← Blog

What Counts as Screen Time on iPhone? How iOS Measures It

iPhone Screen Time counts an app only while that app is frontmost on the screen. Apple’s developer documentation says so in those words. The day’s total is a second number: how long the screen stayed on. Pickups are a third, split into pickups that opened an app and pickups that did not. Many “Screen Time is wrong” puzzles come from mixing those three up.

I build Sparky, the app where you earn screen time by finishing habits. Sparky reads these numbers through Apple’s API, so I can show you, from Sparky’s own code, exactly where an app’s count and the Settings app part ways. Some of those gaps are Sparky’s fault.

What does iOS count as screen time?

iOS counts foreground time. Apple’s developer page for DeviceActivityEvent defines device activity as “the amount of time an application, category, or web domain is frontmost on the screen.” An app that plays audio behind another app is not frontmost. A website counts when Safari, or a browser that reports to Apple, shows it.

The same Apple page adds two details people rarely see. Web time “includes domains visited in Safari or any third-party browser that contributes web usage via a STWebpageController,” so a browser that skips Apple’s reporting hook leaves its sites out. And activity “accumulates based on the time zone of the scheduled start date.” If you fly across time zones, an app’s daily limit can keep counting on the clock of the place where its schedule started.

The total is measured differently. Apple’s page for a report segment’s total activity duration says the value “represents the total amount of time the device’s screen remained on.” That is why your daily total can run higher than the sum of your apps: time on the Lock Screen and the Home Screen is screen-on time with no app in front.

Does Screen Time count when the screen is off?

By Apple’s written definitions, no. App time needs the app “frontmost on the screen,” and the daily total is the time “the device’s screen remained on.” A podcast playing in your pocket with the screen dark meets neither definition. Both definitions come from Apple’s developer documentation, not from the Settings app.

That last point is the honest limit of this answer. Apple’s iPhone User Guide tells you where the report lives (Settings, then Screen Time, then Daily Average) but never defines the words in it. The developer definitions are the only written ones Apple publishes. I have not found an Apple page that says the Settings report uses different rules, and I have not measured the Settings report with the screen dark.

What counts as a pickup?

A pickup is a separate count, not minutes. Apple’s API splits it in two: “the number of pickups made directly to the application,” counted per app, and “the number of device pickups without application use,” such as checking the time on the Lock Screen. A full pickup count needs both halves.

Sparky 1.6 only adds up the first half. Its Today card sums each app’s numberOfPickups and never reads totalPickupsWithoutApplicationActivity. I checked the 1.6 report extension, UsageReportExtension.swift, while writing this. So every glance at the clock that opened no app is missing from Sparky’s pickup number, and Sparky will show fewer pickups than a full count.

For scale, pickups and minutes measure very different things. A dscout study of 94 Android users over five days in 2016 logged 2,617 touches a day across 76 separate phone sessions on average. That sample is small and not iPhone, so treat it as a sense of size, not a benchmark.

Why does Screen Time show an app I didn’t open?

The usual reason is another device. Apple’s guide says that with Share Across Devices on, “You can also see your combined Screen Time activity across devices.” An hour of YouTube on your iPad can then appear in a report you opened on your iPhone. Turn the setting off, or pick one device in the report, to separate them.

A second cause only affects third-party apps, and Sparky’s code documents it. When a Screen Time app asks for a report on both an app and its category, iOS can return that app twice in the same stretch of data. A comment in Sparky’s report extension says those rows “describe the same segment usage,” so Sparky keeps the larger of the two instead of adding them. An app that adds them would show double the real time for that app.

Why don’t a Screen Time app’s numbers match Settings?

Because each app chooses what to ask iOS for. An app picks the devices, the date range and the apps in its report, then does its own math on the results. Sparky 1.6 asks for iPhone data only, from midnight, and leaves apps used for under a minute off its top-apps lists.

