Can you train decision making under uncertainty?

Yes. Decision making under uncertainty is a behaviour, and behaviour is built through practice under real stakes, not through lecture slides. Serious games place managers in a pressured environment with incomplete information and live consequences, then use a structured debrief to surface how they decided. Repeating that loop builds the habits of separating process from outcome, triaging, and acting on partial data.

Decision Making Under Uncertainty: A Manager's Guide

Q: What is decision making under uncertainty?

Decision making under uncertainty is choosing a course of action when you do not have complete information and cannot reliably predict the outcome. Instead of waiting for certainty that never arrives, you judge the quality of the decision by the process used to make it: the information gathered, the options weighed, and the reasoning applied, not by whether the result happened to go your way.

Q: How is a good decision different from a good outcome?

A good decision is one made with a sound process given the information available at the time. A good outcome is a favourable result. Because chance plays a role under uncertainty, a good decision can still produce a bad outcome, and a reckless decision can get lucky. Managers should evaluate and reward the process, not just the result, or they will punish good judgement and reinforce luck.

Q: What is the difference between a reversible and an irreversible decision?

A reversible decision can be undone at low cost if it turns out wrong, so it should be made fast with limited information. An irreversible decision is costly or impossible to undo, so it warrants more analysis, more input, and slower deliberation. Most workplace decisions are reversible, and treating them as if they were permanent is a common cause of paralysis.

Q: When should you stop analysing and make the decision?

Stop analysing when more information would not change your choice, when the cost of delay exceeds the value of extra certainty, or when you have reached the point of diminishing returns where each new data point barely moves your confidence. For reversible decisions, set a short deadline and decide. For irreversible ones, define in advance what evidence would be enough.

Q: What is optionality in decision making?

Optionality is structuring a decision so you keep multiple futures open rather than betting everything on one prediction. Instead of forecasting exactly what will happen, you make moves that pay off across several scenarios, run small experiments, and delay commitment on the costly, irreversible parts until you have learned more. It is preparing for a range of outcomes rather than predicting one.

Short answer: you are being asked to choose without enough information, and no amount of waiting is going to fix that. So stop trying to remove the uncertainty and start managing it. Judge the decision by how you decided, not by how it landed. Move fast on the calls you can undo and slowly on the ones you can't. Keep options open. And learn to spot the moment when more analysis has stopped helping and you are just delaying.

Small group playing a card game together at a casual table

Key takeaways

Principle	What it means	Practical move
Process over outcome	A good decision can still lose	Review how you decided, not just what happened
Reversible vs irreversible	Most calls can be undone cheaply	Decide fast if reversible, slow if not
Optionality	Prepare for many futures, not one forecast	Keep options open, commit late on costly moves
Stop point	More data eventually stops helping	Decide when extra info will not change the choice
It is a behaviour	Judgement is built by practice	Rehearse decisions under real stakes

Why decisions are hard when the information is incomplete

The discomfort here is not a lack of intelligence. It is structural. You are being asked to spend money, credibility and other people's Saturdays on a future you cannot see. Nobody's brain enjoys that.

So we reach for the two escape hatches. Hatch one: gather more data and hope certainty turns up. It won't, because most business questions are not knowable in advance, only estimable. Hatch two: wait. For someone more senior, for the picture to clear, for the call to become obvious.

Both are decisions. They just wear the costume of patience.

And there are two different things hiding inside the word uncertainty, which is where a lot of bad calls come from. Some unknowns are risk: you can put rough odds on it, because you have launched things before and you know roughly how launches go. Others are genuine ambiguity: you cannot even list the possible outcomes, because nobody in your company has ever gone near this.

Risk rewards calculating. Ambiguity rewards small bets and fast learning, because the only way to shrink it is to poke it and see what happens. Demand a precise forecast in a situation that is genuinely ambiguous and you will still be forecasting next March.

Separate the decision from the outcome

This is the big one, and almost every management culture gets it backwards.

A decision is the choice you made with what you knew at the time. An outcome is what happened afterwards, which is your choice plus a large helping of luck you had no say in. Since chance is in the mix, all four boxes exist. Good decision, good result. Good decision, bad result. Reckless punt that happened to work. Reckless punt that got what it deserved.

Judging the decision purely by how it turned out has a name. Resulting. And it does quiet damage.

Punish a sound call that lost to bad luck and you have taught your team never to take a smart risk again. Celebrate a reckless punt that happened to land and you have taught them the punt was wisdom. Do both for two years and you have a team that is either timid or lucky. Never consistently good.

The fix is dull and it works. When the result lands, go back and ask what was actually known at the moment of the call, what options were on the table, and whether the reasoning held up given that.

Which means a good review can end with: "we decided well, we still lost, do the same thing next time." Try saying that sentence in an outcome-only culture. It is almost impossible... and it is the exact sentence that builds people who decide well.

Reversible vs irreversible: match the speed to the cost of being wrong

Not every decision deserves the same care, and pretending otherwise is where both paralysis and recklessness come from. One test: if this is wrong, how expensive is it to undo?

A reversible one, a two-way door, costs almost nothing to walk back. A new meeting format. A vendor for a small pilot. Restructuring a report. If it flops you reverse it and you have lost a fortnight. These should be decided fast, on limited information, by whoever is closest to the work. Spending a week analysing something you could undo in an afternoon is not rigour, it is waste with good posture.

An irreversible one, a one-way door, is expensive or impossible to take back. Letting someone go. Signing a long lease. Killing a product line. Promising something to a client in public. These earn the slow version, more voices, a higher bar of evidence.

