Skip to Content Skip to Footer
How Long Will This Take? People’s Estimated Duration of Time Periods is Affected by Category Boundaries

How Long Will This Take? People’s Estimated Duration of Time Periods is Affected by Category Boundaries

Elena Bocchi and Maria Giulia Trupia

Journal of Marketing Research Scholarly Insights are produced in partnership with the AMA Doctoral Students SIG – a shared interest network for Marketing PhD students across the world.

Many consumer decisions involve considerations of time. For instance, some consumers prefer to cook rather than order takeout, or take a taxi rather than walk. All these choices involve an estimation of how long different events will take. Prior research suggests that consumers’ estimates of duration are often biased. However, there is still limited understanding of the different experiential and contextual features that affect duration estimates. A recent Journal of Marketing Research article enhances our knowledge of when and why consumers over- and underestimate the duration of time periods. Kristin Donnelly, Giovanni Compiani, and Ellen R. K. Evers examine how consumers categorize time and how such categorization affects estimated durations, decisions, and valuations.

Across seven experiments and a large field dataset, the authors find that time periods feel longer when they span more time categories, as consumers categorize time according to salient (natural) boundaries. For example, time periods like 1:45 pm–2:15 pm and March 31–April 6 (“boundary-expanded”) feel longer than 1:15 pm–1:45 pm and April 2–April 8 (“boundary-compressed”). This effect has important consequences on consumer decisions. For example, consumers prefer to schedule pleasant activities for boundary-expanded periods and unpleasant activities for boundary-compressed periods. They are even willing to pay more to avoid a long wait when it is presented as a boundary-expanded period rather than a boundary-compressed period. Additionally, a large field dataset from more than 1.8 million rideshare trips confirms that consumers systematically prefer independent rides (e.g., UberX) over shared options (e.g., UberPool) for boundary-compressed time periods, further suggesting how consumers perceptions impact their behaviors. These findings provide significant insights for marketers. For instance, companies should try, if possible, to present negative events (e.g., waiting times) in boundary-compressed form and positive events (e.g., theater shows) in expanded form to improve consumer satisfaction.


Companies should try, if possible, to present negative events (e.g., waiting times) in boundary-compressed form and positive events (e.g., theater shows) in expanded form to improve consumer satisfaction.

We had a chance to talk with the authors to learn more about the current  study and gain additional insights into the research process.

Q: The effects shown in the paper are in the context of prospective duration estimation. Do you have any insights on the implications for people’s subjective experience of their day? For instance, could experiencing a day with many activities performed over boundary-expanded periods (vs. boundary-contracted) affect people’s feelings of productivity or happiness at the end of the day?

A: This is an excellent question. We believe that boundary-expansiveness is unlikely to affect one’s subjective experience of duration because the effect depends on the presence of salient category boundaries. Unless these boundaries are strongly emphasized during the interval itself (e.g., a clock ticking down, or switching classes, as in Study 3), the boundaries tested here (e.g., for hours, months) may not be salient enough to affect the actual experience of duration.

That said, we suspect that boundary-expansiveness would affect retrospective duration estimates in much the same way that it affects prospective ones. This is because human memory isn’t perfect, and people often incorporate external cues to piece together their impressions of what happened. A day may indeed feel longer when it was packed with boundary-expanded events (e.g., a meeting from 9:30am–11am, lunch from 11:30am–1pm) compared to boundary-compressed ones (e.g., a meeting from 9am–10:30am, lunch from 12pm–1:30pm)—provided, of course, that the meeting time is salient (e.g., by looking at one’s calendar). The implications of perceived duration for feelings of productivity or happiness should depend on the nature of the events. One could imagine that a day full of cognitively demanding, boundary-expanded Zoom meetings may increase feelings of productivity but not necessarily happiness (though of course, that might depend on the person). A spa day with boundary-expanded services may make the consumer feel like they got their money’s worth, and perhaps increase satisfaction with the experience. This topic is ripe for future exploration.

Q:  What other implications of the effect on time-related outcomes, such as busyness perception, time anxiety, perceived effort, or perceived procrastination, can you think of?

A: This work has strong implications for numerous behaviors, all of which draw from the assumption that a given point in time feels disproportionately further away when it belongs to a temporal category that is further away from the present. For example, a person may be more likely to procrastinate if a task is due on December 1st instead of November 30th but equally likely to procrastinate if the task were due December 2nd instead of the 1st (this is supported by existing research by Tu and Soman, 2014). One might also imagine that employees may be more likely to be late to a meeting that begins on the hour compared to the half hour, because they overestimate how much time they have before it starts. Similarly, a passenger en route to the airport may be more anxious about missing their flight when it departs just before a new hour instead of just after it. These are just a few examples, but there are a multitude of other situations that might be impacted by boundary-expansiveness.

