Temptation bundling is pairing an instantly gratifying want with a delayed-reward should, and only allowing the want during the should. Milkman, Minson and Volpp (2014) locked page-turner audiobooks to the gym and raised early visits 51% versus control. The pairing is the method — not more willpower, and not a book title used as a page.
Key takeaways
- Temptation bundling is pairing an instantly gratifying want with a delayed-reward should, and only allowing the want during the should (Milkman, Minson and Volpp, 2014).
- In a 226-person university gym trial, locking page-turner audiobooks to the locker raised first-week visits 51% versus control; encouragement without the lock raised them 29% (Milkman, Minson and Volpp, 2014).
- The gym-visit increase in that 2014 trial decayed over nine weeks, especially after Thanksgiving — a locked pair is a starter, not a personality transplant (Milkman, Minson and Volpp, 2014).
- Immediate rewards predict persistence on long-term goals more strongly than delayed rewards, even when people say they are doing the activity for the delayed payoff (Woolley and Fishbach, 2016; 2017).
- A later 6,792-person gym program found that a free audiobook plus bundling encouragement raised weekly workout likelihood 10–14%; an audiobook alone leaked enough of the idea to move the needle too (Kirgios et al., 2020).
What is temptation bundling?
Temptation bundling is the habit of coupling a guilt-tinged want with a delayed-payoff should, then restricting the want so it only happens during the should — not a bribe after you finish, and not the title of a book used as a page. Katherine Milkman, Julia Minson and Kevin Volpp named the method in a 2014 field experiment at a university gym: page-turner audiobooks were the want; aerobic exercise was the should; the lock was a loaned iPod that had to stay in a monitored locker. Full-treatment participants visited 51% more in the first week after intake than control participants. Encouragement to self-impose the same rule, with audiobooks on a personal iPod you could still use on the sofa, lifted visits 29%. The pairing is the method. The book that later popularised the phrase for a general audience stays off this URL. This spoke is the lock you can run without buying a new gadget.
This page hangs off how to build discipline, the Discipline cluster page. Temptation bundling is one device inside that system: you are not topping up a willpower tank; you are making the should pay today. The sister in this cluster is self-discipline examples — what the day actually looks like when a rule is visible. A bundle you announce once and then stream on the sofa is not discipline. It is a slogan with headphones.
Ted O’Donoghue and Matthew Rabin’s 1999 model of present-biased people is the mechanism underneath. Naifs believe tomorrow-them will happily do the should. Sophisticates know tomorrow-them will flinch, and they will pay for a device that ties the hands of that future self. Temptation bundling is a commitment device that does not confiscate money. It confiscates the want until the should is in progress. David Laibson’s 1997 “golden eggs” paper is the older economics of the same fight: people value commitment accounts because unconstrained access lets the present self raid the nest. Here the nest is the next chapter of the novel.
How is this different from a bribe, habit stacking, or just trying harder?
A bribe pays after the should. Habit stacking chains a new action to a cue you already keep. Grit tries to resist the want and do the should anyway. Temptation bundling runs the want and the should at the same time, and forbids the want anywhere else. Those four get mixed in productivity posts because they all mention “rewards.” They are not the same object.
| Practice | When the want happens | What you are doing | What it is not |
|---|---|---|---|
| Temptation bundling | Only during the should | Locking a treat to the hard block | Not a gadget, not a book title |
| Reward after | When you finish | Paying yourself later | A bribe. The sofa still has the show while you stall |
| Habit stacking | After a cue you already do | Cue → new action | No locked want. The stack can be boring on purpose |
| Implementation intention | If situation X, then action Y | Pre-deciding the flinch | A plan, not a treat. Pair them if you want; do not confuse them |
| Trying harder | Never — you refuse the want | White-knuckling the should | The thing the 2014 lock was built to replace |
The last two rows matter. An implementation intention is “After I shut the laptop, I will start the 25-minute block.” That is a situation → action link. Temptation bundling adds: “The episode plays only while the block is running.” You can use both. You cannot substitute one for the other and keep the same mechanism. Angela Duckworth, Katherine Milkman and David Laibson’s 2018 review of self-control strategies puts bundling in the family of devices that change the situation, not the pep talk. Situation-change is the discipline parent talking. Pep talks are what people try first and then call a character flaw when they fail.
Why does the should lose to the want?
