The research behind "make your important decisions in the morning" is weaker than its reputation. The famous parole study found that favorable rulings fell across each session and recovered after a food break, but the pattern reset after lunch as well, and a published reply and a later simulation argue that much of it came from the order the cases were heard in. The willpower theory underneath it, ego depletion, produced effects near zero when 23 and then 36 laboratories tried to repeat it. For a small firm the useful lesson is narrower. A quote priced from scratch while the client waits is a decision you could have made once, in writing, in a calmer hour.
What the parole study printed
In 2011 Shai Danziger, Jonathan Levav and Liora Avnaim-Pesso looked at 1,112 rulings by eight judges on two Israeli parole boards. They reported that favorable rulings fell from about 65% at the start of each session to nearly zero by its end, then returned to about 65% after a food break. Each board was a judge sitting with a criminologist and a social worker, and "unfavorable" included deferrals to a later hearing as well as outright rejections.
The pattern reset after lunch as well as after the morning snack, so it tracked how far into a session a case came. It was not a morning-against-afternoon result. The authors were also careful about the cause. They wrote that they had no direct measure of the judges' mental resources and could not tell whether rest, food or mood made the difference. The paper reads its result as a drift toward the status quo over a run of decisions, interpreted "through the lens of mental depletion" without measuring depletion. The hungry judge of the retellings does not appear in it.
The reply, the rejoinder and the simulation
The same year, Keren Weinshall-Margel of the Israeli courts' research division and John Shapard of the Federal Judicial Center argued in a published reply that the cases were not heard in random order. They reported that the boards tried to finish one prison's cases before a break, and that prisoners without a lawyer usually went last in a session and were less likely to be granted parole. In their own sample of 227 decisions, unrepresented prisoners made up about a third of cases and prevailed 15% of the time, against 35% for prisoners with counsel. Their conclusion was that the post-meal peak was "likely an artifact of the order of case presentation." Their new data did not record case order, so they could not test that directly.
The original authors answered that when they added legal representation to their models, case order and the meal break still predicted the ruling. The letters did not settle it, because each side pointed to data the other did not have.
In 2016 Andreas Glöckner showed by simulation that much of the pattern could come from a judge sensibly not starting a long case just before a break, since favorable rulings took longer. He noted that a drop from 65% to 5% would be an unusually large effect, more than twice the conventional threshold for a large one. He concluded that the original analysis did "not provide conclusive evidence" that extraneous factors swayed the rulings and that the size of the effect was overestimated. He was also explicit that his simulation did not rule out fatigue and explained the original result "only in parts."
So the parole effect is disputed and probably smaller than it looks. It was not debunked. It also never showed that mornings produce better judgment.
Ego depletion, the idea underneath
The theory behind decision fatigue comes from a 1998 paper by Roy Baumeister and colleagues, four lab experiments with about 220 university students in total that suggested self-control draws on a limited resource. In the best known of them, 67 students who had to eat radishes while resisting chocolate gave up on an unsolvable puzzle after about 8 minutes on average, against about 19 to 21 minutes for the other groups. The authors' own word was "suggest." The popular picture of willpower as a tank that drains through the day traces back to that paper, though it never uses the phrase decision fatigue.
In 2016, 23 laboratories led by Martin Hagger ran the same ego-depletion experiment on 2,141 people and found an effect of d = 0.04, with a confidence interval that included zero. The task that replication used was uncommon in depletion research, so in 2021 Kathleen Vohs, one of the theory's own proponents, led a test across 36 laboratories and 3,531 participants using methods closer to the original literature. The preregistered analysis found no significant depletion effect (d = 0.06). Her team concluded that the effect is "likely small (including zero)."
The two largest tests found effects near zero. That leaves room for a small effect and no room for a reserve that reliably runs dry by the afternoon.
