Keywords: deadlines; self-imposed deadlines; procrastination; commitment devices; present bias; replication; time management

Abstract

Setting your own deadlines is one of the most common pieces of advice for people who put work off. For two decades, much of that advice rested on a 2002 experiment reporting that self-imposed deadlines improved performance, though less than evenly spaced deadlines set by someone else. In September 2026 Psychological Science retracted that article, after a close replication found no effect of any deadline condition (Hyndman & Bisin, 2026). This review asks what the remaining peer-reviewed evidence says about whether self-set deadlines help people finish work. Searches of OpenAlex, Crossref, Semantic Scholar and Europe PMC in October 2026 identified 17 eligible journal articles, each appraised with the CRAAP test. In the controlled studies reviewed, many people chose to bind themselves with deadlines when offered the option (Bisin & Hyndman, 2020; Hyndman & Bisin, 2026). Evidence that hard intermediate deadlines raise completion is weak, and two field experiments found lower completion under some deadline conditions (Burger et al., 2011; Bisin & Hyndman, 2020). Softer commitments, such as a nonbinding agreement or small daily tasks with social accountability, improved progress in field experiments with students (Himmler et al., 2019; Felkey et al., 2023). Deadline length also matters: shorter windows raised redemption of gift certificates (Shu & Gneezy, 2010), and longer deadlines led people to judge a goal as harder (Zhu, Bagchi, & Hock, 2018). The evidence supports soft, near-term commitments over rigid self-imposed cut-offs, with the caveat that most studies used students and short tasks.

1. Introduction

A person who misses deadlines is usually told to make more of them: break the due date into milestones, or set a personal deadline a week before the real one. If effort clusters just before a deadline, the reasoning goes, more deadlines should spread it out.

Behavioural research has a name for the pattern this advice targets. Cornelius König and Martin Kleinmann modelled when students studied for exams and found that a hyperbolic discounting function fitted the rise in study time before the exam better than the exponential function of standard economic models (König & Kleinmann, 2005). Economists describe the same tendency as present bias, and formal models predict that a person aware of it will value ways to restrict their own future choices (Hyndman & Bisin, 2021). Gharad Bryan, Dean Karlan and Scott Nelson at Yale University reviewed the evidence on such commitment devices and distinguished hard commitments, which carry real penalties, from soft ones, which work mainly through psychological cost (Bryan et al., 2010).

The most cited test of self-imposed deadlines was an experiment by Dan Ariely and Klaus Wertenbroch, published in 2002. Its abstract reported that people were willing to impose costly deadlines on themselves, that those deadlines improved performance, and that people set them less effectively than an external, evenly spaced schedule. That article has now been retracted. This review asks which part of the conclusion survives once it is removed: do self-imposed deadlines help people finish work, and under which conditions do they backfire?

2. Method

2.1 Search strategy

Searches were run on 13 October 2026 in OpenAlex, Crossref, Semantic Scholar and Europe PMC, supplemented by WebSearch in extended mode for recent journal articles and replications. Search strings combined "deadline", "self-imposed deadline", "interim deadline", "commitment device", "present bias", "procrastination", "time window", "urgency", "completion time" and "midpoint". Reference lists of the included replication and field studies were used to identify further work. A separate search looked for evidence that deadlines reduce completion or performance.

2.2 Eligibility criteria

Sources were included when they (a) appeared in a peer-reviewed journal, (b) reported a controlled experiment, a field experiment, an observational field study, a replication, a formal model or a review published in a major journal of the field, (c) concerned deadlines, completion windows, commitment to a schedule or the timing of work, and (d) studied non-clinical participants. Books, preprints, working papers, theses, blogs and press coverage were excluded as evidence. Where a working paper and a journal article existed, the journal version was cited. Retracted articles were excluded as evidence.

