Why Odysseus Tied Himself to the Mast
You set a goal on Monday and abandon it by Thursday. Not from laziness – from being human. Behavioral economists have a name for the gap between what you plan to do and what you actually do: present bias. As Harvard economist David Laibson demonstrated in his influential 1997 study, people systematically overvalue immediate rewards and discount future ones, even when they know better [1]. Commitment devices – structures that make it costly or difficult to quit – are the most direct fix for this problem. They don’t rely on willpower. They replace it. This guide covers the specific mechanisms behind commitment devices for goals, from financial stakes to identity-based self-binding strategies, so you can pick the right one for the goal you keep dropping.
Commitment device is a choice or arrangement made in the present that restricts future options, making it harder or more expensive to abandon a goal. Unlike general accountability – which relies on another person checking in – a commitment device binds the goal-setter through self-imposed structure, financial stakes, or social contracts.
Present bias is the tendency to overvalue immediate rewards relative to future ones, causing a predictable gap between long-term plans made in the present and choices made in the moment. Present bias is driven by hyperbolic discounting, where the subjective value of rewards drops sharply as their timing shifts from now to soon, but much less steeply between future time periods.
What You Will Learn
- The psychology behind why commitment devices change behavior when motivation fails
- The difference between hard and soft commitment devices – and when each type fits
- How to build a Commitment Stack – our framework for layering self-binding strategies
- Five commitment device examples you can set up today
- When commitment devices backfire and how to avoid common traps
Key Takeaways
- Commitment devices work by making goal abandonment costly rather than relying on willpower or motivation.
- Hard commitment devices (financial stakes, restricted access) outperform soft ones for high-temptation goals.
- Self-imposed deadlines improve performance, though externally imposed deadlines are even more effective [2].
- The Commitment Stack framework layers multiple self-binding strategies for stronger follow-through.
- Precommitment strategies counter present bias – the tendency to overvalue immediate comfort over future gains [1].
- Public pledges activate identity-based motivation, turning a private goal into a social contract.
- Financial commitment devices increased savings by 81% in a randomized controlled trial [3].
- Combining financial and social commitment devices produces higher success rates than either alone [4].
Why do commitment devices work when willpower doesn’t?
The logic of commitment devices starts with a simple observation: your present self and your future self want different things. Nobel laureate Thomas Schelling called this “egonomics” – the idea that a single person contains multiple competing selves, each with their own priorities [5]. Your Monday-morning self plans to exercise after work. Your 5 PM self wants the couch. Commitment devices work by letting the planning self constrain the tempted self before temptation arrives.
This isn’t just philosophy. Laibson’s hyperbolic discounting model [1] predicts that people don’t discount the future at a steady rate. Instead, they sharply overweight the present moment compared to any future period. A dollar today feels worth far more than a dollar tomorrow, but a dollar in 30 days and a dollar in 31 days feel roughly the same. That asymmetry creates a predictable pattern: you make great plans for “future you” and consistently break them when the moment arrives.
Commitment devices interrupt that pattern. They shift the cost of quitting from the future (where you’ll barely feel it) to the present (where it stings). According to MIT researchers Dan Ariely and Klaus Wertenbroch, in their 2002 study, students who set their own binding deadlines for papers – with real grade penalties for missing them – performed better than students given a single end-of-term deadline [2]. The self-imposed cost changed the equation.
Ariely and Wertenbroch found that students who set binding self-imposed deadlines with real grade penalties performed better on academic tasks than students given a single final deadline [2].
But here’s the nuance Ariely found: students didn’t set their deadlines optimally. They spaced them out, but not evenly enough. People recognize their own self-control problems and attempt to fix them through precommitment strategies, yet they consistently underestimate how much binding they actually need. That finding matters. It means a commitment device you design for yourself will probably be too lenient – something to keep in mind when building your own. For more on how to connect goal tracking systems with the enforcement layer commitment devices provide, that guide is a natural companion to this one.
