← Blog

What Can a Screen Time App See? 20 Privacy Labels Checked

A Screen Time app on your iPhone sees less than you might fear. Apple’s API hands it sealed tokens instead of your app choices, and it keeps usage charts in a sandbox with no internet access. What the app collects beyond that, the developer declares on its App Store privacy label. When we read 20 of those labels on 4 October 2026, 11 declared data used to track you. Ours was one of them.

I build Sparky, so this post grades my own app too, and Sparky does not come out on top.

What can a Screen Time app see through Apple’s API?

Much less than most people assume. When you choose apps to block, Apple’s picker lets you do it “without revealing their choices to the app.” The app gets back “opaque values” it can pass to Apple’s blocking tools but cannot read. Usage charts come from a separate extension that Apple locks in a sandbox with no network access.

Apple’s FamilyActivitySelection page says the selection “holds opaque values that represent categories, applications, and web domains selected by the user.” In practice, an app that blocks TikTok for you holds a token for TikTok, not the word TikTok. The FamilyActivityPicker is where the “without revealing their choices” line comes from.

The DeviceActivityReport page is blunter about charts. The report extension “runs in a sandbox,” and “this sandbox prevents your extension from making network requests or moving sensitive content outside the extension’s address space.” Sparky’s usage charts in version 1.6 are drawn this way, so by Apple’s design the main app never receives the numbers behind them.

An app does learn when a usage limit it set is reached, because iOS has to tell it for the block to work. The signal carries the name the app gave the limit, not the apps behind it. That same signal is what misfires on iOS 26, which we covered in why app blocker limits fire instantly.

Can any Screen Time app read your real app usage?

One kind can, in one region. Since iOS 26.4, Apple has an authorization status called approvedWithDataAccess. It lets an app fetch “the actual bundle identifiers of installed applications, domain names of visited websites” instead of sealed tokens. Only devices in the EU, signed in with an EU Apple Account, can grant it.

Apple’s approvedWithDataAccess page adds two limits. “Only one app at a time can hold this authorization status on a given device,” and “on devices outside the EU, authorizationStatus never returns approvedWithDataAccess.” Sparky version 1.6 never asks for it. Outside the EU, Apple’s Screen Time API gives every app tokens only.

What does “Data Used to Track You” actually mean?

It means ad tracking, in Apple’s specific sense. Apple’s definition covers linking data from an app about you or your device “with Third-Party Data for targeted advertising or advertising measurement purposes,” or sharing that data with a data broker. Linking your install to an ad you saw elsewhere counts. The definition never uses the word sell.

Here is Apple’s full sentence, from its App Privacy Details page for developers:

“Tracking” refers to linking data collected from your app about a particular end-user or device, such as a user ID, device ID, or profile, with Third-Party Data for targeted advertising or advertising measurement purposes, or sharing data collected from your app about a particular end-user or device with a data broker.

One of Apple’s examples shows that an SDK alone can put tracking on a label: “Placing a third-party SDK in your app that combines user data from your app with user data from other developers’ apps to target advertising or measure advertising efficiency, even if you don’t use the SDK for these purposes.”

Every label is self-reported. All 20 listings we read carry the same line from Apple: “This information has not been verified by Apple.” Apple tells developers, “You’re responsible for keeping your responses accurate and up to date,” and that they “may update your answers at any time” without shipping an app update.

Which Screen Time apps declare data used to track you?

Eleven of the 20 listings we read on 4 October 2026: Achieve, BePresent, Blocus, Brainrot, Clearspace, Jomo, LockedIn, Opal, Sparky, Unglue and Unrot. The other nine declare none. Three of those nine, DoomIt, PushBlock and Steppin, say Data Not Collected. Only Sparky, Clearspace and BePresent list third-party advertising as a purpose.