Exhibit 1
The popular decision fatigue story claims more than either the parole study or the depletion experiments delivered
The experiments that gave decision fatigue its name
The phrase itself gets its lab evidence from a 2008 paper by Kathleen Vohs and colleagues, a set of small experiments with university students, most with 25 to 42 people each. In one, students who had made a series of choices held their arm in ice water for about 28 seconds on average, against about 67 seconds for students who had not. The field part of the paper was 58 shoppers at a Salt Lake City mall, where those who reported making more choices that day did worse on simple arithmetic. That is an association, and it cannot say which way the link runs.
The results are real at their own size. They rest on the same depletion theory that the large multilab tests of 2016 and 2021 could not reproduce, and none of them involved a work decision.
Too many options is a separate claim
The jam study often gets folded into the same story. In Sheena Iyengar and Mark Lepper's 2000 study, run in one upscale grocery store in Menlo Park, California, on two Saturdays, a tasting booth with 24 jams drew more shoppers to stop (60%) than one with 6 (40%). Nearly 30% of those who stopped at the small display bought a jar, against 3% at the large one. In absolute terms, 31 shoppers bought after the six-jam display and 4 after the 24-jam display.
When Benjamin Scheibehenne and colleagues pooled 50 experiments with 5,036 people in 2010, the average choice-overload effect was "virtually zero," though results varied a lot from study to study. An attempt to repeat the jam study in an upscale German supermarket found no negative effect of the larger display. Choice overload may happen under some conditions, and nobody has pinned down which. It also concerns the number of options on a shelf, which is a different question from a day's worth of decisions.
What is left standing
A 2014 study by Jeffrey Linder and colleagues of about 22,000 primary-care visits found doctors prescribed antibiotics for respiratory infections more often as each four-hour clinic session wore on, with the pattern starting over after the midday break. The authors called the finding "consistent with" decision fatigue, and noted that it came from one health system and that "unmeasured confounding is possible."
That leaves two field settings where default choices rose over a run of decisions and reset after a break, and the cause of one of them is argued over. We could not find a well-designed study showing that people make better business decisions in the morning. Nothing here measured an office, a service business or anyone pricing a job, so none of these numbers transfers to your quotes.
Why the late quote goes wrong anyway
Owners do price badly late in the day. The quote goes out low because the client is on the phone, or high because the job looks annoying, or it sits unsent because nobody wants to think about it. We do not need a depleted reserve to explain that. The simpler account is that the owner is pricing from scratch every time, under time pressure, with someone waiting, and deciding the same questions again: what the hourly rate really is, what a rush costs, whether this client gets the discount the last one got.
Much of that is also the cost of switching tasks in the middle of other work. A quote that has to be reasoned out from first principles between two other jobs will come out different each time, whatever hour it is.
Decide the rate card once
The fix is a pricing rule made once, in writing, with the numbers the firm has agreed to stand behind. Then the quote late on a Friday is a lookup. Our piece on quotes that get answered makes the same case for settling the rate card and the exceptions when nobody is waiting and letting the busy hours apply them.
The rule should hold everything that does not change from job to job:
- the hourly or day rate, and the minimum charge
- the standard packages and what each one includes
- the rush premium and what counts as a rush
- the discounts you offer, and who qualifies
- payment terms and the deposit
What stays open is the scope of this job, the risks particular to this client, and whether to take the work at all. Those are real decisions, and there are far fewer of them once the rest is fixed.
Exhibit 2
A written rate card turns most of a quote into a lookup and leaves only the job-specific questions to decide
Once the rate card exists, drafting a quote from your past quotes becomes a job a person or an AI assistant can do and you can check.
Where the calm hour belongs
The research does not tell you when to work. We recommend setting pricing rules in a quiet hour because it is easier to think about exceptions when no client is waiting. That is our advice, and no laboratory tested it. Use the same hour to go over the quotes that did not fit the rule, the jobs that ran over and the discount you gave under pressure. A weekly review is where those go, and where the rule gets amended so the next quote does not need the argument again.
Exhibit 3
The pricing rule is made in a calm hour, the quote is a lookup, and the exceptions feed back into the weekly review
If you want help writing that rate card and the review that keeps it current, it is the kind of operating work our management consulting covers.