Two apps can report your usage honestly and still disagree by half an hour. Apple builds its Screen Time report from the raw usage log iOS keeps on your device. No third-party app can read that log: iOS tells a blocker only that a usage threshold has been crossed, never who crossed it and never the running total. Every third-party figure is therefore a reconstruction. Add background audio, merged devices and a reporting lag, and both numbers end up right about different questions.
Why can’t a screen time app just read Apple’s number?
Because Apple has deliberately made it impossible, and the reason is privacy.
When you choose which apps to limit, the picker you tap is drawn by iOS itself; the app around it draws the frame and nothing else. What comes back is a set of opaque tokens — meaningless handles that let the app say “shield these” without ever learning that one of them is Instagram. No call turns a token into a name, an icon, or a per-app minute count.
That is an architectural property of Apple’s FamilyControls and DeviceActivity frameworks, not a policy an app could quietly break. It also means a blocker can never reproduce Apple’s chart, because it has never seen Apple’s data. It does not measure you. It asks iOS to tell it something.
What does threshold crossing actually mean?
DeviceActivity works like a doorbell, not a stopwatch. An app registers an interval — usually your day — and one or more events, each saying in effect: wake me when the selected apps have accumulated this much usage inside this interval. iOS watches, and at some point calls back: threshold crossed.
That callback is the entire signal. No stream of minutes, no per-app split, no timestamp for when you opened anything.
So a third-party app’s resolution is the size of its smallest threshold. When your blocker shows 23 minutes, what it can genuinely justify is closer to you have passed one marker, not yet reached the next, and this much wall-clock time has elapsed since. Useful for running a budget. Not a measurement, and no engineering turns it into one.
Why doesn’t the count reset when I close the app?
Because iOS counts cumulatively per interval, not per sitting.
Close the app, leave the phone on the table for an hour, open it again — the counter carries on from where it stopped. In this API there is no such thing as a sitting. There is a bucket that fills until the interval ends, then empties.
This is why careful apps describe a per-session cap as a block of your budget rather than this sitting. You consume a session-sized chunk of a cumulative total, and when the chunk is gone the shield goes up for a while. The behaviour is correct; the vocabulary everyone reaches for — session, sitting, this time — is what does not fit the machinery underneath.
Does background audio count toward the total?
Very often, and it explains most of the totals people find unbelievable. A podcast or a music app playing with the screen off and the phone in a pocket can log against that app for the whole journey. Two hours of driving becomes two hours of Entertainment. Apple’s own report includes this, so it is not something a third-party app got wrong — it is a shared property of how usage is counted on the device.
Before concluding you spent four hours scrolling, open the category breakdown in Settings and subtract the stretches when you were driving, walking or asleep. Years of your life on your phone goes further into how much of a headline total is genuinely attention.
Why does the total jump when I haven’t touched my phone?
Share Across Devices. In Settings > Screen Time there is a switch that merges every device signed into the same Apple Account into a single report. An hour of Safari on the iPad lands in the same total as your phone.
Meanwhile a blocker installed on your iPhone is reacting to the phone in your hand. Apple can be counting three devices while the app counts one, and the gap is exactly the time you spent on the iPad. This one toggle produces more confused support emails than anything else in Screen Time — there is more on it in App and Website Activity.
Why does today’s number keep changing?
Because none of these figures are live. Apple’s own numbers update through the day and settle overnight, which is why this morning’s running total compared against last week’s finished total always looks wrong. DeviceActivity callbacks arrive when iOS decides to wake the extension, not on a metronome — usually promptly, sometimes later if the phone is in Low Power Mode, cold, or busy with something the system considers more urgent. Widgets then refresh on their own system-managed budget on top of that.
Check a widget, the app, and Settings inside the same minute and you can get three different numbers. By tomorrow they will have converged.
Which number should you trust for which question?
| Question | Number to use | Why |
|---|---|---|
| How much am I really on my phone? | Apple’s weekly average, in Settings | It reads the raw usage log |
| Which app is taking the time? | Apple’s per-app breakdown | Blockers only see opaque tokens |
| Has this gone down since last month? | Apple’s weekly average, month against month | Trends survive the noise a single day does not |
| How much budget is left before things lock? | The blocker’s counter | It owns the threshold that triggers the shield |
| Did today count toward my streak? | The blocker | Apple does not track your rules |
| Was that exactly 23 minutes? | Neither | Nothing on iOS answers that question |
The rule underneath the table: use Apple’s report to know, and a blocker to decide. One is a scoreboard you read afterwards; the other is a governor that acts while you are still playing.
