senge, The Fifth Discipline.

Microsoft, 2026 Work Trend Index: Agents, Human Agency, and the Opportunity for Every Organization.