Feedback loops

A system is not just a pile of parts; it is a set of parts wired together so that they influence each other. The wiring that matters most is the feedback loop — a circular chain of cause and effect in which a change eventually loops back to affect the very thing that changed. Once you start seeing loops, much of the puzzling behaviour of organisations, ecosystems, and economies stops looking like a sequence of unrelated events and starts looking like the predictable output of a structure.

Two kinds of loop

There are only two basic flavours, and almost everything else is a combination of them.

A reinforcing loop (also called a positive loop) amplifies change. More leads to more, or less leads to less. Money in a savings account earns interest, which becomes more money, which earns more interest. A growing population has more parents, who produce more children, who become more parents. Reinforcing loops are the engines of growth — and, run in reverse, of collapse.

A balancing loop (also called a negative loop) resists change and seeks a goal. It senses the gap between where the system is and where it “wants” to be, then acts to close that gap. A thermostat compares the room to a target temperature and switches the heating on or off. Your body sweats when too hot and shivers when too cold. Balancing loops are the sources of stability and stubbornness.

A crucial warning about vocabulary: “positive” and “negative” describe direction, not value. A positive loop can be a vicious spiral into bankruptcy; a negative loop can be exactly the stabilising force you want. Do not read them as “good” and “bad”.

Why the labels mislead

Because the names are so easily misread, many practitioners prefer reinforcing and balancing. The test is simple: trace the loop around once. If a nudge in one direction comes back around as a further nudge in the same direction, the loop is reinforcing. If the nudge comes back as a push in the opposite direction, the loop is balancing.

A bank run shows the danger of a reinforcing loop. Worry that a bank might fail prompts people to withdraw their money. Visible queues and falling deposits increase everyone else’s worry, prompting more withdrawals. The belief that the bank is failing can actually cause it to fail. Nothing here is “positive” in the everyday sense, yet it is a textbook positive loop.

Loops combine — and that is where richness comes from

Real systems rarely contain a single loop. They contain several, tugging against each other, with first one and then another in control. Bacteria in a flask multiply through a reinforcing loop, doubling and doubling. But food runs short and waste accumulates, strengthening a balancing loop that slows growth and finally halts it. The visible result is the familiar S-curve: explosive early growth that bends over and levels off as limits bite.

Oscillation comes from the same interplay. When a balancing loop overshoots its target and then overshoots back, you get the swing of a pendulum, the boom-and-bust of commodity prices, or the lurch of an inventory that is repeatedly over- and under-stocked.

Delays: the hidden troublemaker

Loops would be far easier to manage if their effects were instant. They rarely are. Between an action and its visible result there is usually a delay, and delays are responsible for a great deal of bad decision-making.

Picture an unfamiliar shower with a slow response. You turn the tap, feel no change, and turn it further. Suddenly the water is scalding, so you wrench it back, and a moment later it is freezing. The delay between adjustment and response makes you overcorrect in both directions, oscillating around the temperature you actually want. The same pattern appears when managers respond to last quarter’s data, or when a government stimulates an economy that only responds a year later.

The lesson is not to act, but to respect the lag: make smaller moves, wait for the response, and resist the urge to keep pushing while the system is still catching up.

Why this matters for strategy

Events tempt us to fight symptoms one at a time. Loop thinking shifts attention to the structure that generates those events. If a harmful pattern keeps recurring, look for the reinforcing loop driving it and ask how to weaken it, or for the balancing loop that should be holding it in check and ask why it is too weak or too slow. Change the loops, and you change the behaviour — not just this once, but for good.

Feedback loops are the grammar of systems thinking. Master reinforcing loops, balancing loops, and the delays that connect them, and you have the vocabulary you need to read almost any dynamic system you encounter.