The should loses because its payoff arrives later, while the want pays now — and present-biased people overweight now even when they planned otherwise last night. O’Donoghue and Rabin call people who know this about themselves sophisticates. Sophisticates will restrict future choice. Naifs will not, because they still believe tomorrow-them is a different, more disciplined person. Temptation bundling is a tool for sophisticates. If you keep “forgetting” the rule and watching the show on the sofa, you are not missing a podcast recommendation. You are missing a lock that actually binds.
Kaitlin Woolley and Ayelet Fishbach’s 2016 Journal of Consumer Research paper, and the 2017 follow-up in Personality and Social Psychology Bulletin, make the same point from the other side of the desk. People pick workouts and study blocks for delayed rewards — health, grades, a finished draft. Persistence, in the moment, tracks immediate rewards more tightly than those delayed ones. Choosing the enjoyable version of the exercise, or adding music to a classroom, raised persistence compared with attending to the long-term reason. Temptation bundling is that finding turned into a rule: you do not wait for the should to become fun. You attach a want that already is.
Complementarity is the extra twist in the 2014 design. Listening to a cliffhanger on a treadmill is not just “exercise plus audio.” The next chapter is only available if you go back. Habit-forming wants make the should pull you in a way a static playlist does not. That is why Milkman, Minson and Volpp pre-tested novels for being hard to put down, then loaded four onto the gym iPod. Neutral lo-fi is not a want. A want you would sneak on the sofa is a want.
The four moves that make it a method
- Name the should. One delayed-reward action you already mean to do — a hard block, a gym bout, a review hour. If you cannot name it in a sentence, you do not have a should. You have a mood. The two-minute rule is the size check on the start, not on the want: the first physical action of the should should be small enough to begin.
- Name a real want. A page-turner, a show, a podcast you actually crave. If you would not protect it from your own sofa-self, it is decoration. Milkman’s participants ranked addictive fiction. Rank yours.
- Lock the want. Only during the should. Headphones stay off the sofa. Download offline if the app will otherwise keep playing in the kitchen. The 2014 full treatment used a physical locker because a suggestion is leaky. Your version can be a rule plus friction: the show lives in an account you only open at the desk.
- Refresh before it goes stale. The 2014 lift decayed. Thanksgiving closed the gym and the pair cooled. A bundle is not a personality. When the novel ends, load the next one before the should is left holding an empty pair of headphones.
None of those four is a personality. They are a weekly shape. Is discipline a skill is the cluster claim underneath: the lock is practice, not a trait you either received or didn’t. Mental discipline is staying put once the pair is running — the headphones are not a licence to wander into a second tab.
Why does an unlocked want cost more than it feels like?
An unlocked want lets you take the treat without the should, so the should keeps losing and the want keeps arriving with a side of guilt — two self-control problems, paid twice. Milkman, Minson and Volpp built the device to hit both at once: more gym, less sofa-novel. When the lock is off, you still get the novel. You just do not get the visits. The cost is not dramatic. It is a quiet week of skipped blocks that never announce themselves as a decision.
The 2014 paper is honest about decay. Regression-adjusted, the full treatment was +0.48 visits per week in week one against a control baseline of 0.94 — the 51% figure. That advantage shrank by about 0.07 visits per week across nine weeks, and Thanksgiving (gym closed) was a visible break. Acland and Levy’s gym-incentive work, which Milkman cites as the comparison class, shows the same pattern: attendance interventions often fade. Temptation bundling is not exempt. Treat the first month as a window, not a proof that the pair is now “who you are.”
After the study, 61% of participants stated a non-zero price to keep gym-only access to an iPod loaded with one tempting novel — they would pay to be restricted. That is the sophisticate’s signature. It is also not a weight-loss result. The paper’s introduction talks about obesity and exercise rates; the outcome they measured was gym visits among 226 university members, 86% undergraduates, average BMI 23.3. Do not quote it as a clinical weight trial. Do quote it as a cheap attendance device that people later wanted to buy.
When the novelty of the want fades, you still need a motivation stack that does not depend on a locked audiobook. Steven Kotler’s The Art of Impossible is the named review for that next layer: stacking curiosity, passion, purpose, autonomy and mastery so the should starts paying its own dopamine. Temptation bundling is the on-ramp. The book is not this page. The productivity library already has the review doing that job.
