The Best Time to Make a Decision Is Before You Need It

The advice says make your important decisions in the morning. The decision arrives at ten to five.

It came in on a phone call you did not schedule, attached to a deadline you did not set, and it wants an answer while you are at your emptiest. This is the part the advice never covers. You can know everything about your peak hours and it will not move the phone call by a minute.

The question people actually type is "when is the best time of day to make decisions", and it deserves the honest answer, which comes in two parts. There is a best time. And knowing it is worth far less than you have been told.

Is there a best time of day to make decisions?

Roughly, yes. Alertness is not flat across the day. Two forces shape it: an internal clock that rises and falls on its own schedule, and the pressure that builds with every hour you have been awake. Between them they produce the familiar terrain: a climb through the morning, a dip after lunch that no amount of coffee fully argues with, a partial recovery, a long decline into the evening. This much is settled science, replicated in sleep laboratories for decades.

But the terrain is not the same for everyone, and this is where the standard advice starts to fail. Chronotype, whether you run early or late, shifts the whole map. A systematic review published last year, covering sixty-five studies, confirmed what the field calls the synchrony effect: people do demanding analytic work best at their own circadian peak, and late types in the morning perform among the worst of anyone at any hour. "Decide in the morning" is a rule written by morning people. If you are not one, it is not merely unhelpful. It is exactly wrong.

So the clock answer exists, it is personal rather than universal, and it is real. Hold that thought while we look at what it is worth.

Why do decisions get worse later in the day?

Because two separate costs compound. The first is the clock just described. The second is the count: every decision made under any real consequence draws down a finite reserve, and the reserve does not care what time it is, only how much has already been spent. We have written about that depletion before; the short version is that a depleted system does not stop deciding, it economises, and its economies all point the same way: towards whatever ends the moment soonest.

The cleanest demonstration comes from medicine. A study in JAMA Internal Medicine tracked primary care doctors across four-hour clinic sessions and watched what happened to their prescribing for respiratory infections — the coughs and colds that mostly need no antibiotics at all. Prescriptions rose steadily through the session. By the fourth hour, a patient was roughly a quarter more likely to leave with antibiotics than an identical patient seen in the first. Same doctors, same guidelines, same conditions. The only variable that moved was how long the session had run. An Australian team later found the same shape in a national dataset of GP prescribing.

Note what eroded. Not knowledge. The doctors knew the guideline in hour four as well as they knew it in hour one. What eroded was the willingness to pay for the harder right answer when an easier available one would close the consultation. Writing the script ends the moment. Explaining why there will be no script extends it. Late in the day, the ending wins.

That is the true shape of late-day decline, and it is worth being precise about, because it tells you what any fix has to do. Your afternoon failures are not stupidity arriving on schedule. They are a pricing change. The costly option — the deferral confronted, the refund refused, the email answered in two hard sentences — quietly gets repriced beyond what the remaining reserve will pay.

Should you save your hard decisions for the morning?

This is the standard prescription, and it fails three times over.

It fails first for the chronotype reason. Your peak may not be in the morning at all, in which case the advice relocates your hardest calls to your weakest hours with great confidence.

It fails second because the morning is oversubscribed. The same literature that wants your decisions made before noon also wants your deep work there, and your exercise, your writing, and your strategy. The morning is a small room into which the entire productivity canon is trying to move furniture. Something has to stand in the hallway, and it will usually be the thing with no meeting attached, which is to say the decision.

And it fails third, and worst, on the point we started with: decisions do not queue. They arrive attached to other people's schedules. The investor calls when the investor calls. The resignation lands when it lands. The supplier's final offer expires at six. By the time most consequential decisions surface they are already urgent, and urgency does not book appointments for Tuesday at 9.40am. Advice that assumes you control arrival times is advice for a calendar you do not have.

This is the deeper problem with the whole question. "When should I decide?" treats timing as something you learn about yourself. It is not. It is something you either control or inherit. Uncontrolled, decision timing defaults to the worst available slot: the moment of arrival, whenever that is, met with whatever you happen to have left.

How do you take timing out of the decision?

Two moves, and they divide the territory between them.

The first covers everything recurring, which is most of it. Decide the category once, in advance, at full capacity, and the ten-to-five version stops being a decision at all. It becomes an execution. Any discount above a fixed threshold is declined. Any hire is a no unless two named people say yes. A rule made on a clear morning — your clear morning, wherever it falls — spends correctly at every hour afterwards, because nothing further is being spent. The best time to make a decision is before you need it: while it is still abstract, while nothing rides on the next ten minutes, while the version of you deciding is the one you would choose to send. This is the work of Decision Resolution: finding the decisions you make repeatedly and converting them into rules that hold.

The second move covers what cannot be pre-decided, the calls that are new each time. You cannot rule on these in advance, but you can stop letting them pick their own moment. Give them a standing slot at your actual peak, and make waiting the default. When a decision arrives, the only question answered on the spot is whether anything breaks if it waits for the slot. Almost nothing breaks. The few that cannot wait get made immediately, but under protocol rather than under mood, with the sequence fixed before the day began: the threshold named, the action attached, the outcome logged. That protocol is what Decision Throughput installs, and it exists precisely for the hour the research says to avoid and life says you cannot. A decision made at your worst hour under a good rule will beat a decision made at your best hour with none.

Notice that neither move asks you to become a morning person, and neither requires the science to care about you personally. The clock keeps doing what it does. You have simply arranged for less and less to depend on it.

The investor's call still comes at ten to five. What has changed is what it finds when it arrives.