Commitment Device Apps: How They Work and Which One Fits
Commitment device apps enforce an arrangement between your present self and your future self: you lock a decision in now, while you're motivated, and the app attaches a real cost — usually money — to backing out later, when you won't be. Four are worth knowing: Beeminder charges you when you fall off a tracked goal, StickK backs contracts with stakes and referees, Forfeit takes staked money when you fail a proof-required task, and Paywall charges you to unlock blocked apps early.
The classical reference is Ulysses, who had his crew tie him to the mast so he could hear the Sirens without steering into the rocks. He didn't trust his future judgment, so he removed it from the equation — because he already knew what his future self was going to do. The Greeks needed a ship and a crew. You need an app and a card on file.
Below: the evidence, the design principles that separate devices that work from ones that don't, and which of the current apps — Beeminder, StickK, Forfeit, and Paywall — fits which goal.
Do money stakes actually work?
The best-known test of financial commitment is a smoking-cessation trial published in the New England Journal of Medicine (Halpern et al., 2015). Participants could put their own money on the line as a deposit, returned only if they quit. Among those who accepted the deposit arrangement, quit rates were about 13 percentage points higher than in reward-only programs — where the money was a prize rather than something they could lose.
The catch, and it's an honest one: fewer people were willing to sign up for the deposit version in the first place. Losing your own money is unpleasant, which is precisely why it works and precisely why people avoid agreeing to it. A commitment device is something you choose on a good day to protect yourself on a bad one.
For screen-time goals specifically, there's also evidence that plain friction helps: a 2023 study in PNAS found that a simple forced pause before opening a target app reduced openings by about 57%. Friction works. The question each app answers differently is what happens when you decide to push through the friction anyway — and whether that moment costs you anything.
Three design principles (and where apps get them wrong)
1. The stake must be real. A streak counter is not a commitment device. Neither is a notification you can swipe away. If breaking the commitment costs you nothing you actually value — money, in most of these apps — your future self will treat the rule as a suggestion. Your future self is good at that.
2. The consequence must be immediate. A fee that arrives at the end of the month is an abstraction. A charge that lands at the exact moment you're about to lapse is a decision you have to make with the cost in front of you. The closer the price sits to the moment of temptation, the more work it does.
3. The stake must be right-sized. This one has its own famous study. In "A Fine Is a Price" (Gneezy & Rustichini, 2000), day-care centers introduced a small fine for parents who picked their kids up late. Late pickups went up. The fine was so small it converted a guilty obligation into a cheap purchase — parents read it as a price list and paid it happily. Too-small stakes don't just fail; they license the behavior. But a stake large enough to genuinely hurt gets abandoned after one bad week. The workable range is in the middle: enough to sting, capped so it can't spiral. We cover the full economics in does putting money on the line actually break phone habits.
The apps, by what you're trying to change
| App | Best for | The mechanic | Where the money goes | Platforms |
|---|---|---|---|---|
| Beeminder | Quantifiable long-term goals (workouts, words written, weight) | Track data against a "bright red line"; derail and you're charged, starting at $5 and escalating up a ladder ($5 → $10 → $30 → … up to $7,290, cappable) | Beeminder | Web, iOS, Android |
| StickK | Any goal you can self-report, especially if social pressure motivates you | Sign a commitment contract with optional money stakes (minimum $5 per reporting period) and an optional referee | Charity, "anti-charity" (a cause you oppose), or a friend | Web, iOS, Android |
| Forfeit | One-off tasks and daily habits that need proof | Stake an amount on a task with a deadline; submit photo/video/GPS/health-data evidence, reviewed by a human adjudicator; fail and the stake is forfeited | Forfeit | iOS, Android |
| Paywall (joinpaywall.com) | Screen time — the apps you open without deciding to | Blocks chosen apps via Apple's Screen Time controls; unlocking early costs a price you set ($0.50 floor, typically $1–2), with a monthly cap | Paywall | iOS (pre-launch, waitlist open) |
Beeminder — for people with data
Beeminder is the veteran of the category and the most flexible. If your goal produces a number — pages read, kilometers run, hours logged — Beeminder graphs it against a commitment line, and the moment you fall below the line, you pay. It's free to join with three free goals; charges only happen when you derail. The escalating pledge ladder is clever: each failure raises the stake until you find the number that actually stops you. Pick Beeminder if your goal is measurable and long-horizon. Its weak spot for screen time is timing — the charge arrives after the derailment, not during the temptation.