2.3 Appraisal with the CRAAP test

Each source was appraised on the five CRAAP criteria. Currency weighed the publication date against later work on the same question, with older studies kept when they remain the reference demonstration of a pattern. Relevance assessed how closely the task resembles personal work with a due date. Authority required publication in a peer-reviewed journal and was checked against retraction status in Crossref. Accuracy examined random assignment, sample size, the statistics reported and whether conclusions stay within the data. Purpose checked for an academic aim and for commercial framing. The full texts of Bisin and Hyndman (2020) and Hyndman and Bisin (2026) were read. For the other 15 sources, data were taken from the published abstracts and bibliographic records.

2.4 Appraisal results

Source	Design	Appraisal notes	Decision
Hyndman & Bisin (2026)	Close replication, N = 124	Current, directly relevant; full text read; original materials unavailable to the authors.	Included
Bisin & Hyndman (2020)	Field experiment, 278 students	Full text read; randomised; one university.	Included
Burger et al. (2011)	Two field and online experiments	Directly relevant to interim deadlines; data taken from the abstract.	Included
Himmler et al. (2019)	Field experiment with university students	Soft commitment with a reminder-only control; strong design.	Included
Felkey et al. (2023)	Randomised trial, 276 online students	Current; combines a commitment with social accountability.	Included
Kaur et al. (2015)	Year-long field experiment with data entry workers	Workplace setting; commitment via pay contract.	Included
Shu & Gneezy (2010)	Field data and experiments	Window length; pleasant tasks only.	Included
Zhu, Bagchi & Hock (2018)	Experiments	Deadline length and inferred difficulty; consumer goals.	Included
Tu & Soman (2014)	Field and laboratory studies	How a deadline is categorised in time; samples in India and North America.	Included
Zhu, Yang & Hsee (2018)	Five experiments	Urgency against importance; independently replicated.	Included
Mills et al. (2026)	Independent replication, n = 2,374	Current; tests confounds in the original paradigm.	Included
Buehler et al. (2010)	Four experiments	Predicted completion times; closed and open tasks compared.	Included
Gersick (1988)	Field study of eight teams	Dated; teams with a project deadline; small sample.	Included as mechanism
Gersick (1989)	Laboratory study of task groups	Midpoint transitions in groups.	Included as mechanism
König & Kleinmann (2005)	Model fit to study-time data	Describes the deadline rush; no intervention.	Included as mechanism
Bryan et al. (2010)	Review	Hard and soft commitment distinguished; no new data.	Included as theory
Hyndman & Bisin (2021)	Formal model	Predicts who sets binding deadlines; numerical results only.	Included as theory
Ariely & Wertenbroch (2002)	Experiments	Retracted in 2026 (Retraction, 2026).	Excluded

3. Findings

3.1 The retraction and the replication

Kyle Hyndman of the University of Texas at Dallas and Alberto Bisin of New York University replicated Study 2 of the 2002 article, a proofreading task with three texts and three deadline conditions: evenly spaced deadlines set by the experimenters, deadlines chosen by the participants, and a single final deadline (Hyndman & Bisin, 2026). With 124 adults at a large public university in the United States, the deadline condition had no significant effect on errors found, days late or earnings, and no significant effect on the survey measures used in the original study. Evenly spaced external deadlines did not stand out as more effective. Most participants given the option set binding deadlines for themselves. Many participants failed to submit some or all of the tasks, at similar rates across conditions, whereas the original article reported no attrition.

The retraction notice, published in the September 2026 issue of Psychological Science, states that the replication raised questions about the original data, that a dataset believed to relate to the study could not be confirmed as authentic or complete, and that an independent investigation had questioned the veracity of the findings (Retraction, 2026). The notice states that both authors agreed with the retraction. The replication authors note their own limits, including the absence of the original materials and the two decades between the studies. One failed replication leaves open whether self-imposed deadlines help under other conditions. It removes the main experimental support for the claim that they improve performance.

