TL;DR. Screen time is a crude metric, so do not start by cutting. Start by opening your phone's own weekly report and writing down the daily average. Then set a target of roughly that number minus twenty percent, phrased as a daily yes or no you can tick. Then climb the friction ladder from the cheapest rung upward: notifications off by default, greyscale, home screen stripped to four apps, app limits, phone-free zones and times, charging outside the bedroom, and only then deleting anything. Most of the value sits in the first three rungs, and most of the theatre sits in the ones people try first.
Two hours of reading and two hours of doomscrolling produce exactly the same number in your weekly report.
Worth saying first, because almost every article on this topic quotes an average and then implies the number itself is the problem. It is not. Screen time is a proxy, and a fairly bad one. My own total goes up in weeks where I read more, because the book is on the phone.
So this is not about getting a number down for its own sake. It is about the thing people actually want: to stop losing evenings they did not decide to spend.
Step 1: Measure before you cut
You cannot aim at a target you have not read.
On iPhone: Settings, then Screen Time. On Android: Settings, then Digital Wellbeing and parental controls.
Look at the weekly view, never today. A single day tells you nothing except what kind of day it was.
If you want the number in context rather than in isolation, sketching where the whole 24 hours goes puts the screen figure next to sleep, work and everything else. Four hours reads very differently once you can see what it displaced.
Write down three numbers.
- Daily average. This is the one you will track.
- Your top three apps. This tells you which rung to climb first.
- Pickups per day, called unlocks on Android. The number people ignore, and the most diagnostic of the three.
The pickup count separates two different problems. Ninety pickups of two minutes each is a compulsion problem: your hand goes to the phone in every gap. Four sessions of ninety minutes is a substitution problem: the phone has replaced something that used to fill your evenings. Same total, opposite fixes. The first needs the cue removed, the second needs a replacement.
Expect the number to be higher than you guessed. Everyone underestimates it, including me. You now have a baseline, which is the whole purpose. If you want to see where the rest of the day goes, the time audit and the guide to running one do the same job across a week rather than one device.
Step 2: Turn it into a number you can tick
"Use my phone less" is not a target. It has no failure condition, so it cannot be kept or broken, so nothing changes.
Take your weekly average and cut about twenty percent. Write it as a ceiling.
| Your average | Target |
|---|---|
| 3h 00m | Under 2h 30m |
| 4h 30m | Under 3h 30m |
| 5h 40m | Under 4h 30m |
| 7h 00m | Under 5h 30m |
Twenty percent, not half. A fifty percent cut is a sentence, and it will hold for four days.
Then phrase it as a daily binary: "Phone under 4h 30m", ticked yes or no each evening from the report. Not a number you log, not a slider. A binary is the only version you will still be doing in three weeks.
One warning, from the person who built the tracker: do not attach a streak to this. Screen time spikes for good reasons, a delayed flight, a sick child, a weekend of navigating. A counter that resets to zero for a legitimate Tuesday teaches you the metric is stupid, which it half is anyway. That is the case I made in contribution grid versus streaks, and it applies harder here than almost anywhere. Count the days you kept, not the days in a row.
Step 3: The friction ladder
Every rung below adds friction between you and the app. They are ordered cheapest to most drastic. Climb one at a time, give each one a week, and stop as soon as you hit your number, because the higher rungs cost real convenience.
Rung 1: Notifications off by default. Not "manage" them. Off. Then re-enable one at a time, only where a delay would genuinely cause a problem: calls, messages from actual humans, your calendar. Everything else, email included, is something you check when you decide to. Highest-value rung on the ladder, and almost nobody does it properly because they go app by app instead of setting a default.
Rung 2: Greyscale. Costs nothing, works for a while. iPhone: Accessibility, Display and Text Size, Colour Filters. Bind it to the accessibility shortcut so it is a triple-click. Colour is doing more work in your app-opening decisions than you think. You will habituate, and that is fine: it is a rung, not a cure.
Rung 3: Strip the home screen. Four apps, chosen for utility rather than pull. Everything else into the app library or the second page. No browser on the dock. The point is not hiding, it is that the apps you open by muscle memory stop being where your muscle memory reaches. This and rung 1 do most of the work.
Rung 4: App limits and downtime. Both platforms have them, with one large asterisk: the "ignore limit" button is one tap, and within two weeks you will tap it without reading it. Set it anyway, treat it as a speed bump. Downtime, the scheduled block where most apps simply do not open, is the stronger and less used half.
