Deciding Fast Is Not the Opposite of Deciding Well

It arrives on a Thursday morning. A supplier fails, or a senior person resigns without notice, or a client puts something in writing that has to be answered by close of business. You have hours. You do not have a week.

And you notice your own thinking change. Not slow down, exactly. It gets louder. Options multiply without ordering themselves. You reach for a spreadsheet you know will not help. You call someone whose view you can already predict. Somewhere in the middle of this you form the sentence that will be your explanation afterwards, whichever way it goes: if I'd had more time, I'd have got this right.

You would not have. Given a week you would have deliberated for six days and decided on the seventh, under the same compression, missing the same thing you are missing now. The deadline is not what removed your judgement. It only revealed that the judgement had nowhere to stand.

Why does judgement collapse when the clock starts?

Under time pressure the mind does not become generally worse at reasoning. It narrows. Attention constricts onto whatever is most salient, and the slowest part of the process is the first to go.

The slowest part is comparison. Holding several courses of action side by side, projecting each one forward, weighing them against each other — that is expensive, and it is what compression takes away first. What remains is whatever is nearest to hand. The option already in the room. The suggestion made by the most confident voice. The thing you did last time, which was not the same situation, though under load it looks close enough.

So the felt experience is not a shortage of thoughts. It is a shortage of any way to rank them. And ranking is not something you can improvise while the room waits.

Is deciding quickly actually worse than deciding slowly?

The assumption sitting underneath the panic is that speed and accuracy trade against each other at a steady rate — that every extra hour of deliberation buys a measurable increment of correctness.

Past a shallow point, it does not. Extra time mostly buys more of what you already had. More considerations, more caveats, one more person consulted, and a great deal more room for the private negotiation that feels like analysis and is not. The curve flattens early, and after it flattens you are paying in hours for nothing.

The clearest evidence for this comes from people whose deadlines are measured in seconds. In the mid-1980s Gary Klein and his colleagues set out to document how experienced fireground commanders weighed their options inside a burning building. Twenty-six commanders, averaging twenty-three years of service, were taken back through 156 real decision points. The researchers expected to find option comparison under extreme constraint. They found almost none of it. In fewer than twelve per cent of those decision points was there any evidence of two or more options being held up against each other at all.

What the commanders did instead was recognise. The situation registered as an instance of a type they already knew. A course of action came with the recognition. They ran it forward mentally, and if nothing in that simulation broke, they gave the order. One option, considered once, acted on.

This is not carelessness dressed up as expertise. The comparison work had been done — years earlier, across hundreds of fires, and stored as pattern. By the time the clock started, it was no longer decision-making. It was retrieval.

What do people who decide well under pressure do differently?

They arrive with the decision partly made.

There are two separate pieces of work that both go by the name of deciding, and confusing them is the whole of the problem. The first is general: what matters here, what threshold triggers what response, who holds authority over what, which risks are acceptable and which are not. That work is slow, but it is cheap, because it can be done in a quiet week when nothing is on fire and nobody is waiting.

The second is specific: which case is this, and does the rule reach it. That work is fast by nature, and it is the only part that genuinely has to happen under load.

If none of the first has been done, all of it arrives in the second. You end up asking, in ninety minutes with a client on hold, what your actual risk tolerance is — a question that deserved a quiet afternoon and never got one. That is what the panic is. Not a failure of nerve. A general question being asked for the first time at the worst available moment.

Why doesn't "stay calm and think clearly" work?

Because calm is an output, not a lever, and the advice mistakes one for the other.

Look at what the standard counsel actually asks of you. Breathe. Sleep on it. Do not decide while angry. Get a second opinion. Write down the pros and cons. Every item requires spare capacity to initiate a process, and spare capacity is precisely what the compression has taken. Sleep on it assumes a deadline that permits sleep, which is the exact condition that does not hold. The advice describes what deciding well looks like from the outside and calls the description a method.

The second opinion deserves particular scepticism. Under compression, consulting someone is rarely a way of thinking harder. It arrives phrased as a request for information and functions as a request for permission — a way of making the decision jointly owned before it has been made at all. You can tell which one is happening by whether you would accept an answer you did not want.

How do you decide fast when there is genuinely no time?

Three things are available in the moment, and each of them works by having been fixed beforehand.

Name the decision type before you hunt for the answer. Is this reversible, and if so, at what cost? A reversible decision that is cheap to undo should be made immediately, because the price of being wrong is one further decision, and deliberation costs more than that. Most decisions that feel urgent are reversible. Almost none of them get treated that way, because urgency and consequence feel identical from the inside and are not the same thing at all.

Fix the threshold before you look at the number. Ask what would have to be true for each course of action to be correct, and set that condition first. Once you have seen the figures, every threshold you set afterwards is contaminated by them, and you will find yourself building a case rather than testing one. This costs about four minutes and it is the difference between deciding and justifying.

Then stop at the first workable option, and mean it. The commanders in Klein's study were not searching for the best course of action; they took the first that survived contact with the situation. Under pressure the instinct runs the other way — the stakes feel like they demand a thorough search, so the search continues. But a search with no stopping rule does not conclude. It gets interrupted by the deadline, and a decision interrupted by a deadline carries the cost of the deliberation and none of its benefit.

None of these three is a technique to summon in the moment. Each one is a structure that either exists before the Thursday morning or does not. That is the only variable that separates the people who decide well under compression from the people who describe themselves afterwards as having been unlucky with the timing.

This is what Decision Throughput is built for: the acute case, where the queue has already outrun your capacity and the deadline is not negotiable. It does not teach composure. It installs the thresholds and stopping rules that make composure unnecessary, so that when the clock starts the general questions are already answered and only the specific one remains.

The next compressed decision is already on its way toward you. You do not know what it will be about, and you do not need to. You need to know what you will do when it arrives, and that is a question with a quiet afternoon in it, not a Thursday morning.