The five ways screen-time apps try to stop you

The category looks crowded and mostly is not. There are about five mechanisms, and every product is one of them, occasionally two. What follows is what each one does, who it suits, and the specific way each fails — including ours, which fails in ways worth knowing before you try it.

1. The interruption

What it does. Puts something between the tap and the app: a breathing exercise, a few seconds' pause, a question about why you are opening this. The app still opens. The point is that opening it becomes conscious.

Who it suits. People whose real problem is the unconscious open — the thumb that has already launched the app before any intention formed. This is a much larger group than it sounds, and for them a two-second pause deletes a surprising share of sessions.

How it fails. Habituation, and fast. The pause becomes part of the ritual, then part of the anticipation. Within a few weeks many people report tapping through it without registering the content of the screen. It is also, by design, powerless once you have decided you want in — which describes every session that actually matters.

2. The scheduled block

What it does. Apps or sites become unavailable during set hours, or after a set quantity of use. This is the platform default — iOS Screen Time app limits, Android Digital Wellbeing timers — and what most third-party blockers extend.

Who it suits. People whose use is situational rather than compulsive: focus during work hours, no social apps after eleven. If you generally respect a rule once it exists, this is the cheapest effective option and you already own it.

How it fails. The dismiss button, and the settings screen. Platform limits ship with defaults that make them advisory — no passcode, "Block at End of Limit" switched off, deletion and reinstallation available, the clock adjustable. The full list is in why your screen time limits stopped working. Hardened properly, this mechanism is genuinely decent. Unhardened, which is how nearly everyone runs it, it is a notification.

3. The reward

What it does. Gives you something for not using the phone — a growing tree, a streak, a score, a small forest of sessions completed.

Who it suits. People motivated by accumulation, and people using it for bounded focus sessions rather than all-day control. For a 25-minute work block it is legitimately effective.

How it fails. The stake is symbolic, and symbolic stakes are re-priced under pressure. A 40-day streak feels valuable right up to the moment you want something more, at which point it costs you a number. Streaks also have a distinctive failure: once broken, the entire structure evaporates, and the day after a broken streak is reliably the worst day.

4. The physical object

What it does. Moves enforcement off the phone. A tag you must tap to unlock apps, a lockbox with a timer, a second stripped-down device.

Who it suits. People who have already tried software and watched themselves defeat it. This is the most genuinely rigid mechanism available to a consumer, and it is the honest answer for the hardest cases.

How it fails. Logistics rather than psychology. You have to carry it, charge it, and not leave it at the office. The failure mode is not dramatic — you simply stop using it, over a few weeks, without ever deciding to. It is also the most expensive category by a distance.

5. The stake

What it does. Attaches a real cost to going past your own rule. Money, forfeited or spent, in an amount you agreed to in advance.

Who it suits. People whose problem is not awareness and not scheduling but negotiation — who reliably talk themselves past every barrier that only costs time. The economics research on commitment devices is summarised in our piece on them; the short version is that it works, modestly, for the minority who choose it.

How it fails. Three ways, and they are worth naming plainly. If the stake is too small it becomes a toll you pay without noticing. If paying resets the constraint, you have bought a subscription to breaking your own rule. And there is an obvious conflict of interest in any company that collects when you fail — which is the objection we get most, and which we answer by making blocking free forever, publishing every price in advance, and tracking the share of users who spend nothing as a success metric.

The property that predicts failure

Read the five together and one variable sorts them: does the cost of defeating this fall as your motivation to defeat it rises?

For interruptions, blocks and rewards, yes. All three charge you in seconds, taps, or symbolic value, and all three are cheapest at 1am — the exact moment they are tested. That is not a flaw in any particular product. It is a property of the mechanism.

For physical objects and financial stakes, no. Metres do not shrink at 1am. Neither does a price. That is the entire reason both work better than their user-interface polish would suggest, and the reason they are the two most annoying options to live with.

An honest table

What we are, and where we lose

Last Ferry is mechanism five with mechanism two underneath. Your chosen apps share one daily budget — one between them, so closing Instagram and opening TikTok buys nothing. When it runs out they lock, with a real screen rather than a notification. Going past the lock costs money from a balance you topped up in advance: $0.50 for the first unlock of the day, $1.00 for the second, $2.50 after that, resetting at midnight and never resetting because you paid.

Where we lose:

• Browsers are not covered. We block apps. The mobile web versions work fine, and for some people that alone is disqualifying.
• You can uninstall it, in about a minute, like anything else. No consumer app can honestly claim otherwise — see the app blocker you cannot delete does not exist.
• It costs money to fail, and some people find that objectionable in principle. It is a reasonable objection and we would rather you decide it now than after installing.
• Android alpha only at the time of writing, with iOS in development and no store listing yet.

If your problem is the unconscious open, use an interruption app; it will serve you better than we will. If you respect a rule once it exists, harden the platform limits you already have and spend nothing. Come to a stake when the other four have already been tried and defeated — that is the situation it is designed for, and it is not a better product for anyone else.