Q: The effect is very convincing. Which psychological process do you think could be driving the effect? Did you consider exploring alternative mechanisms, and if so, what made you decide not to pursue that route?

A: The mechanism presented in the article is the only one that we considered. We argue that units of time like hours, years, or months serve as categories, and our effect stems from consumers applying a well-documented categorization principle in which two stimuli within the same category seem more similar to one another than equidistant stimuli in different categories. For example, 3:58pm and 3:59pm, which are both in the “3pm” category, seem more similar than, say, 3:59pm and 4:00pm, and this makes the former pair feel closer together temporally. Our studies collectively work to support this categorization-based explanation. For example, we demonstrate that consumers are sensitive to categorical hierarchy, such that they attend to the highest-level category (i.e., largest unit of measurement) that differentiates a period’s start and end time. We also distinguish our account from that of left-digit bias, showing that boundary-expanded periods continue to feel longer than boundary-compressed when the category information is in the middle (rather than leftmost). We further demonstrate that the effect holds for nontemporal categories like discrete events or even seasons.

Q: According to your research, participants prefer to schedule pleasant activities for boundary-expanded periods and unpleasant activities for boundary-compressed periods. Do you think the type of activities performed, before or after the specified time interval, could moderate this behavior? For instance, would people still prefer to schedule an unpleasant activity for a boundary-expanded period, if this interval was preceded by an unpleasant (vs. pleasant) activity?

A: We suspect that most of the time, people prefer to spend more time doing pleasant activities and less time doing unpleasant activities, regardless of whatever they did before (or will do after) those activities. If so, people should continue to prefer boundary-expanded periods for pleasant experiences and boundary-compressed periods for unpleasant ones.

Q: The data and the analytical approach employed in the last study (Uber trips) are impressive. What other contexts did you consider, in addition to Uber, in the process of searching for real-world data?

A: We initially searched for ways to obtain airline data, hoping to look at which flights consumers choose and whether they disproportionately select boundary-compressed options. This proved too difficult to obtain. We also considered other contexts in which boundary-expanded times might be less appealing (e.g., dental cleanings, HR trainings) but again were unable to find the relevant data. The rideshare dataset presented a unique and compelling opportunity to explore our effect within a common consumer context.

Q:  Do you predict similar patterns of results within cultures that put less focus on values such as the efficient and productive use of time? Relatedly, certain cultures keep track of time differently; for example, they use a 24-hour system instead of the 12-hour system (i.e., am–pm.)  How would such indication of time affect time perception?

A: In cultures that place lower value on time being used efficiently and productively, we would still expect to see the basic effect (i.e., that boundary-expanded periods feel longer than boundary-compressed). There is no reason to think that valuing time efficiency interacts with one’s perception of how long a prospective period feels when it is boundary-expanded compared to boundary-compressed. However, we might observe differences in downstream consequences that do reflect how much a person values efficiency. For example, if people in cultures that de-prioritize efficiency are generally less sensitive to delays or long waiting periods, we might also find that they have a weaker preference for waiting periods that feel shorter (i.e., boundary-compressed) instead of longer (i.e., boundary-expanded). This might be reflected in smaller differences, say, in how much money they might be willing to pay to avoid a boundary-expanded waiting period (as opposed to a boundary-compressed one). Cultures with different time-tracking conventions should still show the effect. For example, we show that periods that span more months (boundary-expanded) feel longer than equivalent periods spanning fewer (boundary-compressed), regardless of whether they are evaluated in month/day format by people in the U.S. (e.g., 3/30–4/10 feels longer than 4/2–4/13) or day/month format by people in the U.K. (e.g., 30/3–10/4 feels longer than 2/4–13/4). We believe that the effect documented here results from a basic categorization bias—two times within the same category feel closer temporally than times within different categories—and this should hold regardless of whether the category in question is represented on a 12- or 24-hour system.

Read the full article:

Kristin Donnelly, Giovanni Compiani, and Ellen R.K. Evers (2022), “Time Periods Feel Longer When They Span More Category Boundaries: Evidence from the Lab and the Field,” Journal of Marketing Research, 59 (4), 821–39. doi:10.1177/00222437211073810

Go to the Journal of Marketing Research

Elena Bocchi is a PhD Candidate, Bayes Business School, City, University of London, UK.

Maria Giulia Trupia is a Post-Doctoral Scholar, University of California Los Angeles, USA.