What is leverage point?

A leverage point is a spot in a system where modest effort changes the outcome far more than effort elsewhere. Finding one saves money and time, because pushing on the wrong part of a system usually achieves little or makes things worse. The idea is associated with the systems thinker Donella Meadows.

How it works

Leverage points differ widely in strength. Adjusting a number, such as a tax rate or a buffer size, is easy but often weak, since the surrounding feedback loops absorb it. Changing the structure is stronger: adding a feedback loop, shortening a delay, or changing who sees which information. The strongest changes alter the goals or the rules of the system. People tend to push hardest on the weakest levers, and sometimes push them in the wrong direction. Testing candidates on a model before acting is a cheap way to tell them apart.

Counterintuitively, the strongest leverage points are often found where people least expect them, and may even feel wrong at first. Teams therefore benefit from checking a surprising candidate with a model before dismissing it.

Example

Say a clinic has long waiting times and considers three changes. Adding one more nurse raises capacity by 5 percent, which trims the wait slightly. Moving paperwork to the patient's home before the visit cuts each visit by 4 minutes out of 20, which raises capacity by 25 percent. The second change is the stronger lever, even though it looks less direct. This is an illustration with round numbers.

How to model it in Machinations

Build the system with Pools, Sources, Drains and state connections, and expose the candidate changes as Levers. Play the diagram once for each candidate and compare the same chart across Plays. The change that moves the outcome most per unit of effort is the leverage point.