3.2 People choose deadlines, and completion did not improve

The strongest remaining field evidence comes from an earlier experiment by the same pair (Bisin & Hyndman, 2020). Students at New York University had to alphabetise long word lists online for pay, either one task in a week or three tasks in two weeks. Some were given no intermediate deadlines, some were given evenly spaced deadlines, and some could set their own. Between about a third and three fifths of students given the option chose a binding deadline for the first task. In the single-task treatments, completion was 57.1% with the option to self-impose a deadline and 45.6% without, a difference that was not statistically significant. In the three-task treatments, completion was highest with no intermediate deadlines (47.0%), lower with external deadlines (40.6%) and lowest with self-imposed deadlines (36.8%), and the gap between self-imposed and no deadlines was significant at p < .05.

The authors trace part of this to perseverance. Each additional day before the deadline at which a student started a task raised the probability of completing it by about 2.6%, so deadlines that shortened the available time also shortened the time to recover from a failed attempt. Students who gave up on one task rarely attempted later ones. The authors also conclude that students set deadlines suboptimally, which they attribute to partial naivety about their own self-control and to overconfidence about finishing. Their formal model predicts the same split: people who are aware of their present bias tend to impose binding deadlines, whereas naive ones do not (Hyndman & Bisin, 2021).

Nicholas Burger, Gary Charness and John Lynham found a similar result with a longer task (Burger et al., 2011). In their first experiment, evenly spaced interim deadlines for a lengthy task lowered completion rates. The aggregate data showed no rising pattern in hours worked over time, which the authors read as evidence of fairly weak present bias, together with wide variation between individuals.

3.3 Soft commitments produced gains in randomised trials

Results differ when the commitment carries no formal penalty. Oliver Himmler, Robert Jäckle and Philipp Weinschenk ran a field experiment in which university students could sign a nonbinding agreement to stay on track for graduation (Himmler et al., 2019). Students offered the agreement were more likely to register for exams, take them and pass them. A reminder-only treatment did not change behaviour, which the authors take as evidence that the effect did not arise from salience alone. Students who procrastinated benefited most.

Amanda Felkey, Eva Dziadula and Eric Chiang randomly assigned 276 students in online courses to microcommitments, small daily tasks with a soft commitment and social accountability on an online platform, or to a control group (Felkey et al., 2023). Treated students were nearly twice as likely to complete some of the assigned work more than a week before the due date. In both trials the commitment was framed as an intention with social or personal weight, and missing it carried no fine.

Hard commitments can also raise output when the stakes are real and the work recurs. Supreet Kaur, Michael Kremer and Sendhil Mullainathan ran a year-long field experiment with full-time data entry workers (Kaur et al., 2015). Workers raised their output by voluntarily choosing contracts that penalised low output and paid nothing extra for high output. Effort also rose as randomly assigned paydays approached, and within workers the contract and payday effects were strongly correlated, a link that grew with experience. The setting differs from personal deadlines: the commitment was a daily pay target in paid work, repeated over many months.

3.4 The length of the window changes behaviour

Two lines of work suggest that a deadline's length matters as much as its existence. Suzanne Shu of the University of California, Los Angeles and Ayelet Gneezy of the University of California, San Diego found that people delay even pleasant experiences when the window is open-ended (Shu & Gneezy, 2010). In a study across several cities, people with unlimited time windows delayed visiting desirable landmarks, and people redeemed gift certificates with long deadlines less often than those with short deadlines, which produced lower overall redemption. Participants predicted the opposite.

Meng Zhu, Rajesh Bagchi and Stefan Hock identified a mere deadline effect (Zhu, Bagchi, & Hock, 2018). When a deadline was longer, even for incidental reasons that said nothing about the task, people inferred that the goal was more difficult and committed more time and money to it. The authors report that the inference can also backfire, since elevated resource estimates were linked to more procrastination and a higher likelihood of quitting.

How a deadline is mentally categorised also matters. Yanping Tu and Dilip Soman studied farmers in India and students in North America and found that people were more likely to start a task when its deadline fell in a time category they saw as like the present (Tu & Soman, 2014). The authors explain the effect through an implemental mindset that a near-category deadline triggers.