Rung 5: Phone-free zones and times. This is where behaviour actually changes, because it removes the cue instead of asking you to resist it. Pick two, no more:
| Zone or time | What it fixes |
|---|---|
| The dinner table | The interrupted conversation |
| The bedroom | The 1am hour and the first ten minutes of the day |
| The first hour after waking | The phone framing your whole morning |
| The toilet | The unglamorous truth about the pickup count |
Rung 6: Charge it outside the bedroom. On its own rung because for most people it is worth more than the five above it combined. Buy a cheap alarm clock, the phone lives in the hall. If you only ever do one, do this. It is the rung I kept from my own detox weekend.
Rung 7: Log out and remove the app. Not delete. Logged out and off the home screen is a real wall, and re-entering a password makes the decision conscious.
Rung 8: Delete it. Only if you keep reinstalling from rung 7. Use the website instead, which is worse on purpose and therefore self-limiting. Deleting the account is a different and much larger decision.
Which rungs are theatre
Screen time goals with no replacement. A target and nothing else is a number that makes you feel bad in the evening. The gap the phone was filling is still there.
Focus modes with twenty rules. The setup is satisfying and the maintenance is not. People build an elaborate schedule of modes, use it for nine days, then stop trusting it the first time it silences something important.
Buying a dumbphone in week one. Occasionally the right answer. Usually a purchase that substitutes for the change, and usually resold within two months.
Checking the dashboard daily. Checking your screen time is screen time. Weekly is enough.
Why "just use it less" fails
Because it is an instruction to a person who is not present at the moment of the decision.
Nobody decides to spend forty minutes in a feed. You unlock the phone to check one thing while tired, bored or waiting, and the app you open next is chosen by whatever your thumb reaches first. There is no decision in there to apply willpower to.
Every rung above changes the conditions instead of the intent: fewer notifications means fewer unlocks, a stripped home screen means the thumb lands on nothing, the phone in the hallway means the 1am hour never starts. The advice that keeps failing is the advice aimed at your intentions. Your intentions are fine. It is the conditions that are against you.
Making the number feel like something
Numbers in hours per day are hard to feel. The screen time calculator converts yours into units you actually have intuitions about.
You enter phone hours and minutes, computer hours and minutes separately, roughly what share is social, plus your age, a life expectancy and your waking hours. It returns hours per week, full days per year, and the share of your waking hours the total represents, banded from low through moderate and high to very high. There is a fixed scenario for cutting one hour a day, which is 365 hours a year or about fifteen full waking days, and comparisons converting the time into books at six hours each, films at two, miles walked, meals cooked, workouts, podcast episodes, and long ones like a language at 600 hours.
Three caveats. The lifetime totals assume your usage stays exactly flat for the rest of your life, which it will not. The reduction scenario is fixed at one hour and not adjustable. The comparison rates are flat constants, so treat "you could read 900 books" as a shape rather than a fact. Set the computer field to zero if most of your screen time is paid work, otherwise you are measuring your job. Everything stays in your browser.
What to expect
Weeks one and two are worse, not better. The reaching continues after the app is gone, and the gaps the phone was filling come back before you have anything to put in them. That discomfort is the cost, paid up front.
By week three the pickup count is usually down more than the total time, which is the better signal of the two.
The number will still spike sometimes. Across the habits people set up in Loggd, about 45% are never checked off even once and roughly two in three die at or before the second check-in, which is the subject of habits people quit fastest. The ones that survive are the ones where a bad Tuesday was allowed to be a bad Tuesday.
If you want the harder reset first, a 48-hour detox weekend shows you which rungs you actually need, and the digital detox tracker will run the week after it in 7, 14 or 30 day blocks.
Measure. Set one number. Climb one rung a week. Stop when the evenings come back.
Last updated: November 2026.
Written by Eusebiu, the solo founder building Loggd in public. I build it in the evenings around contract work and a small daughter, so my own screen time report is not something I would like published. Rungs 1, 3 and 6 are the ones I actually kept. I share what the aggregate data keeps showing on Threads.
Track the ceiling, not the streak. Loggd records "phone under 4h 30m" as a daily yes or no on a forgiving contribution grid, so a bad travel week stays a lighter square instead of a reset. Start free.