Two honest caveats. The 2014 sample is one university gym, mostly undergraduates who already belonged to that gym and typically used a stationary aerobic machine. The 51% figure is the first-week, regression-adjusted contrast for the full lock — not a nine-week average, and not a promise about your living room. Kirgios, Milkman, Duckworth and colleagues (2020) later found that simply handing gym-goers an audiobook leaked enough of the bundling idea that extra verbal encouragement added only a modest extra lift. A free want in a should context is already a hint. A hint is not a locker.
How does the pair run in a real week?
Pick one should, one want, one place the want is allowed, and run that pair on the days the should already belongs on the calendar — do not invent a second life. The mistake is building a cinema. The method is attaching Tuesday’s hard block to a show you already cannot stop.
Name the should in the same language you would use on a calendar: “Wednesday 7:30, 30 minutes on the bike,” or “the first 45 minutes after lunch, draft the client memo.” If the should is “be healthier,” you do not have a should. You have a slogan. Personal accountability is the named owner of that sentence. A bundle with no owner is a mood board.
Put the want behind a door the should opens. Offline download that only exists on the gym phone. A streaming profile you do not use in the kitchen. Physical media in a drawer by the treadmill. Milkman’s full treatment used a locker because the researchers could check it. Your version needs some check you cannot laugh off at 9 p.m. Friction on the sofa is the device. Honour-system “I just won’t press play” is the intermediate condition, and it underperformed the lock.
Refresh on a schedule, not when you are already bored. When two chapters remain, queue the next novel. When the season ends, pick the next show before the should is left with news-radio. Decay is the default. Keystone habits is the Habits cluster page if the pair becomes load-bearing — exercise that then changes evenings. That cascade is not guaranteed, and this page does not claim it. Temptation bundling can be a small, local lock and still be worth running.
What are the common mistakes?
- Paying yourself after you finish. A biscuit at the end of the report is a bribe. The report still has to beat the sofa without the biscuit. Bundling is simultaneous. If the want can happen on the sofa, the should has nothing the sofa lacks.
- Picking a fake want. Ambient focus music is not a page-turner. Milkman pre-tested novels for being hard to put down. If you would not steal time for it, it will not pull you to the gym.
- Leaving the want unlocked. The 29% encouragement arm is the warning: a suggested rule, with the iPod in your bag, is weaker than a locker. Design the leak out.
- Bundling a should you do not actually mean. Pairing a thriller with “learn Japanese” when you have no Tuesday slot is costume. Name a should that already has a time, then attach the want.
- Never refreshing the want. The 2014 lift faded. An ended novel plus an open gym membership is the original problem with extra hardware.
- Using the pair as a diagnosis. Skipping the gym with or without headphones is not a clinical finding. This page is a work-and-habits device. How to build discipline stays the parent; this lock is one tool in that system, not a verdict on your character.
When is it more than a pairing habit?
If a locked want still cannot get the should started — sleep debt, a depleted week, or a block you do not actually control — you are on a different layer than headphones. Temptation bundling assumes a should you can physically begin. It will not open a gym that is closed, a job that forbids the block, or a body that cannot take the bout. Do not turn a device into a verdict. This page is not an evaluation, a symptom list, or a treatment.
A second miss is encoding. Students sometimes want the want (a favourite show) bundled with studying material they must later recall without notes. The show can help you sit down. It can also split attention so the sitting-down produces nothing you can use. Pair a want with a should whose success is “I did the bout,” not “I now know the chapter.” Learning pages on this site are for the second job.
A third miss is anxiety dressed as a stack. Adding three shows, two podcasts and a fancy new headset because the last pair “wasn’t motivating enough” is how the lock becomes a shopping trip. One pair. Refresh the want. Leave the gadget aisle.
What does the evidence actually show?
Every figure above traces to a specific published study. Here is the short version of what each one found — and, honestly, what it does not prove.