3.5 Urgency can crowd out importance

Deadlines create urgency, and urgency carries its own pull. Meng Zhu, Yang Yang and Christopher Hsee showed in five experiments that people chose tasks with lower payoffs over tasks with higher payoffs when the lower-paying task merely appeared urgent, for example through an illusion of expiration (Zhu, Yang, & Hsee, 2018). An independent replication at the University of Nottingham with 2,374 participants confirmed the effect (Mills et al., 2026). The effect weakened as controls for confounds were added, and it remained significant in every test. For personal planning, the finding suggests that an artificial deadline attached to a minor task can draw time away from more valuable work that has no date on it.

3.6 Deadlines organise effort at transition points

Connie Gersick followed eight project teams from start to finish and found that they did not progress through steady stages (Gersick, 1988). Long periods of inertia were broken by a burst of change, and the timing of that change was linked more to the team's awareness of time and its deadline than to the amount of work completed. A laboratory follow-up showed groups making deliberate shifts in their approach at the temporal midpoint of their allotted time, and described what happened when that transition failed (Gersick, 1989). The samples were small groups, and the studies suggest that a deadline serves as a reference point for checking progress.

A prediction of finishing time can be read as an informal deadline. Roger Buehler and Johanna Peetz of Wilfrid Laurier University and Dale Griffin of the University of British Columbia used anchoring to push participants toward earlier or later predictions of when they would finish a task (Buehler et al., 2010). Across four experiments, earlier predictions led to earlier completion for closed tasks, which are done in one continuous session. Predictions had no such effect on open tasks that stretched across several sessions. The authors conclude that predictions help people start, and that their influence fades over longer, multi-stage projects.

4. Discussion

4.1 What survives the retraction

Two claims from the earlier literature remain well supported. Many people want to commit themselves, and they will accept binding deadlines when offered (Bisin & Hyndman, 2020; Hyndman & Bisin, 2026). Effort tends to rise as a deadline or payday approaches (König & Kleinmann, 2005; Kaur et al., 2015). The claim that has lost its support is the practical one: that setting your own hard intermediate deadlines improves the result. In the controlled studies reviewed here, such deadlines produced no measurable gain, and in some conditions lower completion (Burger et al., 2011; Bisin & Hyndman, 2020; Hyndman & Bisin, 2026).

The studies point to a mechanism. A hard intermediate deadline removes slack. When a task takes longer than expected, or a first attempt fails, the person who has set a near cut-off has less time left to recover, and missing one deadline can lead them to abandon the series (Bisin & Hyndman, 2020). A soft commitment keeps the planning benefit of a date while leaving room to recover, which may explain why nonbinding agreements and microcommitments improved progress (Himmler et al., 2019; Felkey et al., 2023). This explanation is an inference across studies with different designs and has not been tested directly.

The window findings add a second consideration. Long or open windows invited delay (Shu & Gneezy, 2010) and made goals seem harder (Zhu, Bagchi, & Hock, 2018), while a deadline perceived as near made starting more likely (Tu & Soman, 2014). These results favour near-term targets with some slack, a combination no study has tested as a personal planning strategy.

4.2 Limitations of the evidence

The evidence base is small and leans heavily on students. The two field experiments with the clearest negative results come from one research pair at one university, and the replication by the same authors could not use the original materials. The tasks were artificial, such as alphabetising word lists or proofreading generated text, and lasted days or weeks. The positive soft-commitment results come from academic settings with exams as outcomes, and Kaur et al. (2015) studied paid work with a daily target. Several window and urgency studies come from consumer research with choice and redemption outcomes. No included study randomised adults to self-imposed deadlines for their own work over months and measured completion. For most sources, data were taken from abstracts, so moderators reported only in the full texts may be missing.

5. Practical implications

The recommendations below are graded by the strength of the evidence behind them.

