Digital friction is a deliberate cost placed between the impulse to open an app and the app opening. A wait, a breathing exercise, a question you have to answer, a habit you have to finish. The app still opens. That is what separates friction from blocking.
Most screen time apps use friction of some kind, but they do not all use the same kind, and the differences decide whether the tool still works in week three. This page sorts them into four types, says when each one earns its place, and gives you a way to check whether the one you picked is doing anything.
The four kinds of friction
| What sets the cost | Resets | Example | |
|---|---|---|---|
| Fixed | You, once. Same cost every open | Never | one sec: a breathing exercise before the app opens |
| Escalating | The number of opens so far today | At midnight | ScreenZen: a delay you can make grow with each open of the day |
| Conditional | A task you complete in exchange | Per unlock | Jomo: finish a habit and the app opens. Roots: perform a ritual |
| Adaptive | Your own recent behavior, measured | Recalculated nightly | Munen: the pause and the access window follow your zone |
Competitor details here come from each app's public App Store listing, website and help center, checked on 17 September 2026.
Fixed friction: the same cost every time
You set a wait or a prompt and it does not change. one sec is the clearest case: a breathing exercise stands between you and the app, every time, and it is the same exercise at 9 am and at midnight.
Fixed friction is the best studied kind. The one sec mechanic was tested in a 2023 PNAS paper and the interruption itself, not its length, did the work. Munen caps every pause at 10 seconds for that reason, which we go through in why the pause is never longer than ten seconds.
Its strength is legibility. You always know what opening Instagram costs, which makes the cost easy to plan around and easy to accept. Its weakness is the same thing: a cost that is right for a calm Tuesday is too light for the night you are actually in trouble, and you cannot raise it without raising it for the calm Tuesday too.
Escalating friction: the cost grows with the day
The wait gets longer with each open, then resets. ScreenZen does this: the delay before an app opens can be configured to grow every time you open it during the day.
This fixes the worst case of fixed friction, the tenth open costing exactly what the first one did. What it does not do is carry anything across days. Midnight wipes the slate, so a week of bad nights starts every morning identical to a week of good ones.
Conditional friction: pay in another currency
Instead of waiting, you do something. Jomo will open YouTube for 15 minutes when you finish a habit you defined, or unlock Netflix at 5,000 steps. Roots asks you to complete a ritual, breathe or pet a virtual dog, before it lets you through.
This is the most interesting category and the least predictable. When the task and the urge are compatible it is excellent, because you get something out of the impulse rather than just surviving it. When they are not, and at 11 pm they usually are not, the task becomes an obstacle you resent rather than a trade you accept.
Adaptive friction: yesterday sets today's cost
The cost is not chosen by you in advance or by a counter that resets. It is derived from how you actually used the apps, and it changes between days.
This is what Munen does. A zone between 1 and 5 is recomputed every night from five signals of your flagged apps: night use at 32 percent of the weight, time in the apps at 26, opens at 16, intensive sessions at 14, and how you handled the pause itself at 12. The zone sets both numbers you feel. On the balanced intensity the pause runs 10, 8, 5, 3 and 0 seconds from the hardest zone to the easiest, and the access window after you get in runs 3, 4, 6, 8 and 10 minutes. A rough night makes tomorrow slower. A good day makes it lighter. The zone moves at most one step per day, and never mid-day.
The claim adaptive friction makes is narrow and worth stating plainly: the day you most need the cost to be high is a day you cannot be trusted to set it yourself, because you set your rules on a calm afternoon and you break them at midnight. Reading yesterday takes that decision off you.
What you give up is legibility. You cannot know today's pause without looking, which is exactly why the weights above are published rather than hidden in a model. If an app is going to decide how hard it is to open Instagram, you should be able to read the arithmetic. We print the whole table in what a screen time app actually measures about you.
Blocking is not friction
A blocker removes the option. Opal sessions, Roots' Monk Mode and Jomo's Strict Mode all do this, and Strict Mode goes further by stopping you from deleting the rule or uninstalling the app.
That is a different tool with a different failure mode. People delete blockers, usually around week two, on the day they genuinely needed in and the wall did not care. The mechanism has a name, reactance, and we cover it in why people delete app blockers after two weeks.
Friction and blocking are not rivals. Munen uses both: friction for ordinary opens, focus sessions when you want an app gone outright. The distinction that matters is who holds the override, and for how long.
How to tell if friction is working for you
Most people judge a screen time app by whether their daily total went down. That single number hides the two ways friction fails.
Check opens against minutes, not minutes alone. If your opens dropped and your minutes did not, the friction is working at the door and failing inside: you go in less often and stay much longer. A friction that only guards the entrance can produce this. What it needs is a limit on the way out, which is what an access window is for.
Check night use separately. It is the signal with the strongest evidence behind it and the one a daily total buries, because 25 minutes at midnight and 25 minutes at lunch look identical in a weekly average. If the evening did not move, nothing important moved.
Count how often you override. Every friction has a way through. If you take it on most opens, the cost is too low or aimed at the wrong moment, and the number itself is the finding.
Give it three weeks. Week one flatters every tool, because novelty is doing the work. What you want to know is whether the thing survives the first night you really want in.
All four are readable from Apple's own Screen Time report if your app does not surface them.
Common questions
Is adaptive friction just a smarter blocker? No. A blocker removes the option; adaptive friction changes what the option costs. The app still opens in every zone, including the hardest one.
Does adaptive mean machine learning? Not necessarily, and not in Munen's case. The zone is a weighted sum with five published weights and fixed thresholds. No model, no training.
Can I combine two friction apps? Technically yes, since they use the same Screen Time framework, but two costs stack on the same open and the result is a wait neither app designed. Pick one mechanic per app.
Which kind is best? For a predictable cost you can reason about, fixed. For an impulse that is worth redirecting into something, conditional. For the 11 pm open you did not decide to make, adaptive, because it is the only one that knows last night happened.
The short version
Fixed friction charges the same every time. Escalating friction charges more as the day goes on and forgets at midnight. Conditional friction charges you in tasks. Adaptive friction charges based on how you have actually been using the app, and is the only one that carries a bad night into the next day.
See how this plays out against specific apps: Munen vs one sec, Munen vs ScreenZen, Munen vs Opal, Munen vs Roots and Munen vs Jomo.
Sources
- Gruening, D. J., Riedel, F., and Lorenz-Spreen, P. "Directing smartphone use through the self-nudge app one sec." Proceedings of the National Academy of Sciences, 2023.
- "InteractOut: Leveraging Interaction Proxies as Input Manipulation Strategies for Reducing Smartphone Overuse." CHI Conference on Human Factors in Computing Systems, 2024.
- Hunt, M., Marx, R., Lipson, C., and Young, J. "No More FOMO: Limiting Social Media Decreases Loneliness and Depression." Journal of Social and Clinical Psychology, 2018.
- Brehm, J. W. A Theory of Psychological Reactance, 1966.
- App Store listings and help centers for one sec, ScreenZen, Opal, Roots and Jomo, checked 17 September 2026.
- Munen figures read directly from its source code on 17 September 2026.