Why Does Waiting Feel Longer Than It Is?
In a 2023 study at a Chinese eye clinic, how long people thought they had waited predicted how satisfied they were. How long they actually waited had no direct effect (P = .365, Medicine (Baltimore), 2023). Two clocks run in every waiting room, and only one of them reached the survey.
That split is the answer to the question in the title. Waiting feels longer than it is because felt time and clock time are two different quantities. Researchers measure them separately. They don’t track each other reliably, and it’s the felt one that turns up in how people rate the wait.
How much longer, then? Nobody can tell you honestly. The field measurements that exist are nowhere near each other, because they were taken in wildly different places, and almost all of them are too old to quote here. The figure repeated most often online has no study underneath it at all.
What follows is what the evidence does support. Why the two clocks come apart, why not knowing stretches a wait, and what actually happens when you tell people how long they’ll be.
Key Takeaways
- The gap between how long a wait feels and how long it takes is real, but there is no single multiplier for it. The measurements that exist disagree enormously depending on the setting, and the figure quoted most often has no traceable source behind it.
- Telling people how long they will wait is the standard advice, and the recent trial evidence for it is mixed at best. A 2025 randomised trial found no difference in satisfaction between patients given an estimated wait and those not given one. Separately, in the same trial, 96 of the 118 patients surveyed at that hospital’s triage entrance said they wanted one displayed (Journal of Family & Community Medicine, 2025).
The gap is real, but it isn’t one number
The gap between the felt wait and the measured one is well attested. Its size is not. In a 2023 study at a Chinese eye clinic, 292 patients filled in a valid questionnaire about their visit. Perceived waiting time, expected waiting time and tolerance each had a significant effect on satisfaction. Actual waiting time had no direct effect (P = .365, Medicine (Baltimore), 2023). The clock still mattered. In the same analysis it reached satisfaction through what people expected and what they perceived.
The perceived effect was graded rather than all or nothing. In that sample, every extra minute of perceived waiting cost 0.17 points of satisfaction on a 0 to 100 scale (Medicine (Baltimore), 2023). The clock on the wall was accurate. It just wasn’t what patients were answering about.
One clinic, one country, one questionnaire. The authors say eye patients may not speak for patients in general. That’s the honest scope of it, and it’s also where the next question runs into trouble. How much longer does a wait feel? Read across the studies that have measured both numbers in the same place and the answers don’t line up. A traffic light, a hospital corridor and a bus stop are not the same wait.
There’s a second problem, and it’s the reason this question is so badly served. Ask the internet and you’ll get a confident percentage back, often as a plain answer at the top of the page. Follow it. It hops from one post to the next, each borrowing from the one before, and the chain ends without a study at the bottom of it. Nearby figures carry the names of real journals that don’t appear to contain them.
So the honest answer to “how much longer” is that no agreed figure has ever been established, not that the right one is hard to find.
One more thing is worth saying, because almost nothing written on this topic says it. Waits don’t always feel longer than they are. In a 2021 randomised trial, parents estimated the wait in an Iranian children’s emergency department. The group with a queue management system averaged below the clock. The control group ran above it (Archives of Academic Emergency Medicine, 2021).
Those measurements are too old to cite as figures in an article about current evidence, so no numbers from them appear here. The direction they point in still stands. “Waiting feels longer” is a tendency under particular conditions, not a law of nature.
Is it the wait that feels long, or the time that feels slow?
Those are two different questions and they get two different answers. “How long did that take?” is a duration judgement. “How slowly did that pass?” is a passage-of-time judgement. Psychologists measure the two apart, because they behave apart.
A 2026 study in Attention, Perception & Psychophysics tested both at once. Ninety-four volunteers saw emotional pictures on a screen, each shown for 2, 4 or 6 seconds. Eighty-seven of them were kept in the analysis. After each picture they rated either how long it had lasted or how fast the time had gone, along with how the image made them feel.
The two judgements came apart cleanly. How arousing or how pleasant a picture was had no effect on duration judgements. It had a significant effect on passage-of-time judgements (Attention, Perception & Psychophysics, 2026).
The direction is the surprise. Higher arousal made time feel faster, not slower.
Now the scope, because it decides how far this travels. Those were pictures on a screen lasting a few seconds, rated on a sliding scale. Nothing in that experiment is evidence about a twenty-minute queue, and the paper doesn’t claim otherwise. What it establishes is that the two judgements can be pulled apart at all.
That is still worth carrying into a room with people waiting in it. A customer who says “that felt like forever” and a customer who guesses “about twenty minutes” are answering different questions. Change something about the room and you might move one of those answers while the other stays exactly where it was.
