Design

Why waiting feels worse when design gets it wrong

Physical and digital wait

Waiting on public transport. Credit author.

Waiting is one of those experiences that often becomes visible only when something goes wrong.

A train platform with no countdown. A loading spinner that keeps spinning. A customer service chat that says “you are in the queue” but never says where in the queue. A public-sector form that has been submitted, but gives no clear answer about what happens next.

Waiting is often treated as empty time. The space between the user and the thing they actually came for. The train. The appointment. The answer. The confirmation. The completed upload.

People read the environment around them. They look for signs. They check whether the system seems to be working. They try to understand if they should wait, act, refresh, leave, call someone, or simply accept the delay.

That makes waiting a surprisingly important design material.

The stress of not knowing

A five-minute wait can feel harmless. It can also feel endless.

This difference sits at the center of David Maister’s well-known work on the psychology of waiting lines. Maister argued that organizations should not only pay attention to waiting time, but also to how that waiting time is experienced. A wait of the same length can feel very different depending on whether people are occupied or unoccupied, anxious or relaxed, informed or uninformed, treated fairly or ignored (Maister, 1985/2005).

One of the most consistent themes in waiting research is that uncertainty changes the experience of time.

Maister described uncertain waits as feeling longer than known waits. This does not mean that people enjoy long waits when they are explained. It means that uncertainty adds another layer to the experience. When people do not know how long something will take, why they are waiting, or whether the system has forgotten them, the wait becomes harder to tolerate (Maister, 1985/2005).

Hui and Tse (1996) studied how different types of waiting information affect consumer reactions. They examined waiting-duration information and queue information, and found that information can affect perceived waiting duration, the acceptability of the wait, and the emotional response to it. Their work is useful because it does not suggest that information is a simple universal fix. The type of information and the length of the wait both matter.

Users do not simply need “more information”. They need the right information at the right moment.

The bus stop as an interface

Public transport is one of the clearest examples of waiting as design.

A bus stop is not only a place where people stand. It is an interface between the passenger and the transport system. The passenger needs to understand whether they are in the right place, whether the service is running, when the next vehicle will arrive, whether there is a delay, and whether they should choose another route.

When that information is missing, the wait becomes harder to interpret.

Waiting at a busstop. Credti author.

Mishalani, McCord, and Wirtz (2006) studied passenger wait-time perceptions at bus stops and examined the relationship between actual and perceived waiting time. Their work is useful because it shows that the value of real-time arrival information is not only operational. It is also experiential. Real-time information helps passengers understand the wait.

Watkins, Ferris, Borning, Rutherford, and Layton (2011) studied OneBusAway, a real-time transit information system in Seattle. Their research found that mobile real-time information reduced both perceived and actual waiting time. One of the important findings was that riders using real-time information waited almost two minutes less than riders using traditional schedule information.

The interesting part is that real-time information changes the wait before the bus even arrives.

A countdown display turns waiting from an unknown state into a known state. The bus is not just “late” or “somewhere.” It is arriving in six minutes. Waiting unknowingly for six minutes might for some feel like an eternity. But knowing how long you have to wait could make the feeling and frustration more barrable.

The countdown displays at crosswalks also works very well when people are waiting to cross the crosswalk. To be able to see how long you have until you are allowed to safely cross the road, does benefit both pedestrians and drives safety a lot.

Countdown timers for pedestrian crosswalks. Credit: Traffic Solution

This is where transit information and UX design overlap. Both are concerned with the same basic question: does the user understand what is happening?

The digital waiting room

Digital products have their own waiting rooms. And when we’re talking about waiting in these rooms a lot of different things happens that do not happen out in the physical world, that makes waiting in digital room often feel longer and more annoying.

These rooms do not have chairs, ticket machines, platforms, or reception desks. But they still place users in moments where something is supposed to happen and the user has to wait for the system to respond.

A login screen is a waiting room. So is a file upload. So is a payment confirmation and so on. Almost every type of digital interface online have some sort of “waiting room” depending on what has to be loaded.

These moments often feel small because they happen inside interfaces. But from the user’s point of view, they can be highly sensitive. They often appear at moments where the user has already committed effort, attention, money, personal data, or trust.

In a physical queue, people can often read the situation around them. They can see whether the line is moving. They can count the people ahead of them. They can notice whether staff are present. They can ask someone what is happening.

In a digital queue, most of that context disappears.