Hard vs. soft commitment devices: which type do you need?
Bryan, Karlan, and Nelson’s 2010 review in the Annual Review of Economics drew a key distinction that shapes how you should think about behavioral commitment [6]. Hard commitment devices impose tangible costs for failure – money lost, access restricted, consequences enforced. Soft commitment devices create psychological costs – identity friction, social discomfort, broken streaks. Both work. But they work for different situations.
Hard commitment device is a precommitment structure that imposes tangible, external costs for failing to meet a goal – most commonly financial loss, restricted access to resources, or enforced consequences that activate regardless of the goal-setter’s willingness to pay them. Hard devices are distinct from soft devices in that the cost of failure is objective and immediate rather than psychological.
Soft commitment device is a precommitment structure that creates psychological or reputational costs for failing to meet a goal – most commonly identity friction, social discomfort, or the loss of a streak. Soft devices differ from hard devices in that the cost of failure is internalized rather than enforced by an external mechanism.
| Feature | Hard Commitment Devices | Soft Commitment Devices |
|---|---|---|
| Mechanism | Financial loss, restricted access | Identity pressure, social friction |
| Cost of failure | Tangible (money, resources) | Psychological (guilt, reputation) |
| Best for | High-temptation, clear-outcome goals | Ongoing habits, identity-level change |
| Examples | Deposit contracts, locked savings, deadline penalties | Public pledges, streak tracking, identity labels |
| Risk | Excessive stress, punitive spirals | Too weak for strong temptations |
| Research backing | Ashraf, Karlan, Yin 2006 [3]; Gine, Karlan, Zinman 2010 [4] | Bryan, Karlan, Nelson 2010 [6] |
Hard commitment devices are most effective for goals with a clear binary outcome – quit smoking, save a specific amount, finish a project by a deadline. The CARES smoking cessation program tested by Gine, Karlan, and Zinman offered Filipino smokers a savings account where deposits were forfeited if they failed a nicotine test after six months [4]. Eleven percent signed up voluntarily, and those who did were more likely to quit. The financial stake changed behavior that years of intention hadn’t.
Soft commitment devices suit goals that are more continuous and identity-driven. Telling people “I’m a runner” creates different psychological pressure than putting $500 on the line. The bind is real, but it works through who you believe yourself to be rather than what you stand to lose. For ongoing habits like exercise, reading, or creative work, soft devices often sustain motivation longer – since you can’t pay a fine every day for the rest of your life. This connects directly to precommitment psychology research, which explores how identity and self-binding interact.
How do you build a Commitment Stack?
Most people try one device, find it insufficient, and conclude that the whole approach doesn’t work. That’s like trying one ingredient and deciding cooking doesn’t work. We call the better approach the Commitment Stack – layering multiple self-binding strategies so each one covers the gaps left by others.
Commitment Stack is a layered combination of hard and soft commitment devices applied to a single goal, where each layer addresses a different failure mode. A financial stake handles temptation, a public pledge handles accountability, and an environmental change handles friction – together creating a stronger bind than any single device alone.
The Commitment Stack works on a simple principle from Schelling’s self-command research [5]: no single constraint beats a determined present self. But stacking constraints makes the escape routes progressively harder to find. Think of it as closing doors rather than locking one.
Layer 1: Environmental commitment (remove the easy exit)
Before adding stakes or social pressure, change the environment. Schelling himself used this as his first principle of self-command [5] – don’t keep cigarettes in the house, don’t store junk food at eye level, don’t leave your phone next to the bed. These devices increase the friction of quitting. They don’t make quitting impossible, just annoying enough that your present self often can’t be bothered.
For a writing goal, this might mean an app like Cold Turkey that blocks your browser during writing hours. For a savings goal, it might mean setting up automatic transfers on payday so the money leaves before you see it. Environmental commitment devices work by making the default action align with the goal, so deviation requires active effort rather than passive drift.