Do not count on a hard personal deadline to rescue a task. In controlled studies, binding self-imposed and evenly spaced interim deadlines produced no gain or lower completion (Burger et al., 2011; Bisin & Hyndman, 2020; Hyndman & Bisin, 2026). Strength: moderate, students and short tasks.
Prefer a soft commitment with some accountability. A nonbinding agreement and small daily tasks with social accountability improved progress in field experiments (Himmler et al., 2019; Felkey et al., 2023). Strength: moderate, academic settings.
Start well before the due date and leave room for a failed attempt. Starting earlier relative to the deadline raised the chance of completion (Bisin & Hyndman, 2020). Strength: limited, one field experiment.
Keep the window short and close to the present. Shorter windows reduced delay (Shu & Gneezy, 2010), deadlines categorised as near raised the likelihood of starting (Tu & Soman, 2014), and long deadlines raised perceived difficulty (Zhu, Bagchi, & Hock, 2018). Strength: moderate, consumer and student settings.
Check that urgent items are also the valuable ones. Apparent urgency drew choices toward lower-value tasks in replicated experiments (Zhu, Yang, & Hsee, 2018; Mills et al., 2026). Strength: moderate, laboratory choice tasks.
Use the halfway point as a review date. Teams with a deadline revised their approach at the midpoint of their time (Gersick, 1988, 1989). Strength: limited, small samples of groups.

A planning tool can apply the third recommendation by placing the work early in the available window and keeping later time free for recovery. Flowo schedules tasks against their due dates and re-plans the remainder when a session runs over, which leaves the deadline in place while moving the work. Whether a scheduled plan of this kind changes completion has not been studied. The guide to re-planning a day that has slipped covers what to do once a block has run long.

6. Limitations of this review

This is a narrative review with structured appraisal, carried out by a single reviewer. It was not preregistered, only English-language sources were searched, and publication bias was not assessed. The full texts of two sources were read; for the other 15, data were taken from abstracts and bibliographic records. Working papers and the commentary that preceded the retraction were not used as evidence. The reviewer has a commercial interest in planning software, declared below.

7. Conclusion

With the 2002 experiment retracted and its close replication null, the case that self-imposed deadlines improve performance rests on thin ground. People do want to bind themselves, and effort does gather near due dates. Controlled tests of hard personal deadlines, however, found no benefit or a cost, in part because a near cut-off leaves no time to recover from a setback. Softer commitments and near-term windows have better support, though mostly from students and short tasks. The most useful next study would randomise adults to soft and hard self-set deadlines for their own work over several months and measure completion.

Frequently asked questions

Do self-imposed deadlines work?

In the controlled studies reviewed here, hard deadlines that people set for themselves did not improve completion, and in one field experiment they lowered it (Bisin & Hyndman, 2020). The classic study that reported a benefit was retracted in 2026 (Retraction, 2026).

Was the Ariely deadline study retracted?

Yes. Psychological Science retracted Ariely and Wertenbroch (2002) in September 2026, after a close replication found no effect of deadline condition and the original data could not be verified (Hyndman & Bisin, 2026; Retraction, 2026).

Are evenly spaced interim deadlines better?

In two field experiments, evenly spaced deadlines set by the researchers did not raise completion and in some conditions lowered it (Burger et al., 2011; Bisin & Hyndman, 2020).

Why do I work better with a short deadline?

Shorter windows reduced delay in field data on gift certificates (Shu & Gneezy, 2010), and longer deadlines led people to see goals as harder (Zhu, Bagchi, & Hock, 2018). Both findings are consistent with that experience.

