The short answer
Every self-imposed screen time limit has the same structural flaw: the person who set the limit and the person tapping "ignore" are the same person, with different amounts of willpower. Apple's built-in Screen Time is the clearest case, because the passcode protecting the limit is a passcode you chose and therefore know.
So the useful way to sort this category is not by features. It is by who holds the override: you, a piece of friction, or another person. Each tier survives a different kind of failure, and picking the wrong tier is why the last three apps you tried are still installed and still ignored.
Key Takeaways
- The category splits three ways by who can lift a limit: you, friction, or another person.
- Limits you hold yourself work against the absent-minded reach and fail against genuine wanting. That is not a character flaw, it is what "self-imposed" means.
- Friction is the honest answer for most people. It costs nobody anything, and if it holds for two weeks you can stop reading.
- The two-minute test for any app: hit the limit on purpose, then count the seconds, the taps, and the people who have to agree.
- If nothing has to change hands when you push back, you do not have a limit. You have a reminder.
- Ignore the claim that accountability lifts success rates to 95%. The strongest experimental evidence nearby is modest: 3 percentage points on a six-month smoking outcome (Giné, Karlan & Zinman, 2010).
This page covers one mechanism applied to one problem. For the wider map, see the four mechanisms of accountability.
Why self-set limits fail at the exact moment they matter
A self-imposed limit is a message from the version of you with slack on a Tuesday afternoon to the version of you with none at 11:40pm, and the second one holds the key. Nothing about the limit was wrong. The enforcement was handed to the person the limit was aimed at.
Watch the shape of it. You cap Instagram at 45 minutes while sitting at a desk, feeling reasonable. The cap fires on Friday night, when you are tired, slightly bored, and have already decided what you are going to do. The block screen offers a way past. You take it. Two weeks later it is furniture you walk through without reading.
Apple's Screen Time makes the problem visible because it is honest about it. App limits, downtime and the passcode are all set by you, in Settings, at a moment when getting past them is not on your mind. When it is on your mind, Settings is four taps away. Screen Time is genuinely useful as a measuring instrument and a speed bump, not as a wall.
The uninstall problem sits underneath every third-party app too. Whatever a blocker does to your phone, it does with permissions you granted, and it stops when the app is gone. Some products make deletion harder than others. None can make it impossible.
The test of a limit is not whether it stops you on a normal Tuesday. It is whether it stops you on the night you actually want through. If it has never been tested that way, you do not yet know whether it works.
You do not need a statistic to decide whether this is your problem, either. Published averages for daily screen time disagree by wide margins depending on source, year and population, and none of them describe your week. Open Settings, look at last week's average, and notice whether the number surprised you.
Three places the override can live
Ask one question of any tool: when the limit fires and you want through, what has to happen? There are only three answers, and they define the tiers.
You hold it
A limit you can lift with a passcode you chose, in an app you can delete. This is Apple's Screen Time and, by default, most blockers on the App Store. It stops one real thing: the reach you never decided to make, where your thumb finds the app while the kettle boils. Interrupting that is worth doing, because a lot of phone use is not a decision at all. It cannot stop a decision, and no configuration changes that.
Friction holds it
A delay before the app opens, a pause screen, an extra step, a physical object you have to fetch. Nobody is stopping you. The tool is buying time, and time is the thing an impulse cannot survive. An urge is not steady pressure that grinds you down until you give in; it rises, peaks and passes. A few seconds of enforced waiting is often enough for it to expire, and for you to notice you did not want it much. For most people this tier is the honest answer.
Someone else holds it
A limit only another person can lift, or a miss with a consequence you cannot quietly reverse. This is the only tier that survives you actively wanting through, because what stands between you and the app is no longer your own resolve. The price is real: it spends someone else's attention, it means asking for something, and it can turn an ordinary evening into a negotiation. Reach for it once the cheaper tiers have failed you, with evidence.
| Who holds the override | What it looks like | What it stops | What it does not stop | What it costs |
|---|---|---|---|---|
| You | A limit behind a passcode you set, in an app you can delete | The unconscious reach | A decision. If you want through, you go through | Nothing, which is also the problem |
| Friction | A delay, a pause screen, an extra step, an object to fetch | Most impulses, because the urge expires before the delay does | A determined hour. You can wait it out | A few seconds, every time, forever |
| Another person | A limit only they can lift, or a miss you cannot reverse alone | Wanting through, because wanting is no longer sufficient | Deleting the arrangement, though somebody notices | Their attention, and having to ask |
Friction first, and often that is enough
Before you recruit anybody, run pure friction for two weeks, because for a lot of people it settles the problem and costs nobody a favour. The cheapest versions need no app at all.
Delete the app and use the browser version, which is slower and worse on purpose. Log out, so opening it needs a password you have to go and find. Take it off the home screen, so reaching it requires search and therefore a decision. Charge the phone in another room overnight, which removes the two worst hours of the day. Each of these adds a gap between the reach and the feed, and the gap is where the impulse dies.