Layer 2: Social commitment (make the goal visible)
Public pledges tap into something deeper than embarrassment. When you tell people “I’m training for a marathon” or “I’m writing a book,” you’re creating what psychologists call an identity claim. Walking that back carries a cost that goes beyond what anyone says to you – it’s the cost to your self-concept. This layer sits between soft and hard binding. The stakes are real, but they’re reputational rather than financial.
A word of caution: research on accountability partner strategies and Locke and Latham’s extensive work on goal-setting suggest that public pledges work best when they’re specific (“I’ll run 20 miles per week”) rather than vague (“I’m getting healthier”) [8]. Vague pledges are too easy to reinterpret. Specific ones create a clear pass/fail that your social circle can actually track, which connects to the broader follow-through framework for sticking to goals.
Layer 3: Financial commitment (raise the cost of failure)
This is where commitment devices get their sharpest teeth. Platforms like stickK let you deposit money that goes to a charity – or an “anti-charity” you disagree with – if you fail to meet your goal [4]. The CARES study found that smokers randomly offered the commitment contract were 3 percentage points more likely to pass a cotinine test after six months than control groups [4]. The effect persisted in surprise follow-up tests at twelve months.
This layer works through loss aversion – the well-documented finding that losing $50 feels roughly twice as painful as the pleasure of gaining $50 [7]. You’re not bribing yourself to succeed. You’re making failure hurt in a way your present self actually respects. The Ashraf, Karlan, and Yin study on commitment savings accounts in the Philippines showed this clearly: clients who locked their savings in restricted-access accounts saw savings balances 81 percentage points higher than the control group after twelve months [3].
A randomized controlled trial by Ashraf, Karlan, and Yin found that clients who locked savings in restricted-access accounts had savings balances 81 percentage points higher than the control group after twelve months [3].
A related approach is temptation bundling, where an immediately enjoyable activity – such as an audiobook you love – is paired with the goal activity – such as exercise – so the reward is only available during the target behavior. A 2014 field experiment by Milkman, Minson, and Volpp at the University of Pennsylvania found that participants who could only access tempting audiobooks at the gym visited significantly more often than control participants, demonstrating that pairing present pleasure with future-focused behavior strengthens follow-through [9].
Commitment Stack Builder
Check off each layer you’ve added to your goal:
One layer is a nudge. Two layers is a commitment. Three layers is a bind.
Five commitment device examples you can set up this week
These commitment device examples range from identity shifts to formal financial contracts, ordered from softest to hardest. Pick the intensity that matches your goal’s difficulty and your history of sticking to goals. If you already track progress for personal goals, think of these as the enforcement layer on top of tracking.
1. The deadline deposit
Give a trusted friend an amount of money that matters to you – not devastating, but enough to sting. Set a specific deadline and a measurable outcome. If you hit it, you get the money back. If not, it goes to a cause you’d rather not support. This mirrors the structure Ariely and Wertenbroch found effective [2], but with a sharper edge than self-imposed academic deadlines.
2. The calendar lock
Block time on your calendar for your goal and treat it as immovable – the same way you’d treat a meeting with your boss. Then add the environmental layer: use website blockers, put your phone in another room, or physically go to a location dedicated to the task. The commitment isn’t the calendar entry itself. It’s the environmental structure you build around it that makes deviation effortful.
3. The public scorecard
Share your weekly progress with a specific audience – a group chat, a social media post, or a shared spreadsheet with an accountability partner. The key is specificity and regularity. “I’ll post my word count every Sunday” creates a recurring social obligation. Vague updates like “going well” don’t create enough friction to qualify as a real commitment device.
4. The identity declaration
This is the softest device on the list, yet surprisingly durable. Instead of saying “I’m trying to write more,” say “I’m a writer.” Instead of “I want to get fit,” say “I’m an athlete in training.” Identity-based commitment devices reframe goal pursuit as self-expression rather than self-denial, which reduces the psychological cost of maintaining the commitment over months and years. The shift from “what I do” to “who I am” creates a self-binding strategy that doesn’t expire or get spent. Pairing identity declaration with a habit-building approach extends its durability even further.
