Mobile playtime per DAU is down: Session length is what moved
min read
Mobile daily playtime is down from two years ago. The decline isn’t about how often players open games. It’s about how long they stay per visit.
Median daily playtime in a mobile game is lower today than it was two years ago. This is a finding our own data supports, and multiple independent sources confirm at the aggregate level. Looking inside the numbers, the component that moved isn’t the one that would most obviously explain the drop.
Our own Benchmarks Hub puts the median mobile game at 13 minutes of daily playtime per DAU in October 2026. Our 2024 benchmark report (technically 2025, but looking at data from 2024) puts the same median at around 22 minutes. Methodology has evolved between the two, so the two figures aren’t a perfect like-for-like, and the gap should, therefore, be interpreted as directional rather than exact. But the shape is clear: Mobile playtime per DAU has shifted down.
External sources observe the same direction
Independent sources tracking mobile engagement have reported directionally similar findings over the same period. Adjust’s 2026 Mobile App Trends report observes a decline in global daily time spent in mobile games between 2024 and 2025. Adjust data cited by eMarketer shows a 2% year-over-year decline in North American daily time spent in mobile games in 2025, with install volume down 5% over the same period. eMarketer’s own forecast places mobile gaming time spent at a plateau, with further decline not expected.
Each source is measuring a slightly different cut of the mobile ecosystem, and the absolute numbers aren’t directly comparable with ours. What matters for this piece is that the trend direction is the same across every source that has looked at the question. Mobile players are spending less daily time in games than they were two years ago.
Sessions per DAU has held steady
Playtime per DAU is a composite of two things: how often players open the game in a day (session count per DAU) and how long each sitting lasts (session length). Either could have moved, but it was not session count. The current Hub shows the median at 4.51 sessions per day. Our 2024 report put the median at 4. The same directional caveat applies here as it does to the playtime comparison, but what the data shows clearly is that players are opening games about as often as they did two years ago, and slightly more often if the comparison is taken at face value.
Session length is the component that moved
Our 2024 data showed the median session running between 5 and 6 minutes, but the current Hub shows it at 3.14 minutes. Players haven’t walked away from mobile games as they’re coming back about as often as before. What changed is how long they are staying per visit.
Third-party coverage places this number consistently. Aggregators citing our 2025 report show the median mobile session at 3.1 to 3.5 minutes, which is in line with what the Hub shows for 2026. The 3-minute median session was already visible in 2025 data, so this component of the shift happened earlier than the surface numbers might suggest.
Session length is the one metric on the Hub that’s moving directionally within 2026, and the movement is upward. The median rose from 2.97 minutes in March to 3.14 minutes in October. Playtime per DAU is holding roughly flat over the same window. The new baseline looks stable, and there’s a small upward drift worth watching into 2027.
What this means for planning
The shift has three practical implications.
- For UA, the ad impression opportunity per DAU has held roughly flat, since impressions scale with session count more than with session length, and session count is where the data is most stable.
- For LiveOps and game design, the planning constraint has shifted. A 3-minute median session is a different design problem than a 5-minute one. Depth-per-moment matters more than extending time-on-task, because extending time-on-task is running against the structural pattern.
- For 2027 planning, the question is whether the small upward drift in session length within 2026 is noise or the start of a slow reversal. Treating the 3-minute baseline as structural is the safer assumption. Treating any further decline as unlikely, given the floor the data now shows, is also reasonable.
Want to see where your own game sits inside this shift? The current percentile data for playtime per DAU, sessions per DAU, and session length is live on the Benchmarks Hub, refreshed monthly, with a copy-paste prompt you can drop into any AI tool to compare your game directly.
