New research on habit formation and decision fatigue shows that willpower isn’t the deciding factor in discipline. The environment is. Here’s what that means for how you build systems that hold.
Every January, the same thing happens at gyms across the country. Attendance spikes in the first two weeks, then falls off a cliff by early February, back down to nearly the same numbers as December. The people who signed up were not lying to themselves about wanting to change. Most of them meant it completely. What they lacked was not desire. It was a structure sturdy enough to survive the first inconvenient evening.
This is where the free preview ends. This essay explains “designing the context,” its research on willpower depletion, temptation bundling, and a method to make desired behavior easier.
This pattern shows up everywhere, not just at gyms. Someone commits to writing every morning and manages four days before life intervenes. Someone decides to finally get serious about their finances and tracks every expense diligently for three weeks before the spreadsheet quietly disappears. The explanation people reach for is almost always the same: not enough willpower, not enough discipline, not enough grit. It feels like a personal failing because it happens to a person, privately, in a moment nobody else sees.
But that explanation does not hold up against what researchers who actually study behavior have found. Willpower is not the deciding factor in whether a habit sticks. The environment around the behavior is. And once you understand why, the entire project of building discipline starts to look completely different, less like a test of character and more like a design problem with a knowable set of solutions.
The Discipline You Think You Have Isn’t Doing the Work
Wendy Wood, a psychologist at the University of Southern California who has spent over three decades studying habit formation, found something that surprises most people the first time they hear it. Roughly 43 % of what a person does in a day is not the product of a deliberate decision at all. The surrounding context, the same place, the same time of day, the same sequence of small cues, automatically trigger this repeated behavior, which runs outside conscious awareness. Wood has been blunt about the implication in recent interviews: habits do not form through motivation, and popular advice that treats them as a willpower problem gets the mechanism backward from the start.
That finding reframes almost everything about how people typically try to change their behavior. If nearly half of daily actions run on autopilot triggered by context, then the sensible target for change is not the moment of decision. The context decides before a person ever decides consciously. Someone who wants to stop reaching for their phone first thing in the morning is not fighting a weak will. They are fighting a phone sitting on the nightstand within arm’s reach, a cue their brain has already learned to respond to automatically.
This is worth sitting with for a moment because it inverts a belief most people carry without ever examining it. The common assumption is that a disciplined person consciously chooses the right action, over and over, through some internal resource that the rest of us are simply worse at accessing. Wood’s research suggests something closer to the opposite. Disciplined people are not making more moral decisions than everyone else. They are making fewer decisions, period, because they have already engineered a context in which good behavior happens automatically, the same way an undisciplined context makes inappropriate behavior happen automatically for someone else.
Why Willpower Runs Out Faster Than You Think
That willpower is a limited resource is not just intuitive. It has real research behind it, including one of the more striking findings in behavioral science. In 2011, researchers Shai Danziger, Jonathan Levav, and Liora Avnaim-Pesso studied more than a thousand parole decisions made by experienced judges in Israel over ten months. Two food breaks broke into three sessions of the judges’ day. At the start of each session, roughly 65% of rulings were favorable. By the end of each session, that number dropped toward zero, then reset back to 65% immediately after the judges ate. The cases were not getting weaker. The judges were.
Researchers have scrutinized this finding since its publication, with some arguing that willpower depletion alone partly explains the effect, along with the order of cases within each session. That debate is worth taking seriously rather than treating the original study as settled fact. But even the critics of the parole study do not dispute the broader pattern found across dozens of other studies: repeated decision-making measurably degrades the quality of decisions that follow it, regardless of the decision-maker’s intelligence, training, or personal resolve. Judges are among the most disciplined, highly selected professionals in any field, and their judgment still eroded on a predictable schedule tied to something as unglamorous as blood sugar and mental fatigue.
If experienced judges cannot out-discipline this pattern through sheer will, an ordinary person trying to stick to a morning workout after a draining workday is not failing because they lack character. They are running the same experiment on themselves and getting the same predictable result. The lesson is not that willpower does not exist. It is that willpower behaves like any other finite resource: present in the morning, diminished by evening, and completely unreliable as the sole mechanism for holding a system together over months or years.
This has a practical consequence that most discipline-focused advice ignores entirely. If willpower depletes throughout the day, then scheduling the hardest, most identity-defining habits for the moment willpower is weakest is close to designing them to fail. The person who tries to write every night after a full workday is fighting the parole judges’ problem in miniature, asking a depleted decision-making system to override a comfortable, well-worn default. Move that same habit to the morning, before the day’s decisions have accumulated, and the odds shift substantially, not because the person suddenly has more discipline, but because they are asking less of a resource that predictably runs low.