Set a checkpoint at two weeks and be honest at it. If usage on the app you care about dropped and stayed down, you are done. Adding social accountability then would be paying a real cost for a solved problem, and it teaches the people around you that your requests do not mean much. Friction fails in one specific place: the evening you have a reason. A bad day, an argument, a deadline you are avoiding. If your two weeks collapsed on nights like that rather than on ordinary ones, that is the signal to move the override out of your own hands. Those evenings run on a familiar script, and we have written about why "I'll do it tomorrow" keeps winning that argument.
The tools, and what we could not check
These are the products people mean when they search for this, and we could not verify the third-party ones first-hand, so we are not going to tell you what they do. Their features change monthly and a stale table is worse than no table. The column that decides your choice is one you can fill in yourself.
| Tool | Who holds the override | How we know |
|---|---|---|
| Apple Screen Time built into iOS |
You. Limits, downtime and the Screen Time passcode are set by you, so lifting them is a trip to Settings. The exception is the family configuration, where a parent holds the passcode for a child's device | Described only at the level of who holds the passcode. We did not re-fetch Apple's documentation on the day of writing, and these menus move between iOS versions |
| Opal | Not verified | Not checked first-hand |
| one sec | Not verified | Not checked first-hand |
| Brick | Not verified | Not checked first-hand |
| ScreenZen | Not verified | Not checked first-hand |
| Freedom | Not verified | Not checked first-hand |
| Jomo | Not verified | Not checked first-hand |
| Roots | Not verified | Not checked first-hand |
| Commit (ours) | Mixed. A miss can block the apps you chose; extra minutes are earned with exercise, and the commitment itself is visible to the people you shared it with | Read-only audit of our own iOS source, August 2026. iPhone only, public beta |
Provenance. "Not verified" means we did not confirm that row ourselves and will not repeat another roundup's claim about a competitor. It does not mean a tool is weak or a feature is missing. Several of these are well regarded, and at least one is a physical device rather than an app, which changes the answer entirely. Check the current listing before paying for anything.
The two-minute test that places any of them
Run it on the app itself rather than reading about it, because the answer lives in the block screen and nowhere else.
- Set a limit for something trivial and hit it on purpose. One minute on a game you do not care about. See the block screen while you are calm, not at midnight while you are not.
- Count the seconds and the taps. Under five seconds and entirely in your own hands means you bought a speed bump. Useful, but know what you bought.
- Ask who has to agree. If the honest answer is nobody, the limit is yours to lift, and it will be lifted on the night it matters.
- Try to uninstall it. Whatever happens next is your real floor, and every tool has one. A product that admits this is more trustworthy than one implying it built a wall.
Where Commit fits
Commit is ours, and its screen time features sit in the third tier: the limit is attached to a commitment other people can see, and some routes past it need somebody else to say yes. Everything in this section is iOS-only, because it is built on Apple's Screen Time frameworks.
Screen Time as the proof, not just the limit
Commit runs on evidence rather than self-reported ticks, and iOS Screen Time is one of the nine proof types, alongside photo, video, screenshot, file, GPS check-in, live time-lapse, Apple Health and none. So "no Instagram after 9pm" becomes a commitment whose evidence is your own usage data, verified like any other proof. The reading comes from iOS, not from you remembering how the evening went. Small reframe, large consequence: the limit becomes something you pass or fail in front of people, rather than a setting that quietly stopped working weeks ago.
A miss can block the apps you chose
Most trackers respond to a missed commitment by drawing a grey square. Commit lets the consequence land on the thing that caused it. Pick a screen time block as your stake, and missing the commitment blocks the apps you nominated, enforced by iOS itself. There is an optional daily soft cap if a full block is more than you want to sign for.
Exercise to unlock
You can earn screen time back with pushups, jumping jacks or squats. This is the feature I would keep if we had to delete the rest, because it does something smarter than saying no: it changes the price. You are not forbidden from opening the app, you just pay for it in a currency your evening self is reluctant to spend. It is self-limiting in a way a passcode never is. Nobody does two sets of pushups to check a feed twice, and if you want the minutes badly enough to earn them, you have made a decision rather than obeyed a reflex.
Who reviews your proof
Proof goes to friends you nominate or to Commit AI, so the structure does not depend on one person staying attentive. Share the commitment with a group, up to 10 people, a hard cap, and every consecutive day everyone shows up moves one group streak. A rejection is not final either: you can appeal, appeals are judged at a deliberately lower bar, and a successful appeal restores your streak. The longer argument for choosing your reviewer well is in how to find an accountability partner, and the group version in habit trackers with friends.
You also start with 10 grace tokens, and spending one within 48 hours of a miss marks the day skipped and restores your streak, because the most common way people quit is missing once and abandoning everything. Money stakes exist too, from $1 to $500 and charged only when you miss, but screen time is the more interesting currency on this page.
The honest ceiling, and it applies to us as much as to anything else here: you can still delete Commit. No app can make itself unremovable. What changes is that deleting it is not silent, because the people you shared the commitment with are still there tomorrow.