“My blocker said 26 minutes and Settings said 58, and I genuinely thought I had been sold something broken. Then I remembered the podcast running on the drive in, and that my iPad shares the same report. Both numbers were fine. They were answering different questions and I had only ever asked one.” — Tomás, logistics coordinator, 38
So is the blocker’s number useless?
No, and people get this backwards. The one thing a blocker is not estimating is enforcement: when the threshold is crossed, iOS applies the shield, and that event is exact and system-level. What is soft is the pretty figure in the middle — the running total, the widget count, the minutes remaining. Those are a fuel gauge, and a gauge a few percent off still gets you to the petrol station. Nobody argues with the moment the engine stops.
Which tells you what is worth being annoyed about. A chart that disagrees with Settings is expected. A shield that fails to appear after you blow through the budget is a real fault.
It also tells you how much precision you actually need, which is less than you want. The gap between 3h40 and 4h10 decides nothing; direction over weeks and what the time displaced are the real question in how much screen time is too much. Chasing an exact figure is usually a comfortable substitute for the harder one about what you would rather be doing — which is where most people who ignore their limits get stuck.
How does Unscrol help?
The honest part first. Apple’s Screen Time is free, already on your iPhone, and the most accurate picture of your usage that exists anywhere. If your question is where the time goes, open Settings and stop reading. App Limits and Downtime are free too, and for many people a Downtime window is the entire answer.
Unscrol is for the case where knowing has not changed anything. It is an iPhone and Apple Watch app built on the same Screen Time frameworks, so blocks are applied by iOS rather than drawn as an overlay you can swipe away. The core is two numbers you choose: a daily total (45 minutes by default) and a per-session maximum (5 minutes by default). Use a session-sized chunk of the budget and the chosen apps lock for 15 minutes. Spend the whole daily budget and they lock until tomorrow. A streak counts every day you stay under.
Everything above is why the app talks the way it does. Usage comes from crossed thresholds and iOS counts cumulatively per interval, so Unscrol says a block of your budget rather than pretending to time your sitting to the second, and its 28 widgets show the remaining daily budget on the system’s refresh schedule rather than live. The privacy position is architectural rather than a promise: iOS hands over opaque tokens, so the app can never see which apps you picked, their names, their icons, or your per-app usage, and none of it reaches Unscrol’s servers.
Around the two numbers sit a Shield you can schedule by time of day — switching it off mid-session is deliberately slow, a breathing exercise plus a wait rather than a one-tap dismiss — a one-minute Mind Break reflection, and around 117 challenges. Free to download, tracking one challenge at a time; premium adds Streak Saver and five slots. The app never tells you off about a number. It puts a wall up and hands the decision back.
Frequently asked questions
Why doesn't my screen time app show the same number as Apple's Screen Time?
Because they are measuring different things. Apple builds its report from the raw usage log iOS keeps on your device, with full access to every app and every minute. A third-party app never sees that log. iOS hands it opaque tokens and tells it only when a usage threshold has been crossed, for example that the chosen apps have now accumulated five minutes inside the current interval. The app reconstructs a total from those crossings. Add background audio, devices merged by Share Across Devices, and a reporting lag, and two figures can differ by tens of minutes while both are correct.
How accurate is a third-party screen time app on iPhone?
It is exact about the thing it controls and approximate about everything it displays. Apple's FamilyControls and DeviceActivity frameworks deliberately withhold app identities and per-second totals from third-party apps for privacy reasons, so a blocker knows when a threshold has been crossed but never holds a running count. That is enough to enforce a limit reliably, because the shield is applied by iOS itself at the moment of crossing. It is not enough to publish a precise usage report. Read a blocker's minutes as a fuel gauge, and Apple's Settings report as the measurement.
Does background audio count toward screen time on iPhone?
Often yes, and it is the most common reason a total looks impossible. A podcast or music app playing with the screen off and the phone in your pocket can log against that app for an entire commute, which is why many people's Entertainment category is a car journey rather than a binge. Apple's own report includes this, so it is not a third-party bug. Before reacting to a large number, open the category breakdown in Settings and mentally subtract the stretches when you were driving, walking, cooking or asleep.
Which screen time number should I actually trust?
Use Apple's Settings report to know, and a blocker's counter to decide. If the question is how much time you spend and where it goes, Apple's weekly average and per-app breakdown are the best data on the device, and comparing this month's weekly average against last month's is far more meaningful than any single day. If the question is how much budget remains before your apps lock, the blocker's figure is the one that governs. No number on iOS, Apple's included, should be read to the exact minute.