The Want Versus Should Conflict, and a Way Around It
Even with better timing, plenty of habits still compete directly against something more immediately rewarding. Wharton professor Katy Milkman has spent years studying exactly this tension, what she calls the conflict between “wants,” things that feel good right now, and “shoulds,” things that pay off later but require exertion at the moment. Her own struggle with this conflict led to one of the more elegant pieces of applied behavioral research in the field.
Milkman loved audiobooks, particularly guilty-pleasure fiction she felt slightly embarrassed enjoying. She also struggled to get herself to the gym consistently. So she made a rule: she could only listen to those audiobooks while exercising. The result was immediate. She achieved her gym goal by merging a want with a need.
She later turned this personal experiment into a formal field study with Julia Minson and Kevin Volpp, giving gym-goers iPods loaded with addictive audiobooks that were only accessible at the gym. The results held up under real experimental conditions. Participants restricted to gym-only audiobook access attended the gym 51% more often than a control group, and 29% more often than a group whom the researchers simply encouraged to save the audiobooks for their workouts. The encouragement group barely got meaningful results. The restriction group, the one where the environment itself enforced the pairing rather than relying on memory or resolve, showed a dramatic and repeatable effect.
This is worth pausing on because it shows something Wood’s research only implies: environment design does not just remove friction from good behavior. It can actively manufacture desire for it by tying a behavior a person avoids to a reward a person already craves. The habit stops requiring willpower not because the person becomes more disciplined, but because the structure has quietly changed what the behavior actually feels like to do. Later research on temptation bundling, including a 2020 field experiment that explicitly taught the technique, found that the effect is strongest when participants genuinely enforce the restriction rather than rely on their own memory. This provides further evidence that the mechanism is structural rather than motivational.
Designing the Path of Least Resistance
This is where behavior design research offers something more useful than motivation ever did. Stanford researcher BJ Fogg, who has spent years studying what actually makes new behaviors stick, argues in his book Tiny Habits that the size of a habit matters less than how much friction stands between a person and the action. Fogg’s core insight is almost embarrassingly simple: behavior change succeeds when a desired action becomes easier to do than to avoid, and fails when the opposite is true, no matter how much someone wants the outcome.
This means the actual work of building discipline is not internal. It is architectural. The person who wants to exercise in the morning does not need more resolve. They need their running shoes already by the door, their gym clothes laid out the night before, their alarm set across the room instead of on the nightstand within reach of the snooze button. Daily habits rely on automatic responses, so removing decision points bypasses the need for willpower.
The same logic applies to eliminating unwanted behavior. Wood’s research on friction works in both directions. Just as reducing friction makes a wanted behavior more automatic, increasing friction on an unwanted one makes it less likely to happen, requiring no conscious resistance at all. Someone who wants to stop checking social media compulsively does not need to develop iron discipline against the impulse. They need to make the impulse harder to act on, logging out after each use, moving the app off the home screen, or removing it from the phone entirely during work hours. Friction performs the task that willpower could not do.
This principle scales beyond individual habits into how choices get presented more broadly, an idea economists Richard Thaler and Cass Sunstein popularized as choice architecture in their book Nudge. Thaler and Sunstein argue that the structure of options, specifically whether a choice requires an active selection or occurs by default, influences behavior more consistently than persuasion or information. Compared to a plan that automatically enrolls people, from which they can opt out, a retirement plan that requires employees to opt in receives a fraction of the participation. The financial decision is identical regardless of the method. The difference is entirely architectural. Nobody became more disciplined about saving. The default simply changed what the path of least resistance was.
What This Actually Means for the Way You Build Habits
None of this means effort or intention is irrelevant. Someone still has to decide what environment to build and why. But the decision only has to be made once, at the design stage, rather than fought for repeatedly at every single moment of temptation throughout the day. That is the entire advantage environment design has over willpower: it front-loads the hard decision into a single moment of clear thinking, then lets the structure carry the weight afterward, even on the days when motivation is completely absent.
This also changes what failure actually means. When a habit doesn’t stick, the honest question is not “why didn’t I have more discipline.” It is “what did the environment make easy, and what did it make hard.” A dropped habit is diagnostic information about a structural gap, not evidence of a character flaw. That reframe alone is the difference between someone who gives up on a goal permanently and someone who simply redesigns the structure around it and tries again with better architecture.