5. The commitment contract
The strongest form. Write a formal contract specifying your goal, the deadline, the measurement criteria, the financial stake, and the referee who verifies the outcome. Platforms like stickK handle the mechanics, but you can do this with a handwritten agreement and a trusted friend. The formality matters – it activates a different level of psychological seriousness than a casual verbal promise. The CARES study structure confirms that combining financial stakes with external verification produces the strongest results [4].
| Commitment Device | Type | Strength |
|---|---|---|
| Deadline deposit | Hard | High |
| Calendar lock | Soft + Environmental | Medium |
| Public scorecard | Soft + Social | Medium |
| Identity declaration | Soft | Low-Medium (but long-lasting) |
| Commitment contract | Hard + Social | Very High |
| Commitment Device | Setup Time | Best For |
|---|---|---|
| Deadline deposit | 10 minutes | Project deadlines, one-time goals |
| Calendar lock | 15 minutes | Daily habits, creative work |
| Public scorecard | 5 minutes | Ongoing goals, fitness, learning |
| Identity declaration | 1 minute | Lifestyle changes, long-term shifts |
| Commitment contract | 30 minutes | High-stakes goals, breaking bad habits |
Tools that enforce commitment contracts
If you want a platform to handle the enforcement mechanics, three tools stand out. stickK was built by Yale economists Dean Karlan and Ian Ayres and implements the anti-charity contract mechanism directly: you name a beneficiary you genuinely oppose, deposit money, and a referee verifies your outcome [4]. Beeminder is designed for quantified goals – the platform automatically charges your card when your tracked data (steps, words written, hours logged) falls below your commitment line, removing any need to rely on a referee’s judgment. Forfeit uses a photo-proof model: you submit photographic evidence of completing the task, and a reviewer verifies it before releasing your stake. For lower-stakes commitment contracts without a platform, a shared Google Sheet with a trusted accountability partner and a clear payment agreement can replicate the core mechanics at no cost.
When do commitment devices backfire?
These tools aren’t a universal fix. Research points to several failure modes worth knowing before you start stacking constraints on yourself. Recognizing when self-binding strategies cause harm matters just as much as knowing when they help.
Overly punitive commitment devices create anxiety that undermines performance rather than improving it. If the financial stake is so high that failure feels catastrophic, you may avoid the goal entirely rather than risk falling short. The point is to raise the cost of quitting above the cost of continuing – not to terrorize yourself. Ariely’s finding that self-imposed deadlines help but aren’t set optimally [2] suggests starting with moderate stakes and adjusting upward if needed.
A second risk: commitment to the wrong goal. Locking yourself into a target that turns out to be misguided – or that circumstances make impossible – can trap you in a losing game. Good self-binding structures include an escape clause for genuine changes in direction, not for Thursday-evening laziness. Schelling noted this tension directly: the same mechanism that prevents you from quitting a diet can prevent you from stopping a project that should be abandoned [5].
Third, some people respond to financial stakes by finding loopholes rather than doing the work. Bryan, Karlan, and Nelson found that many commitment device users chose minimal or zero-dollar wagers, which functionally strips out the loss aversion mechanism that gives these devices their power [6]. That’s fine for some goals, but if the core challenge you’re facing is breaking through procrastination, a device without teeth probably won’t cut it.
| Failure Mode | Signal | Fix |
|---|---|---|
| Stakes too high | Avoiding the goal, anxiety, dread | Lower the financial amount or switch to a social-only device |
| Wrong goal locked in | Circumstances changed, goal no longer relevant | Build in a quarterly review clause before setting the commitment |
| Loophole exploitation | Technically meeting the letter but not the spirit | Add a referee who judges spirit, not just metrics |
| Motivation crowding | External stakes replaced intrinsic interest | Use soft devices (identity, social) alongside or instead of financial ones |
| Lenient referee | Commitment contract exists but the friend never enforces it | Choose a referee with a professional relationship or use a structured platform like stickK; avoid referees who prioritize your feelings over your goals |
Ramon’s Take
I should be better at this than I am. Here’s what I’ve learned from struggling with it. For years, I used motivation as my primary tool for sticking to goals – and motivation is rarely a reliable binding mechanism. It’s weather. Some days it shows up, most days it doesn’t.