Here is how the counts line up, from Apple’s guide and Sparky 1.6’s code:

Who is countingDevices in the numberWhere the count startsPickupsWeak spot
Apple Screen Time (Settings)This iPhone, or every device on your Apple Account with Share Across Devices onThe day or week you pick in the reportEvery pickupThe combined view can fold iPad and Mac time into what looks like an iPhone number
Sparky (ours)iPhone data only (devices: .init([.iPhone]))Midnight, for charts and “Used today”Only pickups that opened an appUndercounts pickups, rounds down to whole minutes, and needs iOS 26.0 or later
A usage limit in any Screen Time appThe device it runs onWhen the app starts or restarts watching, unless it turns on includesPastActivityNoneRestarting the watch resets the count to zero

Two smaller rules in Sparky’s code also move numbers. The weekly chart fills in all seven days, because a comment there notes that iOS “omits days with no activity,” and an average over active days only would quietly ignore your zero days. And the reports show “completed minutes only,” so 4 minutes 59 seconds of TikTok reads as 4 minutes.

Why does a limit’s count restart in the middle of the day?

Because the app restarted its watch. A Screen Time app arms a limit by asking iOS to monitor some apps. Sparky’s AppLimitsService.swift records that stopping and restarting that monitor “resets DeviceActivity’s accumulated-usage counter for the current window to zero.” Edit a limit at 3 p.m. and its count starts again at 3 p.m.

The result inside Sparky is two numbers on one screen that disagree. In App Limits, a Pro feature in Sparky 1.6, the “Used today” card in the limit editor counts from midnight. The limit itself counts from your last edit. After 45 minutes of Instagram and an edit, the card says 45 minutes while the limit grants a fresh budget. Sparky’s code calls this “a platform limitation.” And because Sparky keeps every limit on one monitor, a change to any limit restarts the count of all of them.

Apple gives apps one way to count backwards. The includesPastActivity switch decides “whether the system takes into account the person’s device activity before your app starts monitoring the event.” Apple’s example: an app that starts at 1:30 p.m. on a 1 p.m. schedule can include the 1:00 to 1:30 usage. Sparky never turns it on. In a device test on 11 June 2026, we confirmed that it counts the whole day’s use of the chosen apps.

There is also a known iOS 26 bug where a limit fires seconds after you set it, with almost no usage behind it. That is a fault in iOS, not a counting rule, and we covered the evidence in why app blocker limits fire instantly on iOS 26.

What limits does Apple put on how apps measure usage?

Apple caps how apps can watch usage. A monitoring window must last at least fifteen minutes and at most one week. An app and its extensions can monitor at most twenty activities at once. Limits finer than one minute are not reliable.

That last rule is not on Apple’s pages. It comes from a comment in Sparky’s EnergySnapshot.swift: “DeviceActivity silently coalesces sub-minute thresholds, so anything finer than this is aspiration, not measurement.” In practice a limit is measured in whole minutes. Sparky’s limit editor starts at 1 minute and stores every limit as one event on a single daily monitor, which keeps it far under the cap of twenty.

Can a Screen Time app export your usage numbers?

No, and that is by Apple’s design. Usage charts come from a report extension that, in Apple’s words on the DeviceActivityReport page, “runs in a sandbox.” The sandbox “prevents your extension from making network requests or moving sensitive content outside the extension’s address space.” The app can draw your numbers but cannot read them.

That applies to Sparky too. Sparky’s History screen draws your daily average, screen time and pickups inside that sandbox, and the main app never receives the values. So Sparky cannot email you a spreadsheet of your week, and the same sandbox binds every Screen Time app. Our post on what Screen Time apps can see covers the rest of Apple’s privacy model. It also explains why Sparky’s own privacy label declares tracking: Sparky measures its ads through an attribution service after Apple’s tracking prompt, which is separate from these usage numbers.

Which number should you trust?

Trust Settings for the whole picture and the app for what it enforces. Settings covers every pickup and, if you choose, every device. A Screen Time app’s number is the one its limits act on, so it explains a block. If you want an accurate pickup count or one total across iPad and Mac, Sparky is the wrong tool.

The numbers matter because people act on them. In Pew Research Center’s 2024 survey, 38% of US teens said they spend too much time on their smartphone. That kind of judgment rests on a number, and the number depends on who counted it and from when.

If you want a count that ends in a decision rather than a chart, our guide to apps that make you earn screen time compares the apps that turn usage into something you have to earn first.


Sparky turns small habits into screen time you earn. Download it free on the App Store.