Applying this concretely usually breaks down into three separate, answerable questions rather than one vague resolution to “try harder.” First, what decision point stops you from performing the behavior, and can you eliminate that decision point rather than merely overcoming it? Second, could you associate a want with this should, similar to how Milkman paired audiobooks with the gym, so the behavior depends not only on delayed gratification? Third, does your current schedule for this habit coincide with a time when you have already depleted your decision-making capacity? Would moving it earlier in the day improve your chances of success without altering the habit itself?
None of these questions require more discipline to answer. You should spend about twenty minutes thinking honestly about these questions. Ideally, complete this on a day when the goal still feels far enough away to allow for clear thinking, rather than in the exhausted moment when the habit should occur.
Where This Breaks Down, and Why That Matters
It would be dishonest to present environment design as a complete substitute for every form of self-regulation. Friction and defaults cannot eliminate some decisions, particularly those involving other people, ethical judgment, or genuinely novel situations that no amount of pre-planning can expect. Wood herself has been careful to note that habits are powerful precisely because they are inflexible, which makes them poorly suited to contexts that actually require judgment rather than automation. Designing your environment does not make you a better decision-maker in the moments that genuinely call for deliberate thought. It simply frees up mental capacity, preventing the need to re-litigate decisions that never required reconsideration.
This distinction matters because it clarifies what environment design is actually for. It is not a way to avoid thinking. It is a way to stop wasting a scarce resource on decisions that do not deserve it, so that resource remains available for the decisions that genuinely do. A person who has removed the daily fight over whether to exercise has more capacity left, not less, for the decisions in their life that actually require judgment, empathy, or creativity. The goal was never to eliminate willpower from a person’s life. It was to stop spending it on battles a well-designed environment could have already won.
What This Looks Like Outside the Gym
The gym provides an easy example since researchers have built their work around it, yet this same architecture applies to nearly every field where discipline supposedly determines the outcome. A writer who struggles to start each morning is rarely missing motivation at 6 a.m. They are missing a manuscript already open on the screen the night before, a coffee already set to brew automatically, a phone charging in another room so the first hour of the day has no competing cues. A manager trying to give harder feedback more consistently is not missing courage at the moment. They are missing a recurring calendar block that makes the conversation the default rather than something requiring a fresh decision to start every single time.
Even financial discipline, often treated as the ultimate test of self-control, responds to the same logic. Automatic transfers into a savings account the moment a paycheck lands remove the decision entirely, the same mechanism behind Thaler and Sunstein’s opt-out retirement plans. The person is not exercising restraint every payday. They never see the money as available to spend in the first place because the environment made saving the default rather than the exception.
The Gym Membership Was Never the Real Story
The gym membership that goes unused every January is not a willpower story. It is a design story. The people who actually keep exercising months later are not, on average, more disciplined than the people who quit. They have usually just built an environment where showing up requires less friction than skipping, clothes ready, a workout partner waiting, a class already paid for and scheduled, perhaps even a favorite audiobook locked behind the gym’s door the way Milkman’s participants experienced. Discipline is not something you have more or less of as a fixed trait. It is something you build deliberately into the structure of an ordinary day, so that on the days your motivation runs out, the system carries you anyway.
That is the actual shift this research asks for. No more effort. No more resolve. A different question entirely, asked once, at the design stage, before the moment of temptation ever arrives: what would have to be true about this environment for the right behavior to be the easy one.
References
Wood, W. Interview, American Psychological Association Monitor on Psychology, 2026.
Wood, W., Quinn, J. M., & Kashy, D. A. Habits in everyday life: Thought, emotion, and action. Journal of Personality and Social Psychology, 2002. Summary via Wikipedia.
Danziger, S., Levav, J., & Avnaim-Pesso, L. Extraneous factors in judicial decisions. Proceedings of the National Academy of Sciences, 2011.
Weinshall-Margel, K., & Shapard, J. Overlooked factors in the analysis of parole decisions. Cambridge Core, Judgment and Decision Making.
Milkman, K. L., Minson, J. A., & Volpp, K. G. Holding the Hunger Games hostage at the gym: An evaluation of temptation bundling. Management Science, 60(2), 2014.
Kirgios, E. L., et al. Teaching temptation bundling to boost exercise: A field experiment. ResearchGate, 2020.
Fogg, B. J. Tiny Habits: The Small Changes That Change Everything. Houghton Mifflin Harcourt, 2019.
Thaler, R. H., & Sunstein, C. R. Nudge: Improving Decisions About Health, Wealth, and Happiness. Yale University Press, 2008.