And here's the failure I see most. Managers agonise over the two-way doors and then sprint through the one-way doors because by then everyone is tired of deciding things. Sort the door before you sort the answer.

Optionality: prepare for several futures instead of betting on one

If you cannot tell which future is coming, the answer is not a better prediction. It is standing in a way that survives several of them.

In practice that is four habits. Run small cheap experiments instead of one enormous bet, so each one buys information and costs you very little. Stage your commitments in tranches tied to what you learn. Delay the irreversible bits of the plan as long as you sensibly can, because every week of waiting is a week of cheaper information. And prefer moves that pay off across a few scenarios over the one move that only wins if your forecast is exactly right.

Now, this is not indecision with a nicer name. Indecision keeps everything open forever and commits to nothing at all. Optionality commits happily and quickly to the cheap reversible stuff that generates learning, and deliberately holds the expensive doors open until the fog thins.

The aim is to be the team still standing, and still able to move, when the future finally shows its hand.

When to stop analysing and act

Every decision has a moment where analysis stops helping and starts costing you. Spotting it is its own skill. Three tells.

More information would not change your answer. Ask yourself honestly what the next report would have to say to make you decide differently. If the answer is "nothing that's realistically in it", you already decided. Stop.
Waiting now costs more than certainty is worth. Delay has a price. Missed windows, idle people, someone else moving first. When that price goes above what extra confidence buys you, deciding with gaps is the rational move.
The returns have gone flat. The first two hours of digging move your confidence a lot. Hour nine moves it by almost nothing. At that point you are not learning, you are polishing.

Two things that help. On reversible calls, set a short hard deadline up front and decide when it hits, because being wrong is cheap. On irreversible ones, write down in advance what evidence would be enough, so you cannot quietly move the goalposts when the moment arrives.

Deciding at 70% and adjusting as reality corrects you beats waiting for a 95% that was never coming.

Decision making is a behaviour, so build it under real stakes

Now the awkward truth about everything you just read. All of it fits on a slide. All of it is forgotten by lunch.

Because deciding under uncertainty is not knowledge, it is a behaviour. Nobody changes how they decide by being told the principles. People change by deciding badly, somewhere the consequences are visible, with someone holding up a mirror straight afterwards.

That is the entire case for serious games. Put managers in a pressured situation with deliberately incomplete information, real interdependence and live consequences, then debrief what they actually did rather than what they would say in an interview.

And you see it all. Who froze waiting for certainty that was never arriving. Who agonised over a two-way door. Who owned a bad outcome that came from a good decision, and who quietly moved away from it. That is the Put The Player First framework doing its job: their behaviour is the curriculum, the game only exists to make it visible.

Chaos in the Kitchen is the one built for exactly this. Teams run a kitchen while orders arrive faster than anyone can fill them and the information is always half-missing. Strategy, adapting, keeping options alive, deciding what to drop, and then owning the calls in the debrief. It is this whole page as a felt thing rather than a theory.

If your team's problem is more about moving on very thin data, Planetfall is the better fit. A crew on an unknown planet, surviving by reading a place nobody understands yet and committing before the facts are in. Chaos in the Kitchen is about triage and ownership. Planetfall is about deciding, then re-deciding, without treating that as failure.

Both sit inside the wider serious games for leadership development idea. Stop lecturing managers about judgement. Put their judgement on the line where everyone can see it. Then debrief properly. That is how a principle turns into a habit.

Get decision and facilitation playbooks by email

One short email when I have something you can actually use. A new game, a debrief move that worked, an honest look at a real session. No schedule, no filler.

Frequently asked questions

Basics

What is decision making under uncertainty?
Choosing when you don't have the full picture and cannot reliably predict what happens next. You stop waiting for certainty and start judging your own process instead: what you gathered, what options you weighed, whether the reasoning held.

How is a good decision different from a good outcome?
A good decision is sound reasoning on what you knew at the time. A good outcome is just a nice result. Luck sits between the two, so a good call can lose and a reckless one can win. Judge the call.

Methods

What is the difference between a reversible and an irreversible decision?
A reversible one can be undone cheaply, so decide it fast on limited information. An irreversible one is expensive or impossible to take back, so it earns the slow treatment. Most workplace decisions are reversible, and treating them as permanent is where paralysis comes from.

What is optionality in decision making?
Setting things up so several futures stay open instead of betting everything on one forecast. Small experiments, staged commitments, irreversible moves held back until you have learned more, and a preference for choices that pay off in more than one scenario.

When should you stop analysing and make the decision?
When more information wouldn't change your answer. When waiting costs more than certainty is worth. Or when each new data point moves you by almost nothing. Deadlines for the reversible calls, a written evidence bar for the irreversible ones.

Outcomes

How do I stop my team getting stuck in analysis paralysis?
Sort the door first. Push the reversible calls down to whoever is closest and give them a short deadline. Save the deep work for the genuine one-way doors. Nearly all paralysis is people treating a small undoable decision as if it were permanent.

Can you actually train this?
Yes, because it is a behaviour and not a body of knowledge. Put people somewhere pressured with incomplete information and real consequences, then debrief how they actually decided. Run that loop a few times and the habits form.

Logistics

Which game is best for this?
Chaos in the Kitchen, mostly. Running a kitchen under pressure forces strategy, optionality, triage and owning your calls. Planetfall is the better pick when the focus is learning agility and acting on very thin data.