The frustrating truth about complex systems is that the most obvious place to intervene is often the least effective.

When something is not working, organisations usually reach for the visible controls. Change the target. Increase the budget. Add more people. Rewrite the process. These interventions can matter, but they often sit at the shallow end of the system. They adjust the numbers without changing the behaviour that produced the numbers in the first place.

Donella Meadows’ insight was that leverage is counterintuitive. The places where a small shift can produce a large change are not always the places people look first. To find them, you need to understand the system well enough to see what is really governing its behaviour.

What are leverage points?

In systems thinking, a leverage point is a place within a complex system where a small shift can produce large changes in the system’s behaviour.

Meadows described a hierarchy of places to intervene in a system. At the lower end are numbers, constants and parameters. These are things like budgets, targets, staffing levels, deadlines and thresholds. They are easy to see, easy to change and often politically attractive. They are also usually low leverage.

Higher up the hierarchy are the rules of the system, the structure of information flows, the feedback loops that drive behaviour, the goals of the system, and the mindsets or paradigms from which the system arises.

That may sound abstract, but it is very practical. If a sales culture rewards individual short-term performance above customer trust, changing the monthly target may not alter the pattern. Changing the rules, feedback and definition of success might.

Why the obvious interventions rarely work

Most organisations intervene at the numbers level because numbers are visible. If a team is overloaded, hire more people. If sales are down, increase the target. If service quality is poor, add a new metric. If a project is late, shorten the reporting cycle.

There are times when those actions are necessary. The problem is assuming they are sufficient.

Take a service delivery team that is missing response-time targets. Leadership adds more people and tightens the performance dashboard. For a while, the numbers improve. Then the old pattern returns. Complaints rise again. Escalations increase. Morale drops.

The numbers-level intervention did not address the deeper system. Cases were being passed between teams because the rules for ownership were unclear. Information about customer history was fragmented. The feedback loop rewarded speed of closure rather than quality of resolution. The system kept producing the same behaviour through different people.

That is why understanding leverage points matters. It helps a group look beyond the easiest control and ask where the system is actually being shaped.

Higher leverage: rules, information flows and feedback

Rules are more powerful than numbers because they define what is allowed, expected and rewarded. A rule does not need to be written down to shape behaviour. Many of the most influential rules in organisations are informal. Do not challenge senior leaders in public. Do not admit uncertainty. Protect your department first. Get sign-off before taking a risk.

Changing those rules can shift behaviour more than changing a target. If a team is told to collaborate but every review meeting rewards individual performance, the formal message and the real rule are in conflict. The real rule usually wins.

Information flows are another important system intervention point. Who knows what, when? What information reaches decision-makers and what gets filtered out? What do frontline teams see that senior leaders never hear? What do customers know that the organisation has stopped noticing?

A change in information flow can alter the system quickly. When a leadership team starts hearing unfiltered customer stories, or when frontline teams can see the consequences of their work downstream, behaviour often changes without a new policy being written.

Feedback loops sit higher still. If a loop is reinforcing the wrong behaviour, small adjustments will struggle. A performance system that rewards heroic recovery after failure may quietly discourage prevention. A culture that celebrates overwork may keep reproducing burnout, even while wellbeing initiatives are introduced. Understanding which loops govern the current behaviour gives the group a better chance of intervening where it matters.

The highest leverage: goals and mental models

The highest leverage points are also the hardest to change. They sit at the level of goals and mental models.

The goal of a system is not always the goal written in the strategy document. The real goal is revealed by what the system consistently optimises for. A business may say it wants learning, but if it punishes failure, the system is optimising for protection. An organisation may say it wants collaboration, but if status and reward are tied to individual ownership, the system is optimising for internal competition.

Changing the goal changes what the system pays attention to. It changes what success means. That is why it is so powerful and why it can be so difficult.

Mental models sit deeper again. These are the assumptions and narratives out of which the system operates. What do we believe people are motivated by? What do we believe customers value? What do we believe leadership is for? What do we believe must never be questioned?

A shift in mental model can change many visible behaviours at once. But it cannot be forced through a slogan. It has to be surfaced, examined and worked with honestly.

How to find leverage points in practice

You cannot find useful leverage points by guessing from outside the system. You need to bring the system into view.

That is where a LEGO Serious Play Build Level 3 system model can be valuable. The group builds the system physically. They identify agents, connect them, explore relationships and test what happens when change enters the model. The model externalises relationships and feedback loops that are often invisible in normal discussion.

Once the system is visible, leverage points become discussable. The group can ask which connection carries the most influence, which rule shapes the behaviour, where information gets blocked, which feedback loop reinforces the current pattern, and what goal the system is really serving.

This kind of work sits close to complexity and sense-making. In complex systems, the right intervention is rarely obvious at the start. The work is to understand enough of the pattern to act carefully, then observe what the system does in response.

For facilitators who want to work at this level, Advanced LEGO Serious Play Facilitation Training develops the Build Level 3 capability needed to guide this kind of system work.

A note on caution

High leverage points are attractive because they can create significant change. They are also high risk.

Intervening at the level of rules, goals or mental models without understanding the system well enough can create large unintended consequences. A change that looks elegant in a workshop may behave differently once it meets politics, history, incentives and pressure.

Good systems thinking work takes time. It does not rush to the clever intervention. It helps a group see the system, test its assumptions, understand its feedback loops and decide where action is most likely to matter.

Leverage is not about doing the most dramatic thing. It is about finding the place where a careful shift can change the pattern. In complex organisational systems, that is usually where real change begins.

The discipline is patience. Look long enough to understand what the system is protecting, rewarding and repeating. Only then does intervention become more than activity. It becomes a considered attempt to change the conditions that keep the current pattern alive.

For leaders and facilitators, that is a different kind of confidence. Not the confidence of having a quick answer. The confidence of knowing where an answer is most likely to matter.