The first self-binding tool that actually changed my behavior was embarrassingly simple: I told three coworkers I’d have a project proposal done by Friday and asked them to check in. The thought of showing up empty-handed in front of people I respected was far more motivating than any deadline on my calendar.
I’ve since tried financial stakes with mixed results. A $100 bet with a friend got me to finish a certification I’d been putting off for months, but a $200 commitment to exercise four times a week just made me anxious – the stakes were too high for something that required daily willingness rather than a single push. Hard devices work best for goals with a clear finish line.
For ongoing habits, I do better with the identity layer. Calling myself “someone who writes every morning” has been more durable than any contract I’ve put my name on. The Commitment Stack concept came from noticing that my successful goals always had multiple layers – environmental changes plus social pressure plus some kind of tracking. My failures always relied on a single mechanism. Start with your environment, add a person, and only reach for the financial stakes if the first two layers aren’t enough.
Commitment Devices Conclusion: Bind Your Future Self
Commitment devices don’t make goals easier. They make abandoning them harder – and that distinction matters. The research from Laibson, Ariely, Schelling, and others points to a single insight: your planning self is smarter about your goals than your in-the-moment self. Precommitment strategies give your planning self the authority to set the rules before temptation shows up. The Commitment Stack – layering environmental, social, and financial self-binding strategies – creates a structure that no single weak moment can dismantle.
The goal isn’t to remove all freedom. It’s to remove the specific freedoms you’ll use to sabotage yourself.
Next 10 Minutes
- Pick one goal you’ve been struggling to maintain and identify which layer of commitment is missing (environmental, social, or financial).
- Tell one specific person a specific, measurable version of that goal – not “I want to get better at X” but “I’ll do X by Y date.”
This Week
- Set up one environmental commitment device (website blocker, automatic transfer, phone in another room during work hours).
- Create a formal commitment contract for your highest-priority goal – written down, with a referee, a deadline, and a defined stake.
- Schedule a quarterly review date to check whether your commitment devices still match the right goals.
There is More to Explore
For a broader look at tracking and measurement systems, explore our goal tracking systems guide. If you’re curious about the psychology behind why self-binding works differently from external pressure, our guide on precommitment psychology digs into the mechanisms. If you need help setting a strong goal to commit to before using these devices, our goal setting frameworks guide covers the foundations. And for practical systems that help with following through on goals, that framework pairs well with the commitment devices covered here.
Related articles in this guide
- Dependency mapping for goals
- Goal setting frameworks: proven systems for success
- Goal setting systems from psychology you have not tried
Frequently Asked Questions
What is the difference between a commitment device and an accountability partner?
A commitment device is a self-imposed structure that makes goal abandonment costly. An accountability partner, by contrast, is a person who checks on your progress. Commitment devices work even without another person involved – a locked savings account or a website blocker functions on its own. Accountability partners add social pressure but depend on the other person’s consistency and willingness to hold you to your word. Many effective setups combine both approaches [6].
How much money should you put at stake in a financial commitment device?
Enough to sting if you lose it, but not so much that the fear prevents you from starting. Bryan, Karlan, and Nelson’s review found that many users chose minimal wagers, which strips out the loss aversion that makes these devices effective [6]. A practical starting point is 1-3% of your monthly income per goal. You can increase the stake after a successful first round if the initial amount felt too easy to dismiss.
Do commitment devices work for habits or only for one-time goals?