The limitations, plainly. Every screen time feature above needs iOS, so there is no Android equivalent and there will not be a close one. Commit is iPhone only, portrait only, no iPad. It is in public beta through TestFlight and is not on the App Store, so there is no download link we could give you today that would work, and it is free while in beta. Screen time enforcement is one of the newest parts of the app and is still under test, so treat it as beta software rather than a finished lock. If you need something working on an Android phone this evening, one of the other tools here is your answer.
Commit is in public beta
iPhone, free while in beta. We will email you when a spot opens. There is no App Store listing yet and we are not going to pretend otherwise.
Join the betaIf you would rather understand the underlying idea first, start with what a commitment device is.
A setup that survives week two
These five decisions matter more than which app you pick, and all of them are free to make before you install anything.
- Name the app, not the phone. One or two apps account for most of it, and you already know which. A blanket limit on the whole device fails because you genuinely need your phone, so the limit becomes something you route around all day until you switch it off.
- Pick a boundary, not a budget. "Nothing after 10pm" is checkable and has one decision point. "45 minutes a day" invites a negotiation every hour about whether this bit counts, and you will win every one of those negotiations.
- Run friction alone for two weeks, and write today's number down now, so the review is against a figure you recorded rather than a memory you will edit.
- Only if that failed, move the override. Name one person, and hand over the smallest thing that works. A veto over your entire phone is a request they will regret accepting.
- Agree the exception in advance. There will be a night with a real reason. Decide now how many are allowed and how you flag one, because these arrangements die in the argument about the first unplanned exception, not in the exception itself.
Two weeks is also the moment to ask whether the goal is yours. People are remarkably disciplined about things they actually want, so if this keeps collapsing it is worth checking whether you want less phone time or want to be the sort of person who wants less phone time. Only one of those responds to a blocker.
Common questions
How do I stop doomscrolling at night?
Attack the location before the software. Charging the phone outside the bedroom removes the two worst hours from the day and needs no app, no subscription and no willpower after the first week. If it has to stay in the room, a time boundary beats a daily budget, because at 11pm a budget is a question and a boundary is an answer. If both of those have already failed for you, that is the specific signal to move the override to another person rather than to buy a stricter app.
Is there an app that lets a friend control my screen time?
That is the right thing to search for, and the feature to look for is not "share my stats" but "somebody else has to approve the change". Plenty of apps show a friend your usage, which produces mild embarrassment and no override. Our version is that a missed commitment blocks the apps you nominated, and extra minutes are earned with exercise: pushups, jumping jacks or squats. We did not verify whether the other tools above offer anything similar, so check before you decide.
Does Apple's Screen Time actually work?
It works well as a measuring instrument and a speed bump, and both are worth having. It cannot work as a wall, because you set the passcode. The one built-in configuration that moves the override is the family setup, where a parent holds the passcode for a child's device. Two adults can improvise a version of that by one holding the other's passcode, which some couples do successfully, though it is blunt: no rules attached, and no way for either of you to see whether the arrangement is being respected.
Can't I just uninstall whatever I install?
Yes, and any product implying otherwise is overselling. The realistic goal is not to make escape impossible but to make it slow, deliberate and visible. Slow defeats impulses. Deliberate turns an unconscious reach into a decision you own. Visible is what stops the decision being free, and visibility is the only one of the three you cannot arrange by yourself.
Do screen time limits actually reduce phone use?
The evidence here is thinner and noisier than the marketing suggests, and we will not invent a number to fill the gap. The most rigorous nearby experiment is about commitment contracts rather than phones: Giné, Karlan and Zinman (2010) found 11% take-up of a commitment savings product for smokers, and those offered it were 3 percentage points more likely to pass a six-month nicotine test, with the effect persisting in surprise tests at twelve months. Modest and durable is the honest summary, and any page promising a 95% success rate is citing a study nobody can produce.
What if I do not want anyone watching my phone use?
Stay in the friction tier, and treat that as a legitimate choice rather than a failure of nerve. Social accountability suits people motivated by not letting someone down, and corrodes people who mostly feel surveilled. If sharing usage with a friend sounds worse than the doomscrolling, it probably is worse for you, and a delay screen you never have to explain will do more good than a system you resent.
Sources
- Giné, X., Karlan, D. & Zinman, J. (2010). Put Your Money Where Your Butt Is: A Commitment Contract for Smoking Cessation. American Economic Journal: Applied Economics, 2(4), 213-35. aeaweb.org
- Commit feature claims, including proof types, screen time stakes, exercise unlocks, verification and grace: read-only audit of the Commit iOS application source, August 2026.
What we did not verify. The current feature sets, platforms and pricing of Opal, one sec, Brick, ScreenZen, Freedom, Jomo and Roots. We describe Apple's Screen Time only at the level of who holds the passcode, and did not re-fetch Apple's documentation on the day of writing, so confirm current behaviour in Settings. No screen time statistic appears on this page, because the published averages disagree by wide margins depending on source, year and population, and a number you cannot trace is worse than none. If you work on one of these products and want an accurate row, send details through our contact page or to [email protected] and we will verify and publish them.