The waiting page for the annual tax statement self service platform from the Danish Tax Agency. You can close the page and keep your place in the queue. Credit Skat.dk

The user cannot see the server. They cannot see the case worker. They cannot see the payment system. They cannot see whether a document is still uploading. They cannot see whether a login request has failed, frozen, or is simply taking longer than expected.

The screen has to do the work that the physical environment normally does.

It has to show status. It has to explain progress. It has to confirm that the user’s action was received. It has to make clear whether the user should wait, act, retry, or leave the page open. It has to reassure the user that the system has not forgotten them.

A customer service queue is a simple example. “An agent will be with you shortly” gives very little structure. “You are number 12 in line. Estimated wait time: 14 minutes. Now serving: 11” gives the user a clearer model of the situation. It shows position, progress, and expected duration.

How a “waiting room” in the digital space could look like. Credit author

The wait may still be inconvenient. But it is less invisible.

The same applies to login screens and authentication flows. If a user enters a password and nothing happens for several seconds, the interface creates uncertainty. A small loading state, a disabled button, or a clear “checking your details” message can reduce that uncertainty by showing that the system is actively processing the request.

Digital waiting is therefore not only about performance. It is about communication during moments when performance is not immediate.

Progress makes the user feel safe

Digital interfaces have their own version of the platform countdown: the progress indicator.

Progress bars, loading states, skeleton screens, upload percentages, queue positions, and step indicators are all attempts to make invisible system activity visible. They do not necessarily make the task faster. Instead, they make the wait more legible.

Brad Myers (1985) described a percent-done progress indicator as a graphical technique that allows the user to monitor progress through a task. That simple idea still sits behind many familiar interface patterns today. A progress indicator tells the user that the system is alive, that work is happening, and that the user has not been abandoned.

When a person waits in a physical queue, they can often see the line moving. They can count the people ahead of them. They can infer progress from the environment. In a digital system, the work is hidden. The user sees a screen, and the screen either responds or it does not.

The interface therefore has to represent progress on behalf of the system.

An upload flow is a simple example. If the interface only says “uploading,” the user knows something has started, but little more. If it shows file name, percentage, remaining steps, and a clear success state, the wait becomes easier to understand. The user can see that the system is doing something. They can also see when the task is complete.

A typical file upload screen. Credit author

The same applies to multi-step forms. A step indicator does more than show progress. It gives the user a sense of scale. It answers a basic question: how much of this is left?

That question matters because waiting is often harder when the end is invisible.

Not all feedback is useful

A spinner that loops forever does not show progress. It shows activity, but not completion. It can be useful for very short waits, but over time it becomes less reassuring because it does not explain where the user is in the process.

A progress bar that moves quickly to 90 percent and then freezes may be worse than a slower but more honest indicator. It creates an expectation that the task is almost finished, and then breaks that expectation.

This is where the design of waiting becomes closely connected to trust.

Harrison, Yeo, and Hudson (2010) studied how visual design can affect the perceived duration of progress bars. Their work shows that the perception of waiting can be influenced by how progress is visually represented. Later research has continued to explore how different progress-bar patterns affect perceived time, including whether constant or accelerating progress appears faster (Wang, Kang, & Rau, 2022).

But this type of research should not be read as permission to hide poor system performance behind clever animation. A progress indicator can make waiting more understandable. It can make the system feel responsive. It can support the user’s mental model.

Waiting for an upload to complete. Credit author.

If the final result fails, if the user loses data, or if the waiting state was misleading, the interface has not reduced friction. It has only delayed disappointment.

Occupied time feels shorter

Another recurring idea in waiting research is that occupied time feels shorter than unoccupied time.

Maister included this as one of the central propositions in his psychology of waiting lines. The basic idea is simple: when people have something meaningful to do, they are less focused on the passage of time itself (Maister, 1985/2005).

This principle appears everywhere.

Restaurants hand people menus while they wait. Airports place information screens near gates. Public offices use ticket numbers and status screens. Apps show skeleton layouts before the full content loads. Delivery services show order stages instead of leaving customers with a blank confirmation screen.

Waiting room with a waiting screen in a public office. Credit author

The best waiting experiences usually do more than keep people busy. They help people prepare, understand, decide, or feel reassured and of course inform so people don’t get frustrated and impatient.

A delivery app that shows “order received,” “being prepared,” “picked up,” and “nearby” is not only filling time. It is explaining the service process. A public-sector self-service flow that says “your application has been received, and the next step is review” is not only confirming receipt. It is reducing uncertainty.

