Second Order Thinking: How Good Decisions Produce Bad Outcomes

Second order thinking is the discipline of asking what happens after the thing you wanted happens. Most bad business decisions were not stupid at the moment they were made. They were correct at the first step and disastrous at the third, and nobody in the room was looking that far.

You cut prices to win share. You win share. Your best competitor matches you within a month, the category resets lower, and now everyone earns less for the same work. Step one worked exactly as designed.

What First Order Thinking Looks Like

First order thinking asks what happens next. It is fast, it feels decisive, and it is right often enough to be dangerous.

Hire a salesperson to increase revenue. Revenue increases. First order confirmed.

What it misses is that the salesperson sells the thing that is easiest to sell rather than the thing with the best margin, that the founder now spends six hours a week managing rather than building, and that customers arriving through discounting churn at twice the rate. None of that appears in the first step. All of it appears by month nine.

The problem is not that people are bad at this. It is that the first order arrives with evidence and the second order arrives later, quietly, and gets blamed on something else.

Second Order Thinking, Defined

Second order thinking is asking: and then what.

Not once. Three times. What happens next, what happens because of that, and what other people do once they notice.

That last part is the one most people skip. First order thinking treats the world as static. But your competitors, your team and your customers all respond to what you do, and their responses are usually the thing that determines the outcome.

A discount is not a price change. It is a price change plus everything your competitors do about it. A new metric is not a measurement. It is a measurement plus everything your team does to move it.

The Measurement Trap

The clearest illustration is a rule named after the psychologist Donald Campbell. Campbell’s law states that the more any quantitative indicator is used for decision-making, the more it will be corrupted, and the more it will distort the thing it was meant to measure.

Watch it operate. You decide to track calls booked, because calls booked correlates with revenue. Within a month your team is booking calls with people who will never buy, because the number is what they are judged on. The metric goes up. The correlation it was built on is gone, because you destroyed it by measuring it.

Nothing dishonest happened. Everyone did exactly what you asked.

This is why second order thinking matters most when you introduce a target. Every number you elevate changes behaviour, and the behaviour change is the second order effect you did not plan.

Four Questions That Force the Second Order

Before any decision that is hard to reverse, answer these in writing. Ten minutes.

And then what? Three times, out loud. Most bad decisions die at the second iteration.

Who responds, and how? Name the competitor, the team member or the customer who reacts, and what they do. If your plan only works when nobody responds, you do not have a plan.

What behaviour does this reward? Especially for any new metric, incentive or rule. People optimise for what is measured, not for what is meant.

What does this make harder later? Every decision closes doors. Name the door. Cheap customers make premium positioning harder. One large client makes saying no harder.

Four questions. They will not make you slower, because the decisions that need them are the ones you are already agonising over.

Where Second Order Thinking Goes Too Far

The failure mode on the other side is real, and it is more common in thoughtful people than the first one.

You can run the chain forever. Every consequence has a consequence, and at some point the analysis stops improving the decision and starts preventing it. Someone who cannot buy a laptop without modelling the third order effects is not being rigorous, he is stuck.

The rule that keeps it useful is proportionality. Run the chain hard on decisions that are expensive to undo, and stop at the first order on decisions you can reverse in a week. That is the same distinction behind the framework for decisions with no way back, and it is what stops a good habit becoming paralysis.

Second order thinking is a tool, not a personality. Applied to everything, it costs more than it saves.

How to Build It Into a Decision

Do not try to become a person who naturally thinks this way. That is a slow and unreliable project. Build it into the process instead.

Put the four questions at the top of whatever document you use to make significant decisions. A page, a note, an email to yourself. The point is that the questions get asked because they are on the page, not because you remembered.

Then write the answers down before you decide, not after. Written predictions are what let you check yourself in six months, and checking yourself is the only thing that reliably improves judgement over time.

And when you set a target, ask the behaviour question twice. Targets are where second order effects do the most damage and get noticed the latest, which is also why an aggressive growth goal needs a filter attached rather than just a number.

The people who look like they have unusually good judgement are rarely smarter at the moment of decision. They have just been asking “and then what” for twenty years, and it has become the shape of how they see a problem.

The clearest everyday version is a fee charged as a percentage, where the first order cost looks small and the twenty year cost is not.

Written By Victor Lanza
Editor, The Executive Insight