| Study | Design & size | What it found | Used here for |
|---|---|---|---|
| Milkman, Minson & Volpp (2014), Management Science | Three-arm RCT, 226 university gym members; full lock vs encouragement vs gift-card control | First-week visits +51% (full) and +29% (encouragement) vs control; effects decayed, especially after Thanksgiving; 61% later paid for gym-only iPod access | The method, the lock vs a suggestion, and the decay warning |
| Kirgios, Mandel, Park, Milkman, Gromet, Kay & Duckworth (2020), Organizational Behavior and Human Decision Processes | Four-week gym program, N = 6,792 | Audiobook plus bundling encouragement raised weekly-workout likelihood 10–14% and average weekly workouts 10–12% during and up to 17 weeks after; audiobook alone also moved exercise because it leaked the idea | Scale, and why a free want in a gym context is already a hint |
| Woolley & Fishbach (2016), Journal of Consumer Research | Lab and field studies attaching immediate rewards to health and academic tasks | Attending to immediate rewards, or adding them (enjoyable exercise, classroom music), increased persistence more than attending to delayed payoffs | Why the want has to pay now, not after |
| Woolley & Fishbach (2017), Personality and Social Psychology Bulletin | Studies of resolutions, studying, and exercise; meta-analysis across the set | Immediate rewards predicted persistence; delayed rewards predicted intention more than follow-through | Why “remember why you started” is the weaker lever in the moment |
| O’Donoghue & Rabin (1999), American Economic Review | Theory of present-biased naifs vs sophisticates | People who know they will flinch tomorrow will value restrictions on future choice; naifs will not | Why some people pay for a locker and others “just won’t press play” |
| Laibson (1997), Quarterly Journal of Economics | Theory of hyperbolic discounting and commitment assets (“golden eggs”) | Illiquid assets can be rational when the present self would otherwise raid them | The economic cousin of locking the novel in the gym |
| Duckworth, Milkman & Laibson (2018), Psychological Science in the Public Interest | Review of self-control strategies beyond willpower | Situation-change and commitment devices outperform pep-talk models of restraint | Why this page sits under discipline-as-system, not grit-as-identity |
| Ariely & Wertenbroch (2002), Psychological Science | Experiments on self-imposed deadlines | People self-impose costly deadlines and those deadlines reduce delay, though not as well as evenly spaced externally imposed ones | Contrast: a deadline is another commitment device; it is not a locked want |
Frequently asked questions
What is temptation bundling in simple terms?
Temptation bundling is pairing something you crave now with something that pays later, and only allowing the craving during the later-paying work. Milkman, Minson and Volpp locked page-turner audiobooks to a university gym and saw early visits rise 51% versus control. The pairing is the method — not more grit, not a bribe after you finish, and not a book title used as a page. If the show still plays on the sofa, you do not have a bundle. You have a show.
Is temptation bundling the same as rewarding yourself after you finish?
No. A reward after you finish is a bribe. The should still has to beat the sofa without the treat in the room. Temptation bundling runs the want and the should at the same time, and forbids the want anywhere else. That is why the 2014 full treatment used a locker: the novel was physically unavailable on the sofa. If you eat the biscuit at the desk after the report, you have a bribe. If the biscuit only exists while the report is open, you are closer to a bundle — and you still need a want you actually crave.
Does temptation bundling actually work?
It worked on gym visits in the studies we have, with limits. Milkman, Minson and Volpp (2014) found a 51% first-week lift for a locked audiobook and a 29% lift for encouragement, in 226 university gym members, with decay over nine weeks. Kirgios and colleagues (2020) found 10–14% higher weekly-workout likelihood in a 6,792-person program, and noted that a free audiobook alone already leaked the idea. Neither paper is a weight-loss trial. Neither proves the pair will survive an ended novel. Run it as a device. Do not promote it to an identity.
What should I bundle if I do not go to a gym?
Any delayed-reward should you can physically start, paired with a want you would otherwise take on the sofa. A hard writing block with a favourite podcast that only plays in that document. Ironing with a show that does not play in bed. A weekly review with a coffee you do not drink at the inbox. The 2014 paper used a gym because visits were measurable with a university ID. The method is the lock, not the treadmill. Name a should that already has a time. Then attach one want. Do not invent a second life to have somewhere to put headphones.
How is this different from habit stacking?
Habit stacking chains a new action to a cue you already keep: after the kettle boils, you stretch. There is no locked want. The stack can be boring on purpose. Temptation bundling chains a want to a should and forbids the want elsewhere. You can use a stack as the cue that starts the should, then let the want run only inside that should. Do not wait for a later habit-stacking page before you lock one pair this week. The two methods share a sentence rhythm and not a mechanism.
Will the bundle wear off?
Yes, that is the default in the data we have. Milkman, Minson and Volpp watched the gym lift shrink across nine weeks, especially after Thanksgiving closed the gym. Kirgios and colleagues still saw effects into a 17-week follow-up in a larger program, which is more encouraging and still not permanent. Refresh the want before it ends. If the should never pays anything of its own, you will be hostage to the next novel. Kotler’s Art of Impossible is the named review for stacking intrinsic drivers so the should is not only a vehicle for the show.
