Control and capacity
Control is visible. It has artifacts: approval chains, status meetings, escalation paths, permissions, dashboards, titles, sign-offs. It gives the impression that the organization is being held together by responsible people. When something goes wrong, control offers an immediate move: tighten the process, narrow discretion, require more visibility, centralize authority.
Capacity is quieter. It shows up as the ability of the system to get useful work done without constant force. People understand intent. Teams resolve ambiguity close to the work. Decisions move at the right level. Feedback travels quickly. Mistakes are noticed early. Coordination does not require heroic leadership.
Russell Ackoff's distinction between "power over" and "power to" is useful because it separates these two things. Power over is the ability to command, constrain, approve, or prevent. Power to is the ability to accomplish something. The two are often treated as if they rise together: give someone more authority over people, and they should be able to get more done. But in real organizations, they can move in opposite directions.
A leader can gain more power over a team and lose power to produce results.
Cooperation, not obedience
This happens because most meaningful work depends on cooperation, not obedience. Obedience can make people comply with a procedure. It can make them attend the meeting, fill out the template, wait for approval, and avoid visible mistakes. But it cannot make them care about the outcome, surface uncomfortable information early, improvise intelligently, or coordinate generously with another team. Those behaviors require trust, judgment, context, and commitment.
This is why control-heavy organizations often become surprisingly weak. From the outside, they look managed. There are clear owners, strict processes, formal reviews, and executive visibility. Yet the actual capacity of the organization declines. People stop making decisions because decisions are punished when wrong and invisible when right. Teams optimize for approval rather than outcomes. Information is shaped for safety before it is shared upward. Work slows because the people closest to reality are not authorized to respond to it.
Command can compel motion. It cannot manufacture contribution.
The organization has increased power over its parts while reducing the system's power to perform.
When frustration rises
This distinction matters most during periods of frustration. When execution feels unreliable, leaders naturally reach for more control. A missed deadline becomes a new reporting ritual. A bad decision becomes a new approval gate. A cross-functional failure becomes a new committee. Each move feels reasonable in isolation. Each one is meant to prevent a recurrence.
But the cumulative effect can be a system that is harder to move.
The deeper question is not "Who should have more authority?" It is "What would increase the organization's ability to accomplish the thing it wants?" Sometimes the answer is clearer ownership. Sometimes it is a tighter standard. Sometimes it is a constraint. But often the answer is the opposite: remove a gate, clarify intent, shorten the feedback loop, let the team closest to the facts decide, or replace supervision with a shared definition of success.
Power to is created by designing conditions in which capable people can act effectively.
Designing interactions
That requires a different theory of leadership. The leader is not primarily a controller of people. The leader is a designer of interactions. Work breaks down less often because one part is individually bad than because the interactions among parts are poorly designed: handoffs are unclear, incentives conflict, timing is mismatched, feedback arrives too late, or authority sits far away from information.
If performance lives in interactions, then power over the parts is a weak substitute for power to improve the whole. You can pressure every department to perform better and still make the company worse if the pressure causes each group to defend itself locally. Sales promises what delivery cannot support. Product ships what support cannot explain. Finance blocks what operations needs to stabilize. Everyone is "accountable," but the system is incoherent.
More pressure does not fix incoherence. It often intensifies it.
Generative leadership
The same principle applies to planning. Many organizations plan by listing what they do not want: fewer delays, fewer mistakes, fewer escalations, less waste, less confusion. Those are valid frustrations, but removing what is unwanted is not the same as creating what is wanted. A company can reduce complaints and still fail to build a system with greater capacity.
Power over is naturally drawn to deficiency removal. It asks: What went wrong? Who owns it? What rule will prevent it? Power to asks a more generative question: What condition are we trying to create, and what interactions would make that condition normal?
That question changes the work. Instead of adding approvals because decisions were bad, you might ask what information decision-makers lacked. Instead of requiring more status meetings because projects slipped, you might ask why risk became visible so late. Instead of centralizing authority because teams made inconsistent choices, you might ask whether the strategy was concrete enough for local judgment.
The goal is not to romanticize autonomy. Autonomy without context is just distributed confusion. The point is that authority should increase the system's ability to act, learn, and adapt. When authority mainly increases the ability to stop, inspect, or override, it may create the feeling of leadership while weakening the work.
A useful test is this: after a leadership intervention, do people have more ability to produce the desired outcome, or do leaders merely have more ability to monitor and constrain them?
If the answer is mostly monitoring, the intervention may be power over disguised as improvement.
Closing contrast
The best leaders are not powerless. They use authority. They set direction, define standards, allocate resources, make hard calls, and intervene when the system is stuck. But they do not mistake domination for effectiveness. They understand that the highest form of power is not needing to force everything. It is building a system where the right things become easier to do.
Power over asks, "How do I make people do what I want?"
Power to asks, "How do we make the work capable of producing what matters?"
The first question produces compliance. The second produces capacity. And in the long run, capacity is the only kind of power an organization can actually use.