AppData used to track youWeak spot on the label
Apple Screen Time (built into iOS)No App Store labelNo label to compare, because it is part of iOS
Sparky (ours)Purchases, Identifiers, Usage DataTies for the most tracked categories. Lists Device ID for third-party advertising
UngluePurchases, Identifiers, Usage DataTies for the most tracked categories. Links 4 categories to you
ClearspaceContact Info, Usage DataLinks 7 categories to you, Health among them. Lists third-party advertising
LockedInPurchases, Usage DataTracks purchases and usage
AchieveUsage DataTracks usage data
BePresentIdentifiersLists Device ID for third-party advertising
BlocusPurchasesTracks purchases
BrainrotIdentifiersTracks identifiers
JomoUser ContentLinks 6 categories to you
OpalIdentifiersLinks 5 categories to you
UnrotIdentifiersTracks identifiers. Name is linked to you
AtosumaNoneLinks 7 categories to you
BrickNoneLinks 5 categories to you
Habit DoomNoneUsage data for analytics, not linked
one secNoneUsage data for analytics, not linked
PiggyNoneLinks photos or videos to you
ScreenZenNoneUsage and performance data for analytics, not linked
DoomItData Not CollectedUnverified by Apple, like every label
PushBlockData Not CollectedUnverified by Apple, like every label
SteppinData Not CollectedUnverified by Apple, like every label

How we checked: we opened each app’s US App Store page on 4 October 2026 and copied its App Privacy section word for word. “Links” means the label’s Data Linked to You section. Apple’s built-in Screen Time has no label to read, though Apple’s iCloud security overview lists Screen Time data stored in iCloud as end-to-end encrypted.

Tracking and linking are separate questions, and the table shows it. Atosuma, Brick and Piggy track nothing, yet link 3 to 7 data categories to your identity for purposes such as analytics and app features. A label with no tracking can still hold a lot about you.

Why does Sparky’s own label declare tracking?

Because Sparky measures its ads. In version 1.6, Sparky sends funnel events to Atlas, an attribution service I run, and Atlas uses them to learn which ads on Apple Ads and Meta brought people in. Matching installs to Meta ads is ad measurement, which Apple counts as tracking. Nothing goes to Atlas until you answer Apple’s tracking prompt.

Of the funnel events Sparky records, only four are allowed to reach an ad network: onboarding completed, first gate created, paywall viewed, and the paywall button tapped. That list lives in PaidFunnelAnalytics.swift, and the same file blocks free text by design. Its header says a habit name, a gated app name, a survey note, or a raw error string “cannot reach an ad network by accident.” Atlas also logs app opens and sessions by itself, and in a July 2026 test before release, Meta accepted one of those session events. Purchase data reaches Atlas from RevenueCat, the subscription service, server to server. The app sends no revenue events itself.

Sparky asks the tracking question during onboarding, after a screen titled “How Sparky measures ads.” That screen says: “Say no, and nothing changes. Your gates, habits, and earned time work the same.” Atlas is set to wait for your answer before it starts any network collection (AtlasIntegration.swift).

There is a second lock. Sparky 1.6’s privacy manifest lists Atlas’s web domain as a tracking domain, and Apple’s rule for those domains is strict: “If the user has not granted tracking permission through the App Tracking Transparency framework, network requests to these domains fail.” So a “No” blocks Sparky’s reports to Atlas at the iOS level. Our 1.6 release notes record the cost: funnel analytics stop for people who decline.

The one item our label links to your identity is a device ID, listed for third-party advertising and app functionality. That is the honest shape of it. Apple’s definition is not about selling. Sparky does measure ads, and that is what the label says.

Do photo-proof apps see more than blockers?

They can, because a photo is data the app creates itself, outside Apple’s sealed tokens. Habit Doom checks habit photos on the phone and says “Photos never leave your iPhone.” Its label links nothing to you. Piggy, a chore app where kids “snap proof,” links Photos or Videos to your identity for app functionality.

One wrinkle is worth knowing. Habit Doom’s 6 June 2026 post says “The App Store privacy label stays at zero collected data.” On 4 October 2026 its label listed one item: Product Interaction usage data, not linked to you, used for analytics. Labels move. Read the current one, not a blog’s summary of it, ours included.

How do you check a Screen Time app’s label yourself?

Open the app’s App Store page and scroll to App Privacy. Read Data Used to Track You first, then Data Linked to You, then the purposes listed under each. Keep Apple’s line at the top of every label in mind: “This information has not been verified by Apple.” A label is the developer’s own account of itself.

Our label is the bad news in this post. Sparky ties Unglue for the most tracked categories and is one of three apps that list third-party advertising. If a clean label is your first filter, Sparky is the wrong pick. DoomIt, PushBlock and Steppin declare Data Not Collected, and one sec and ScreenZen list only analytics data that is not linked to you.

If the earning model matters more to you than the label, our category guide compares the apps that make you earn screen time. The Opal alternatives list covers the blockers, and the Habit Doom comparison goes deeper on photo proof.


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