So a venue that wants to know how it’s doing has two things to measure rather than one. What that takes in practice is the subject of how to measure customer wait times.
Why does not knowing make it worse?
Because with nothing to look at, the same delay gets rated as longer. An experiment accepted to the ACM CHI 2026 conference put 425 people through a system delay of 10, 30 or 60 seconds during an online visual reasoning task. Some saw a countdown of the time remaining, some saw the time already elapsed, and some saw no display at all.
Averaged across all three delay lengths, the no-display group rated the wait 4.06 on a seven-point scale. The countdown group averaged 3.70, and the difference was significant at p < .01 (Tan and Nov, ACM CHI 2026).
Read that narrowly, because the paper does. It reports no other pairwise differences on perceived duration. So showing nothing was rated longer than a countdown, and it was not rated longer than the elapsed-time display. Scope again, too: this is a screen delay measured in seconds, not a queue measured in minutes.
The idea is much older than the experiment. Harvard Business School published a background note by David Maister in April 1984, item number 684064. It set out “Eight (testable) propositions concerning the psychology of queues” (Harvard Business Publishing). Maister set the eight out himself in 2006. Two of them hold that uncertain waits and unexplained waits feel longer than known, explained ones.
The original is a teaching note with eight propositions, and none of them was ever a statistic.
One piece of the standard advice sits awkwardly next to all this. Maister’s first proposition holds that occupied time feels shorter than unoccupied time, which is why waiting rooms get magazines and counters get screens.
There’s a well-established laboratory finding that runs the other way. An interval with something in it, the filled-duration illusion, tends to be judged longer than an empty interval of the same length (Tan and Nov, ACM CHI 2026). Nothing recent enough to cite here reconciles the two, so it’s an open question rather than a debunking, and occupied and unoccupied waiting time is where that argument lives.
Telling people how long they’ll wait does less than you’d think
The advice is everywhere and the trial evidence behind it is thin. A randomised trial published in 2025 tested it directly in a hospital emergency department in eastern Saudi Arabia (Journal of Family & Community Medicine). Some patients were given an estimated waiting time. Others were not. The reported finding: “No statistically significant difference in the level of satisfaction between patients who were provided with the estimated waiting time and those who were not (P=0.962).”
Then comes the part that makes this trial unusually clean. Inside the same trial, 96 of the 118 patients surveyed at the triage entrance said they wanted an estimated wait displayed (Journal of Family & Community Medicine, 2025).
Those are two different groups within one trial, and the distinction is worth stating plainly. 190 patients took part. 118 were surveyed at the entrance, and 72 were inside the department, split into 35 who were told and 37 who were not. “People said they wanted it” and “it changed nothing measurable” are both findings of the same study.
A second 2025 study narrows down when it does help. In a paediatric outpatient department in China, 2,019 people were divided into a control group of 1,013 and an experimental group of 1,006 (BMC Health Services Research, 2025). Satisfaction differed between the groups only where the actual wait ran past the expected one. Where the wait matched the expectation or beat it, no significant difference showed up.
Put those two together and a conditional appears that neither states on its own. A quoted estimate is insurance against your bad days. It isn’t an upgrade on your good ones.
There’s also a design finding in the 2026 screen-delay experiment, and it cuts against the obvious choice. Showing the time remaining drew significantly more frustration than showing the time already elapsed. That was 30.7 against 24.1 on the NASA-TLX frustration subscale, which runs from 0 to 100, p=.019 (Tan and Nov, ACM CHI 2026). The countdown group also rated the experience as significantly less pleasant (p=.025).
And one result nobody quotes. Those differences in how the wait felt “did not translate into differences in post-wait task performance” (Tan and Nov, ACM CHI 2026). The display changed the experience. It did not change what people did next.
| What was tested | Where, and with whom | What happened |
|---|---|---|
| An estimated waiting time given at triage | Emergency department, eastern Saudi Arabia. 72 patients inside the department, 35 told and 37 not told (Journal of Family & Community Medicine, 2025) | No statistically significant difference in satisfaction, P=0.962. Separately, 96 of the 118 patients surveyed at the entrance said they wanted an estimate displayed |
| Expected waiting time given to outpatients | Paediatric outpatient department, China. 2,019 people, 1,013 control and 1,006 experimental (BMC Health Services Research, 2025) | Satisfaction differed only where the actual wait ran past the expected one, and not where it matched or beat it |
| A countdown, an elapsed timer, or no display | Online experiment, 425 participants, screen delays of 10, 30 or 60 seconds (Tan and Nov, ACM CHI 2026) | No display was rated longer than the countdown on a seven-point scale. The countdown drew significantly more frustration than the elapsed timer |
Perceived time is not a softer version of real time
Notice what keeps happening across those four sets of results. The perceived wait predicted satisfaction directly and the actual wait did not. Arousal moved one kind of time judgement and left the other untouched. What patients said they wanted did not predict what changed their satisfaction. A display altered how a delay felt without altering what anyone did afterwards.