StickK — for people who respond to an audience
StickK came out of Yale behavioral economics research and is built around the commitment contract: a stated goal, a reporting schedule, optional stakes, and optionally a referee who verifies your reports. Its signature move is the anti-charity — your lost stakes can go to an organization you oppose, which for some people is far more motivating than losing money to a neutral party. The platform itself is free. Pick StickK if accountability to another person is what moves you. Self-reported goals rely on your own honesty, which is either fine or the entire problem, depending on the user.
Forfeit — for people who need proof, not promises
Forfeit is the strictest of the three generalists: you stake money on a specific task with a specific deadline, and you must submit evidence — a timestamped photo, a video, GPS or health data — that a human reviewer checks. You pay only when you fail. Pick Forfeit for discrete, provable tasks: gym sessions, morning routines, finishing the thing. Pricing for its premium tiers changes; check current rates in the app.
Paywall — for the specific problem of your phone
Paywall applies the same logic to one narrow domain: the apps you open thirty times a day without deciding to. You choose the apps, Paywall walls them off using Apple's own Screen Time controls, and opening one early costs real money at that exact moment — a price you set in advance, floor of $0.50, with a monthly cap so a bad week stays a line item instead of a crisis. It's the "immediate" principle taken literally: the charge and the temptation happen in the same second. Paywall is iOS-only and currently pre-launch — the waitlist is open ahead of its App Store release. For the broader landscape of blockers with and without stakes, see best app blockers for iPhone; for the pay-per-unlock category specifically, see apps that charge you to unlock.
How to choose
Match the device to the failure mode, not the other way around:
- You abandon long projects → Beeminder. The red line doesn't care that it's been a busy month.
- You keep promises to people, not to yourself → StickK, with a referee.
- You negotiate with yourself about whether you "really" did it → Forfeit. The adjudicator doesn't negotiate.
- Your problem is reflexive app-opening → a screen-time-specific device with an immediate price, like Paywall's screen time tax model.
One honest note that applies to all four: no commitment device is inescapable. Every app can be deleted; every contract can be canceled. The point isn't to make lapsing impossible — it's to make it cost something at the moment it happens, so the easy path and the intended path point the same way more often.
FAQ
What is a commitment device, in plain terms? A rule you set for your future self, with a built-in cost for breaking it. You decide while calm; the device enforces while you're not.
Do commitment device apps actually work? The strongest evidence is for deposit contracts: in the NEJM smoking trial, people who accepted money-at-stake arrangements quit at rates about 13 percentage points higher than reward-only programs. The main limitation is that you have to be willing to sign up.
What's the best commitment device app for screen time? Beeminder can track screen-time data, and Forfeit can stake tasks around it, but the timing argument favors tools where the cost lands at the moment of temptation. Paywall is built specifically for that; it's currently in pre-launch waitlist.
Are commitment apps free? StickK's platform is free (you only lose money you stake). Beeminder is free until you derail — the first pledge level is $5. Forfeit charges only on failure, plus optional premium tiers. Paywall charges per early unlock at a price you set.
Isn't paying money to an app when I fail just losing money? It's buying enforcement. The NEJM evidence suggests the possibility of loss is what moves behavior — the design goal is that you pay rarely, and the rest of the time the stake does its work silently.
Try Paywall
If your particular Siren is a feed, the mast is available in app form. Paywall blocks the apps you choose and charges you real money to open them early. Resisting is free.