Android 17 ships a 50/50 split-screen gaming mode built specifically for foldables, letting a user run a game on one half of the unfolded display while using a second app — chat, a guide, a companion tool — on the other half, alongside a new built-in virtual gamepad overlay for compatible games. It’s a small platform feature with a real instrumentation problem attached: almost every mobile analytics setup assumes one app, one screen, one foreground session at a time. Split-screen mode breaks that assumption outright, and foldables are exactly the device class where it’s now a first-class, Google-promoted interaction pattern rather than an edge case.
The immediate risk is double-counted or misattributed sessions. If your SDK’s foreground/background detection treats “app is visible in half the screen” the same as “app has full user attention,” a session that’s actually a secondary, glanced-at companion app gets counted identically to a primary, fully-engaged one. Engagement metrics built on session duration or screen-time will quietly inflate for foldable users running split-screen sessions, making foldable engagement look stronger than it is — right up until someone tries to explain why foldable retention numbers don’t match qualitative feedback.
Data Points to Track
- Split-screen session rate — how often your app is opened or used while another app is simultaneously visible on a foldable’s unfolded display, versus full-screen single-app sessions
- Screen-share position and size — which half (or what proportion) of the display your app occupies during a split-screen session, since a small companion pane implies different attention than an even 50/50 split
- Foreground-but-secondary engagement, distinguishing genuine interaction (taps, scrolls, input) from a session that’s simply visible but not being actively used
- Gamepad overlay usage, for any game or interactive feature, tracked as a distinct input method from touch, since it changes how session and interaction events should be interpreted
- Fold-state transitions mid-session — how often a user folds, unfolds, or enters/exits split-screen while your app is running, and what happens to engagement immediately after
Setup Steps
- Detect split-screen and multi-window state via the platform APIs available on Android 17 and tag every session event with that state at capture time.
- Separate “visible” from “actively engaged” in your session model — require an interaction signal (tap, scroll, input) within a defined window before counting a split-screen session as engaged rather than merely displayed.
- Instrument gamepad-overlay sessions distinctly from touch-based sessions, since input method affects both engagement patterns and what a “successful” session looks like.
- Log fold-state and window-size transitions as explicit events, so a sudden drop or spike in session metrics can be traced to a device-state change rather than treated as unexplained noise.
- Re-baseline foldable engagement dashboards separately from standard phone dashboards until split-screen behaviour is well understood, rather than blending the two device categories into one engagement number.
Actionable Insights
A high split-screen session rate with low active-engagement signal is a strong sign your foldable engagement numbers are currently overstated by companion-app usage — worth correcting before it feeds into a retention or engagement report that gets compared against phone-only cohorts. A genuinely high split-screen engagement rate, backed by real interaction events, is a signal worth investing in deliberately: it means users are treating your app as a companion tool during another primary activity, which is a different product opportunity than being the primary app itself.
Gamepad-overlay adoption is worth watching as an early indicator of how much of your foldable gaming audience is using the native controls versus touch — a fast-growing share there usually means it’s worth optimising input handling specifically for the overlay rather than treating it as a rare edge case.
Related Resources
Need help tracking this in your app?
Our team sets up analytics pipelines for mobile and web teams every day. Talk to us and get your first events flowing in under an hour.
Talk to an expert