References

Ariely, D., & Wertenbroch, K. (2002). RETRACTED: Procrastination, deadlines, and performance: Self-control by precommitment. Psychological Science, 13(3), 219-224. https://doi.org/10.1111/1467-9280.00441 (retracted; excluded as evidence)
Bisin, A., & Hyndman, K. (2020). Present-bias, procrastination and deadlines in a field experiment. Games and Economic Behavior, 119, 339-357. https://doi.org/10.1016/j.geb.2019.11.010
Bryan, G., Karlan, D., & Nelson, S. (2010). Commitment devices. Annual Review of Economics, 2, 671-698. https://doi.org/10.1146/annurev.economics.102308.124324
Buehler, R., Peetz, J., & Griffin, D. (2010). Finishing on time: When do predictions influence completion times? Organizational Behavior and Human Decision Processes, 111(1), 23-32. https://doi.org/10.1016/j.obhdp.2009.08.001
Burger, N., Charness, G., & Lynham, J. (2011). Field and online experiments on self-control. Journal of Economic Behavior & Organization, 77(3), 393-404. https://doi.org/10.1016/j.jebo.2010.11.010
Felkey, A. J., Dziadula, E., & Chiang, E. P. (2023). Microcommitments: Mitigating procrastination with more than a nudge. Southern Economic Journal, 90(2), 497-509. https://doi.org/10.1002/soej.12659
Gersick, C. J. G. (1988). Time and transition in work teams: Toward a new model of group development. Academy of Management Journal, 31(1), 9-41. https://doi.org/10.2307/256496
Gersick, C. J. G. (1989). Marking time: Predictable transitions in task groups. Academy of Management Journal, 32(2), 274-309. https://doi.org/10.2307/256363
Himmler, O., Jäckle, R., & Weinschenk, P. (2019). Soft commitments, reminders, and academic performance. American Economic Journal: Applied Economics, 11(2), 114-142. https://doi.org/10.1257/app.20170288
Hyndman, K., & Bisin, A. (2021). Procrastination, self-imposed deadlines and other commitment devices. Economic Theory, 74(3), 871-897. https://doi.org/10.1007/s00199-021-01381-6
Hyndman, K., & Bisin, A. (2026). Replication of "Procrastination, deadlines, and performance: Self-control by precommitment." Psychological Science, 37(8), 557-571. https://doi.org/10.1177/09567976261460772
Kaur, S., Kremer, M., & Mullainathan, S. (2015). Self-control at work. Journal of Political Economy, 123(6), 1227-1277. https://doi.org/10.1086/683822
König, C. J., & Kleinmann, M. (2005). Deadline rush: A time management phenomenon and its mathematical description. The Journal of Psychology, 139(1), 33-45. https://doi.org/10.3200/JRLP.139.1.33-45
Mills, R., Cubitt, R., & Starmer, C. (2026). Brief commentary: An independent replication study of the mere urgency effect. Journal of Consumer Research. Advance online publication. https://doi.org/10.1093/jcr/ucag026
Retraction: Procrastination, deadlines, and performance: Self-control by precommitment. (2026). Psychological Science, 37(9), 665. https://doi.org/10.1177/09567976261488042
Shu, S. B., & Gneezy, A. (2010). Procrastination of enjoyable experiences. Journal of Marketing Research, 47(5), 933-944. https://doi.org/10.1509/jmkr.47.5.933
Tu, Y., & Soman, D. (2014). The categorization of time and its impact on task initiation. Journal of Consumer Research, 41(3), 810-822. https://doi.org/10.1086/677840
Zhu, M., Bagchi, R., & Hock, S. J. (2018). The mere deadline effect: Why more time might sabotage goal pursuit. Journal of Consumer Research, 45(5), 1068-1084. https://doi.org/10.1093/jcr/ucy030
Zhu, M., Yang, Y., & Hsee, C. K. (2018). The mere urgency effect. Journal of Consumer Research, 45(3), 673-690. https://doi.org/10.1093/jcr/ucy008

Conflict of interest: the author is a co-founder of Flowo, a planning app. None of the studies cited was funded by or connected to the company.

Flowo is free to download on the App Store for iPhone and on Google Play for Android, with optional in-app purchases.

Do Self-Imposed Deadlines Work? A Review of the Evidence After a Landmark Study Was Retracted