Commitment devices work by removing your future self's ability to backslide on goals, using structured accountability and behavioral psychology to turn intentions into consistent action, a strategy therapists often incorporate into cognitive behavioral therapy to help clients build lasting habits and follow through on meaningful change.
What if you could outsmart your own procrastination before it even starts? Commitment devices work by locking in today's good intentions, so your future self has no real choice but to follow through, even on the hardest, least motivated days.
What is a commitment device?
You already know what you want for your life. You want to save the money, finish the workout, skip the late-night scrolling, keep the promise to yourself. The trouble shows up later, at 9pm, when a different version of you shows up to vote on the decision. A commitment device is a way of making sure that version of you doesn’t get to overrule the plan.
What are commitment devices?
A commitment device is an arrangement you set up in advance that makes it harder, or more expensive, for your future self to do the thing you have already decided you don’t want to do. You build it while you’re thinking clearly, before temptation or fatigue shows up. The classic setup is Odysseus and the sirens, which gets its own telling elsewhere, but the basic move is the same one people use today when they freeze a credit card in a block of ice or hand their car keys to a friend before a night out.
The problem a commitment device solves is simple to state and hard to live with: what you want right now and what you want for your life often disagree. Economists call this time-inconsistent preferences, meaning your ranking of two options changes depending on how close the reward sits. A related idea, present bias, describes why: rewards available immediately feel bigger than rewards arriving later, even when the later reward is objectively worth more on paper. Neither of these is a flaw unique to you. They describe a predictable feature of how choosing over time works for most people.
What is the commitment device theory?
Commitment device theory holds that the value comes from removing a choice, not from adding willpower. Instead of trying to out-muscle temptation in the moment, you take the option off the table ahead of time. This is why the devices work even for people who don’t think of themselves as especially disciplined. The device does the holding, so you don’t have to.
What a commitment device is in economics
Asking what a commitment device is in economics gets an unusual answer, because most economic models assume more options are always better. A commitment device is a voluntary constraint on your own future choice set, deliberately narrowing what later-you is allowed to do. Economists sort these into two categories. Hard commitments make the unwanted action nearly impossible, like locking money away until a set date. Soft commitments make it embarrassing, costly, or simply inconvenient, like telling a friend your goal so backing out means admitting it.
None of this is a character judgment. Needing a commitment device doesn’t mean you lack willpower, it means you understand how choosing works well enough to design around it. Therapies like acceptance and commitment therapy work with a similar idea, helping people commit to actions that match their values even when a passing feeling argues otherwise.
Odysseus and the sirens, the oldest commitment device we know about
In Homer’s Odyssey, Odysseus wants to hear the sirens sing without steering his ship onto the rocks their song lures men toward. His solution is not willpower. He has his crew tie him to the mast and plug their own ears with wax, so he can listen while losing any ability to act on what he hears. This is the Odysseus commitment device, and it still gets cited because it contains every piece the concept needs.
The first piece is timing. Odysseus makes the arrangement before the sirens are audible, while he still agrees with his own plan and has nothing pulling him toward the rocks. The second piece is who holds the enforcement. He does not trust himself to stay tied up once the temptation arrives, so he hands control to his crew and orders them in advance to ignore anything he says once the singing starts. The third piece is what he does not attempt: he never tries to want the sirens less. He accepts that he will want them badly and removes his own capacity to act on that wanting, rather than fighting the desire itself.
Advance setup, external enforcement, an accepted loss of freedom: those three features are why behavioral economics keeps returning to this story, and research tying the myth directly to real-world commitment savings behavior treats it as more than metaphor. The idea shows up in commitment device game theory too, because a promise you cannot break carries more weight than one you can. Credibly removing your own options changes what other people, or your future self, can expect from you next.
How commitment devices actually work
Once you strip away the specific product or app, every commitment device is pulling one of a small number of levers. Understanding those levers, rather than memorizing examples, is what lets you size up any device someone hands you and decide whether it will actually hold. Here is how commitment devices work, mechanism by mechanism.
Restricting your options
The most direct mechanism removes the unwanted choice from the table entirely. Money locked in an account you cannot touch until a set date is not discouraging withdrawal, it is preventing it. This differs from willpower in a basic way: you are not resisting temptation moment to moment, because the option simply is not there to resist. It is the same logic behind structuring an environment so a poor choice cannot be made, rather than trusting yourself to decline it each time it appears.
Imposing a cost on the choice you do not want
A softer version leaves the unwanted action possible but attaches a price to it. A forfeited deposit, a donation routed to a cause you find distasteful, an inconvenience you have to walk through before you can act: none of these make the choice impossible, they just make it more expensive than it was a moment ago. This is where loss aversion does its work. Losing something you already hold tends to feel worse than gaining the same amount feels good, which is why handing over a deposit and then losing it stings more than simply missing a bonus you never had. Devices built on cost are borrowing that asymmetry on purpose.