The popular version of this topic treats felt time as real time with some blur on it. Same quantity, estimated badly by a person instead of counted properly by a clock. The research keeps finding something else. Felt duration behaves as its own variable, with its own inputs.
That has an awkward consequence for anyone running a room full of waiting people. Cutting the actual wait is not the only lever, because felt duration has inputs of its own. Which of them are worth acting on is a separate question, and how to reduce perceived wait time takes it.
There is a harder question underneath it. Once a wait has stretched far enough, some people stop waiting, and how many of them do is barely measured at all: how long customers wait before leaving.
Frequently Asked Questions
How much longer does waiting actually feel?
There’s no reliable figure for it, and the direction isn’t even fixed. In a 2021 Iranian trial, parents at a children’s emergency department gave the estimates. One group came in below the clock, another above it (Archives of Academic Emergency Medicine, 2021). Anyone quoting a tidy percentage is passing on a number with no study under it. What holds up is narrower: the felt wait is the one that directly tracks satisfaction (Medicine (Baltimore), 2023).
Does showing a countdown make waiting feel shorter?
Shorter than showing nothing, yes, in a 2026 online experiment on screen delays of 10, 30 or 60 seconds. A delay with no display was rated longer than the same delay with a countdown of time remaining. That countdown also drew significantly more frustration than a display of time already elapsed, so it wasn’t the comfortable option (Tan and Nov, ACM CHI 2026).
Why do some waits feel fine and others feel endless?
Expectation and uncertainty do much of the work. In a 2025 Chinese paediatric outpatient study, telling people the expected wait mattered to satisfaction only when the real wait ran past it (BMC Health Services Research, 2025). Maister’s 1984 propositions make the same point from the other side: uncertain and unexplained waits feel longer than known ones.
Sources
- Archives of Academic Emergency Medicine, Bidari and colleagues, “Effect of Queue Management System on Patient Satisfaction in Emergency Department; a Randomized Controlled Trial”, published 5 September 2021, retrieved 2026-08-27, https://pmc.ncbi.nlm.nih.gov/articles/PMC8464010/
- arXiv, Tan and Nov, “Counting the Wait: Effects of Temporal Feedback on Downstream Task Performance and Perceived Wait-Time Experience during System-Imposed Delays”, accepted to ACM CHI 2026, posted 4 February 2026, retrieved 2026-08-27, https://arxiv.org/abs/2602.04138
- Attention, Perception & Psychophysics, Bratzke, “Emotional pictures and time: The effects of arousal and valence on the perception of duration and the subjective passage of time”, 88(4):98, published 24 March 2026, retrieved 2026-08-27, https://pmc.ncbi.nlm.nih.gov/articles/PMC13013257/
- BMC Health Services Research, Zhang and colleagues, “Impact of expected waiting time on pediatric outpatient satisfaction: a behavioral experiment study”, published 14 May 2025, retrieved 2026-08-27, https://pmc.ncbi.nlm.nih.gov/articles/PMC12080158/
- David Maister, “The Psychology of Waiting Lines”, the author’s own restatement of his eight propositions, posted 8 September 2006, retrieved 2026-08-27, https://davidmaister.com/blog/201/the-psychology-of-waiting-lines.html
- Harvard Business Publishing, “Psychology of Waiting Lines”, background note 684064, published 2 April 1984, retrieved 2026-08-27, https://store.hbr.org/product/psychology-of-waiting-lines/684064
- Journal of Family & Community Medicine, Almulhim and colleagues, “Effect of the provision of estimated waiting time on patient satisfaction with the Emergency Department: A randomized controlled trial”, published 17 January 2025, retrieved 2026-08-27, https://pmc.ncbi.nlm.nih.gov/articles/PMC11864359/
- Medicine (Baltimore), “Effect of waiting time on patient satisfaction in outpatient: An empirical investigation”, published 6 October 2023, retrieved 2026-08-27, https://pmc.ncbi.nlm.nih.gov/articles/PMC10553012/