Average session length is the mean duration of a single user session — calculated by dividing total session time by the number of sessions over a period.
For product teams, it measures the depth of engagement per visit.
For UA teams, it helps evaluate whether acquired users are genuinely engaging or just opening the app and leaving. This glossary entry covers how to calculate average session length, what affects it, and how to use it alongside other metrics to diagnose engagement problems.
What Is Average Session Length?
A session is a single continuous period of app usage — from the moment a user opens the app to the moment they close it or it goes to background. Average session length is the mean duration across all sessions in a given period.
It's a behavioral metric, not a business metric. It doesn't directly measure revenue or retention — but it correlates with both. Users who spend more time per session tend to engage more deeply with content, progress further, and convert to paying at higher rates.
The key word is "tend." A dictionary app with 45-second average sessions isn't broken. A mid-core RPG with 45-second sessions probably is.
Learn more about average session length →
How to Calculate Average Session Length
Formula:
| Average Session Length = Total Session Time / Number of Sessions |
|---|
Example: Users collectively spend 1,000 minutes across 200 sessions. Average session length = 5 minutes.
Note: this is average per session, not per user. A user with three 5-minute sessions contributes three sessions to the denominator — not one. That's different from Total Daily Time Spent, which measures average time per user per day.
Outliers matter. A single unusually long session — a user who left the app open overnight — can skew the average significantly. If your average session length seems implausibly high, check the distribution and consider using the median session length alongside the mean.
What's a Normal Average Session Length?
Session length varies significantly by app type and genre. There's no universal benchmark — but directional ranges help set expectations.
| App / Game type | Typical session length |
|---|---|
| Hyper-casual games | 2–5 min |
| Casual games (match-3, puzzle) | 5–10 min |
| Mid-core games | 10–20 min |
| RPG / Strategy | 20–40 min |
| Utility / productivity apps | 1–3 min |
| Social apps | 5–15 min |
| Streaming / media | 20–60 min |
These are directional ranges based on market data. Your numbers will vary by genre, audience, and product design.
What matters more than the absolute number is the trend — and how session length compares across user segments.
What Affects Average Session Length?
Content depth. More content — more levels, longer narrative arcs, deeper progression systems — gives users more reasons to stay. A session that ends because a user ran out of things to do is a design problem.
Session design. Some products are designed for short, frequent sessions (casual games, news apps). Others are designed for long, immersive sessions (RPGs, streaming). Session length should reflect deliberate design, not just happen.
Onboarding quality. New users often have shorter sessions than retained users — they're still learning. A sudden drop in average session length after a UA campaign can mean the new cohort isn't engaging as deeply as existing users.
Technical issues. Crashes, loading times, and bugs shorten sessions. A sharp drop in average session length is often one of the first signals of a technical regression — it appears before DAU or retention metrics move.
How to Use Average Session Length
Track trends, not absolutes. A single session length number tells you little. What tells you something is a consistent decline over two weeks, a spike after a content update, or a gap between new and retained users.
Segment by user type. Average session length for new users vs. Day 30+ users often differs significantly. If retained users have much longer sessions than new users, the product rewards time investment — which is good. If new users have longer sessions than retained ones, they may be getting stuck rather than engaging.
Correlate with retention. Users with longer sessions typically retain better — but the relationship isn't linear. Segment users by session length bucket and compare their D7 retention. If users with 10+ minute sessions retain at 2× the rate of users with under 2 minutes, session depth is a leading indicator of retention — and you should optimize for it.
Use as an early warning signal. Session length reacts to product changes faster than retention or revenue metrics. If average session length drops the day after a release, investigate before the retention impact compounds.
Average Session Length vs Total Daily Time Spent
These two metrics are related but measure different things:
| Average Session Length | Total Daily Time Spent | |
|---|---|---|
| Formula | Total session time / Sessions | Total session time / DAU |
| Measures | Depth per visit | Total daily engagement per user |
| Rises when | Sessions get longer | Users have more sessions OR longer sessions |
| Use case | Session design, content depth | Overall engagement level |
A product can have short average sessions but high total daily time spent — if users open it many times per day. That's typical of messaging apps and casual games with energy mechanics.
Takeaways
- Average Session Length = Total Session Time / Number of Sessions. It measures depth per visit, not per user.
- There's no universal benchmark — what matters is whether session length matches your product's design intent and how it trends over time.
- Outliers skew the average. Consider the median session length alongside the mean for a cleaner picture.
- A sudden drop in session length is often the first signal of a technical issue — it appears before DAU or retention metrics move.
- Segment session length by user type: new vs. retained users often show very different patterns.
- Correlate with retention: users with longer sessions typically retain better. Use it as a leading indicator.
See your average session length by segment, cohort, and acquisition channel. Explore devtodev →