Thinking in scenarios: how to see consequences before they happen

A decision gets made. It's well-reasoned, backed by data, and everyone in the room nods along. Six months later, the result is the opposite of what was intended.

This pattern is more common than most people assume, and it rarely comes down to bad decisions in themselves. It comes down to decision-makers thinking in a straight line – "we do X, therefore Y happens" – when reality actually consists of people, incentives, and countermoves that respond to X in ways no one accounted for.

Scenario thinking is the countermeasure. It isn't about predicting the future; it's about systematically asking: if we do this, how might different actors reasonably respond, and which of those responses lead somewhere other than where we intended?

Why a straight line isn't enough

Most decision processes rest on an implicit model: intervention leads to effect. Raise the price, sell less. Introduce a rule, get more of the desired behavior. That model is rarely entirely wrong, but it's almost always incomplete, because it misses how people adapt their behavior to new rules rather than simply following them.

Scenario thinking means moving from a line to a tree. Instead of asking "what happens if we do X," you ask "what are the three or four most plausible ways people, competitors, or the market might respond to X, and what does each of those lead to." It's uncomfortable, because it forces scenarios into view where the original plan doesn't work. That discomfort is exactly what makes the method valuable.

The cobra effect: when the fix becomes the problem

One of the clearest examples of what happens when that step gets skipped is known as the cobra effect. The term comes from a story about British colonial administration in India, which offered a bounty for dead cobras to reduce the snake population in Delhi. Locals responded by breeding cobras specifically to collect the bounty. When authorities discovered this and cancelled the reward, the breeders released their now-worthless snakes – leaving the city with more cobras than before the program began.

The term is now used more broadly for any situation where an action creates an incentive that undermines its own purpose. The pattern is the same every time: a rule targets a symptom, people find the cheapest way to satisfy the rule on paper, and that method happens to make the underlying problem worse.

The Mexico City example

One of the better-known modern examples is Mexico City's traffic program "Hoy No Circula" ("Today you don't drive"), introduced in the late 1980s to cut congestion and air pollution. The principle was simple: based on the last digit of a car's license plate, each vehicle was banned from driving one day a week.

The effect on paper seemed obvious: one less day of driving per car, less congestion, cleaner air. What actually happened was that many households who could afford it bought a second car, often an older, cheaper model with worse emissions, purely to have something to drive on the day the original car was banned. The number of cars in the city increased, and because the newly purchased cars were often older and dirtier than average, air quality didn't improve to the extent hoped for. The rule was followed to the letter – and the underlying problem it was meant to solve remained, in some respects worse than before.

What was missing from the planning wasn't data or good intentions. What was missing was the question: if we make this mandatory, what's the cheapest way for a driver to satisfy the rule without satisfying its purpose? That question is the core of scenario thinking, and it's exactly the kind of reasoning that could have surfaced the second-car scenario before the program launched.

Worth being upfront about: the second-car hypothesis is the most widely repeated explanation for why Hoy No Circula didn't deliver the air quality improvements hoped for, but the evidence isn't settled. Only a handful of the studies that have examined the program test the hypothesis directly, and the support is mixed rather than confirmed. The example is included here not as airtight proof, but to illustrate the pattern itself – how a rule that's followed to the letter can still miss its purpose if the people it targets have a cheaper way to adapt.

Why this is a strategy question, not just a policy question

It's tempting to read the examples above as curiosities from urban planning and colonial history. But the same pattern shows up in boardrooms, sales organizations, and product decisions every week: a bonus model that optimizes for the wrong behavior, a KPI that becomes the goal itself instead of a measure of the goal, a cost cut that moves the cost somewhere else in the organization instead of removing it.

What separates organizations that avoid this from those that don't is rarely intelligence or good intentions. It's a habit: before a decision is finalized, ask how the people affected will reasonably adapt – and assume that adaptation will happen at the lowest possible cost to them, not the effect you were hoping for.

The cobra effect as a phenomenon deserves its own, deeper treatment – why it emerges, how to spot it before it hits, and more examples from business. That article is next in this series.