Both, but the type of device matters. Hard financial commitment devices work best for clear-endpoint goals. For ongoing habits, repeated financial penalties can create commitment fatigue – resentment that undermines intrinsic motivation over time – while identity-based devices actually strengthen with repetition. A practical protocol: use an identity declaration for the first 30 days, add a public scorecard at day 31, and only introduce financial stakes if the first two layers fail by day 60.
Can commitment devices backfire and make goals harder?
Yes, and one underappreciated failure mode is the accountability paradox: when the person verifying your commitment is too lenient or too close to you emotionally, the device loses its teeth. A friend who cares more about your feelings than your goals will let you off the hook. Choose a referee willing to enforce the rules even when you plead for an exception – professional relationships or structured platforms like stickK often work better than close friends for high-stakes financial commitments.
What is a precommitment strategy in behavioral economics?
A precommitment strategy is any action taken now that restricts your future choices, preventing you from acting against your long-term interest. The concept was formalized by Schelling (1978) and modeled by Laibson (1997) through hyperbolic discounting [1][5]. A classic example: automatic payroll deductions that save money before you can spend it.
Are commitment devices effective for people with ADHD?
Commitment devices can be especially valuable for people with ADHD, since ADHD is associated with heightened difficulty valuing delayed rewards and greater sensitivity to immediate consequences. Environmental commitment devices – removing distractions, using app blockers, setting up automatic routines – tend to work better than financial stakes for ADHD brains. Pairing environmental devices with a short feedback loop (daily check-ins rather than monthly reviews) improves results.
What apps or tools support commitment contracts?
stickK.com is the most research-backed platform, built by Yale economists who study commitment devices [4]. Beeminder tracks quantified goals and charges you when you go off track. Forfeit requires photo proof of completed tasks or you lose your stake. For simpler setups, you can use a shared Google Sheet with an accountability partner, an automatic savings transfer as a savings commitment device, or even a handwritten contract witnessed by a friend.
How does the Commitment Stack differ from just using multiple strategies?
The Commitment Stack differs from random strategy combination by systematically addressing three distinct failure modes – friction (environmental), accountability (social), and temptation (financial) – with each layer covering gaps the others miss. Random combinations may leave the same weakness uncovered twice. The Stack framework ensures all three domains are covered before concluding that commitment devices don’t work for a given goal.
This article is part of our Goal Tracking Systems complete guide.
References
[1] Laibson, D. “Golden Eggs and Hyperbolic Discounting.” The Quarterly Journal of Economics, 112(2), 443-478, 1997. DOI
[2] Ariely, D. and Wertenbroch, K. “Procrastination, Deadlines, and Performance: Self-Control by Precommitment.” Psychological Science, 13(3), 219-224, 2002. DOI
[3] Ashraf, N., Karlan, D., and Yin, W. “Tying Odysseus to the Mast: Evidence from a Commitment Savings Product in the Philippines.” The Quarterly Journal of Economics, 121(2), 635-672, 2006. DOI
[4] Gine, X., Karlan, D., and Zinman, J. “Put Your Money Where Your Butt Is: A Commitment Contract for Smoking Cessation.” American Economic Journal: Applied Economics, 2(4), 213-235, 2010. DOI
[5] Schelling, T.C. “Egonomics, or the Art of Self-Management.” American Economic Review, 68(2), 290-294, 1978. JSTOR
[6] Bryan, G., Karlan, D., and Nelson, S. “Commitment Devices.” Annual Review of Economics, 2, 671-698, 2010. DOI
[7] Kahneman, D. and Tversky, A. “Prospect Theory: An Analysis of Decision under Risk.” Econometrica, 47(2), 263-292, 1979. DOI
[8] Locke, E.A. and Latham, G.P. “Building a Practically Useful Theory of Goal Setting and Task Motivation.” American Psychologist, 57(9), 705-717, 2002. DOI
[9] Milkman, K.L., Minson, J.A., and Volpp, K.G.M. “Holding the Hunger Games Hostage at the Gym: An Evaluation of Temptation Bundling.” Management Science, 60(2), 283-299, 2014. DOI