Is this just for people who lack willpower?
No. Duckworth, Milkman and Laibson’s 2018 review treats situation-change and commitment devices as the grown-up layer, not as a consolation prize for people who “lack grit.” O’Donoghue and Rabin’s sophisticates are the people who know they will flinch and therefore value a restriction. Paying for a gym-only iPod after the 2014 study is that signature. Trying harder while the show sits unlocked on the sofa is the naif’s plan. This page sits under how to build discipline for that reason: the system is the point, not a verdict on your character.
What should I read after this page?
Start with the Discipline parent, how to build discipline, if you need the system this lock sits inside. Read self-discipline examples if you want the day-to-day look of a visible rule. Implementation intentions cover the if-then cousin. Then the named review: Steven Kotler’s The Art of Impossible, for the motivation stack that remains when a novel ends. None of those replace locking one want to one should this week. The sofa will still play the show if you do not.
Sources
- Milkman, K. L., Minson, J. A., & Volpp, K. G. (2014). Holding the Hunger Games hostage at the gym: An evaluation of temptation bundling. Management Science, 60(2), 283–299. https://doi.org/10.1287/mnsc.2013.1784
- Kirgios, E. L., Mandel, G. H., Park, Y., Milkman, K. L., Gromet, D. M., Kay, J. S., & Duckworth, A. L. (2020). Teaching temptation bundling to boost exercise: A field experiment. Organizational Behavior and Human Decision Processes, 161, 20–35. https://doi.org/10.1016/j.obhdp.2020.09.003
- Woolley, K., & Fishbach, A. (2016). For the fun of it: Harnessing immediate rewards to increase persistence in long-term goals. Journal of Consumer Research, 42(6), 952–966. https://doi.org/10.1093/jcr/ucv098
- Woolley, K., & Fishbach, A. (2017). Immediate rewards predict adherence to long-term goals. Personality and Social Psychology Bulletin, 43(2), 151–162. https://doi.org/10.1177/0146167216676480
- O’Donoghue, T., & Rabin, M. (1999). Doing it now or later. American Economic Review, 89(1), 103–124. https://doi.org/10.1257/aer.89.1.103
- Laibson, D. (1997). Golden eggs and hyperbolic discounting. Quarterly Journal of Economics, 112(2), 443–478. https://doi.org/10.1162/003355397555253
- Duckworth, A. L., Milkman, K. L., & Laibson, D. (2018). Beyond willpower: Strategies for reducing failures of self-control. Psychological Science in the Public Interest, 19(3), 102–129. https://doi.org/10.1177/1529100618821893
- Ariely, D., & Wertenbroch, K. (2002). Procrastination, deadlines, and performance: Self-control by precommitment. Psychological Science, 13(3), 219–224. https://doi.org/10.1111/1467-9280.00441
Related reading on The Growth Reads
The books this method draws on
- The Art of Impossible summary & review — the named unused review; the motivation stack that remains when a locked novel ends
Related articles
- How to build discipline — the Discipline parent; the system this lock sits inside
- Self-discipline examples — the sister; what a visible rule looks like day to day
- Is discipline a skill — the lock is practice, not a trait
- Mental discipline — staying put once the pair is running
- Implementation intentions — the if-then cousin, not a locked want
- What is personal accountability — the named owner of the should
- The two-minute rule — the size check on starting the should
- Keystone habits — if the pair becomes load-bearing; not a new habits parent
Go deeper
How this article was researched
The Growth Reads editorial team writes from primary sources: the books themselves, the authors’ published essays, and the peer-reviewed literature behind the claims. Every study cited above was read in the original publication or its published abstract, and every figure is quoted with its design and limit so you can judge its weight rather than take ours. Where a finding is a university-gym attendance trial, a present-bias model, or a practitioner motivation stack rather than a clinical weight-loss result, we say so in the body rather than in a footnote.
This page is a productivity method, not a book review of any single title, and not medical or clinical advice. Persistent inability to start work across contexts belongs with a clinician, not with a pair of headphones.
How we work: every article on The Growth Reads is built from the books in our library plus the research they rest on, and is reviewed and date-stamped when the evidence changes. Read our full methodology.