How a session is defined
A session is one continuous stretch of active use. It typically starts when the app comes to the foreground and ends when the app goes to the background, or after the user has been idle for longer than a timeout.
The timeout decides how sessions are counted. If a user checks a message, switches to another app for 20 seconds, then returns, is that one session or two? With a 30-second background timeout it is one; with a 10-second timeout it is two. Pick a timeout that matches how your app is used and keep it fixed, or session trends will shift for reasons that have nothing to do with users.
Common session metrics
- Sessions per DAU: how many times an active user opens the app on a typical day.
- Average session length: total session time divided by session count.
- Time in app per user: sessions per user multiplied by average length.
- Sessions to conversion: how many sessions it takes before a first purchase or subscription.
A worked example: on one day, 20,000 active users produce 56,000 sessions totalling 3,920,000 seconds. Sessions per DAU is 56,000 / 20,000 = 2.8. Average session length is 3,920,000 / 56,000 = 70 seconds. Time in app per user is 2.8 x 70 = 196 seconds, a little over three minutes.
What sessions tell you
Session metrics are engagement signals. A rising number of sessions per user usually means the app is becoming a habit. A falling average session length can mean users find what they need faster, or that something on the first screen is pushing them away; the other metrics around it tell you which.
Sessions also feed other metrics. DAU is often defined as users with at least one session that day, and retention is often measured as a user starting a session on day N after install.
Sessions and attribution
Attribution links sessions back to where users came from. An MMP like Linkrunner attributes each install to its source and records the in-app events you send, then reports retention by campaign (classic, rolling and cumulative). That tells you whether a campaign's users keep coming back, which is what session counts measure at the user level. For deeper session analytics, many teams pair their MMP with a product analytics tool; Linkrunner connects to Amplitude, Mixpanel, PostHog and others so attribution data travels with the user.