There is a difference between occupied time and noisy time.

A playful loading animation may be charming once, but irritating if it appears too often. A hospital waiting room television may occupy attention, but it may not reduce anxiety if patients still do not know when they will be called. A skeleton screen may help users understand the shape of the coming page, but it cannot compensate for a slow or unreliable product.

Good waiting design gives the user something useful while they wait.

The first ten seconds

In interaction design, waiting begins almost immediately. Jakob Nielsen’s well-known response-time limits describe three important thresholds: around 0.1 seconds for an interaction to feel instantaneous, around 1 second for the user’s flow of thought to remain mostly uninterrupted, and around 10 seconds before attention is likely to drift and more explicit feedback becomes necessary (Nielsen, 1993).

These thresholds are useful because they show that waiting is not only about long delays. It is also about tiny breaks in the relationship between action and response.

A button that reacts immediately feels directly connected to the user’s action. A button that waits a moment before changing state may leave the user unsure whether they clicked correctly. A form submission that takes several seconds without feedback may encourage duplicate clicks, refreshes, or abandonment.

A disabled button after submission. A small loading state. A clear “saving” message. A success confirmation. These details may seem small, but they protect the user from uncertainty at exactly the moment where uncertainty begins.

The invisible design of waiting often starts before the user thinks of themselves as waiting.

Waiting as part of the service

One reason waiting is often poorly designed is that it is treated as outside the main experience.

The restaurant experience begins when the guest is seated. The transport experience begins when the train arrives. The digital experience begins when the page has loaded. The customer service experience begins when a human answers.

The passenger is already using the transport system while standing on the platform. The customer is already evaluating the company while waiting on hold. The user is already forming an impression of the product while watching the loading state. The patient is already experiencing the healthcare system in the waiting room.

This changes how design should approach it. A waiting state should not be treated as a leftover screen. It should be designed with the same care as the “real” interface.

What does the user need to know? What are they worried about? What can they do now? What should they expect next? What would make the wait feel fair, understandable, and controlled?

The research does not suggest that design can magically remove the cost of waiting. Long waits are still long waits. A progress bar cannot fix a broken system. Real-time information cannot make a cancelled train arrive. A confirmation email cannot make case handling faster.

But design can reduce the unnecessary frustration that comes from silence, uncertainty, poor feedback, and lack of orientation.

Research across waiting-line psychology, consumer behavior and public transport points toward the same broad conclusion: people do not experience waiting as pure time. They experience it through context.

That is why waiting deserves more attention in design.

References

Harrison, C., Yeo, Z., & Hudson, S. E. (2010). Faster progress bars: Manipulating perceived duration with visual augmentations. Proceedings of the SIGCHI Conference on Human Factors in Computing Systems, 1545–1548. doi:10.1145/1753326.1753556

Hui, M. K., & Tse, D. K. (1996). What to tell consumers in waits of different lengths: An integrative model of service evaluation. Journal of Marketing, 60(2), 81–90. doi:10.1177/002224299606000206

Maister, D. H. (2005). The psychology of waiting lines. Original work published 1985.

Mishalani, R. G., McCord, M. M., & Wirtz, J. (2006). Passenger wait time perceptions at bus stops: Empirical results and impact on evaluating real-time bus arrival information. Journal of Public Transportation, 9(2), 89–106. doi:10.5038/2375–0901.9.2.5

Myers, B. A. (1985). The importance of percent-done progress indicators for computer-human interfaces. ACM SIGCHI Bulletin, 11–17. doi:10.1145/317456.317459

Nielsen, J. (1993). Response times: The three important limits. Nielsen Norman Group.

Wang, Q., Kang, X., & Rau, P. L. P. (2022). The magic of slow-to-fast and constant: Evaluating time perception of progress bars by Bayesian model. arXiv. doi:10.48550/arXiv.2211.13909

Watkins, K. E., Ferris, B., Borning, A., Rutherford, G. S., & Layton, D. (2011). Where is my bus? Impact of mobile real-time information on the perceived and actual wait time of transit riders. Transportation Research Part A: Policy and Practice, 45(8), 839–848. doi:10.1016/j.tra.2011.06.010

Thank you for reading.

Hi, I’m Marcus Fleckner — a UX/UI designer and freelance photographer. I write about my passion for design, maps and photography. Join my newsletter📨 to get fresh insights and inspiration about UX, design and maps.