Using witnesses and reputation
Telling specific people what you intend to do converts a private lapse into a social event. The mechanism here is not financial at all, it is reputational: once someone else is watching, backing out costs you standing, not money. A related mechanism works without any stake at all, by adding friction to the unwanted path and removing it from the wanted one, so the easier option quietly becomes the better one.
The strongest devices often stack two or three of these mechanisms at once, pairing a real cost with a witness, for instance. But stacking too many tends to make the arrangement brittle, since each added condition is one more point where the whole thing can snap or feel punitive rather than useful. When you are offered a device, ask three questions: what does it remove, what does it cost, and who actually enforces it. Structuring choices this way has a cousin in dialectical behavior therapy, which also works by reshaping the options available in a moment rather than relying on willpower alone.
Commitment device examples people actually use
The mechanisms behind commitment devices only matter if you can point to something and say, that’s one. Here’s what they look like when they show up in an ordinary week.
Financial commitment devices
Money is the oldest lever. A savings account with a withdrawal penalty or a fixed maturity date makes spending a decision you have to fight for, not a click. Automatic transfers scheduled for payday move money before you see it, so there’s nothing left to talk yourself out of saving. Cash-only envelopes cap what you can spend in a category once the bills inside them are gone, and some people leave their cards at home on purpose, so a purchase requires a trip back rather than a tap.
Digital and attention commitment devices
Website and app blockers work the same way, except the friction is a forced delay before you can turn them off rather than a bank’s rules. Leaving your phone charging in another room does the same job without any software at all. Some people go further and use a single-purpose device, like a phone with no apps or a stripped-down music player, so the temptation isn’t sitting in your pocket to begin with.
Health habits borrow the same logic. Prepaid class packages turn a decision to skip into a decision to waste money already spent. A training partner who expects you at a set time makes flaking a social cost, not just a private one. Buying food in single portions instead of in bulk means the excess simply isn’t in the house to reach for.
Commitment contracts and stickK
stickK, an online platform, lets users sign a commitment contract, name a referee to verify their progress, and put money at stake that goes to a charity, an anti-charity, or a friend if they fail. Public pledges do something similar without money: announcing a goal to a named group, or posting progress somewhere the same audience will notice if it stops.
James Clear’s Atomic Habits frames commitment devices as one-time decisions that lock in better behavior later: buying smaller plates, cancelling a subscription, setting a timer that shuts a machine off. Institutions build the same structure in, through automatic retirement contributions, deposit-based programs, or deadlines someone else sets for you.
None of this works because of the brand attached to it. The strength of any example here comes from the mechanisms it borrows, not from stickK’s name or a bestselling book’s endorsement.
The psychology underneath, present bias and the two selves problem
Commitment devices work because they stop treating self-control as a single, unified will and start treating it as a negotiation. Consider two versions of yourself. One is the planner, who sits down on a quiet Sunday and decides that next month’s version of them will exercise, save money, or finally send the difficult email. The other is the doer, who shows up in the actual moment of choice, tired or distracted or staring at something more immediately rewarding. This is the planner doer model, and it matters because these two selves do not have equal power when the moment actually arrives.
The planner has good intentions and a long view, shaped in part by present bias, the tendency to weigh immediate rewards more heavily than future ones. But the planner cannot stand next to the doer at every decision point, whispering reminders about the goal. Its only real leverage is arranging, in advance, what the doer will find in the room. A commitment device is that leverage put into practice.
This reframes what failure usually means. A broken resolution is rarely proof of weak character. More often it is a plan handed to a self who was never actually positioned to keep it, a design problem disguised as a discipline problem. That is also why willpower-first advice tends to disappoint: it asks the doer to win a fight the planner set up badly, with no support and no change in surroundings.
There is a real cost here, too. Every commitment device trades away flexibility, and sometimes the future self who wants an exit has a legitimate reason, not just weaker resolve. This is where sophistication versus naivety comes in: people who expect to be tempted build structure around that expectation, while people who assume they will simply choose correctly in the moment often skip the structure entirely. Self-compassion belongs in the design itself, not as a prize for succeeding, because a system you can only sustain by punishing yourself for slipping rarely holds for long.
What the research shows about commitment savings
Most of what people say about willpower is opinion. This part is not. Economists ran a field experiment on a commitment savings product in the Philippines, offering bank clients a savings account that locked away withdrawals until they hit a date or a dollar amount they picked themselves. Researchers then compared their balances against clients who were never offered the account. It is one of the clearest real-world tests of whether a commitment device actually changes financial behavior, rather than just sounding like it should.
The effect was not small. Clients who opened the commitment savings account increased their savings balances by 81% in the first year, compared with those who kept an ordinary account. For one typical account holder, that means the simple act of restricting your own access to your money, on terms you set, moved far more cash into savings than good intentions alone managed to. The effect faded over time, settling to a 33% increase after two and a half years, which suggests the lock helps most early on, when the pull toward spending is strongest.