Average Session Duration
Average session duration measures mean time per session — a directional engagement signal that's easy to misread. Here's how it's calculated, what's normal by vertical, and when to trust it.
Average Session Duration
The mean time visitors spend on your site per session, calculated as total session time divided by total sessions.
Average session duration is one of the oldest engagement metrics in web analytics: total time across all sessions divided by the number of sessions in the period. In GA4 it's reported as `Average session duration` and is derived from the time between the first and last event in a session.
It looks simple but reads badly in isolation. A long session can mean a shopper is comparing variants in detail — or that they're lost in your size guide and about to bounce. Treat it as a directional signal you slice by segment, not as a goal in itself.
The metric sits in the engagement family alongside pages per session, scroll depth, and engaged sessions. None of them are revenue metrics — they're proxies you use to diagnose why conversion rate moved, not numbers you optimise directly.
Two structural quirks bite first-time users. Bounced sessions in legacy Universal Analytics counted as zero seconds, which dragged the average down. GA4 changed the model: it measures elapsed time between the first and last event, so single-event sessions still register near zero unless an engagement event fires.
Average Session Duration = Total Session Time / Total Sessions
Total Session Time
Sum of all session durations
Sum, in seconds, of elapsed time between first and last event of every session in the period.
Total Sessions
Session count
Total sessions started in the same period, including single-event ones.
A Shopify apparel store reviews last week's traffic in GA4.
Total session time: 1,820,000 seconds (≈ 505.5 hours)
Total sessions: 42,000
→ 43.3 seconds average session duration
That's on the low end for apparel. Before chasing it, the team segments by source: paid social sessions average 18 seconds (mostly bounces), while email sessions average 2 minutes 10 seconds. The 'fix' isn't a site change — it's tightening the paid social audience.
Benchmarks below are realistic ranges for online stores on Shopify, WooCommerce, or Magento with GA4 set up correctly. Verticals with high consideration (electronics, home) run longer than impulse categories (beauty consumables, accessories). Mobile sessions are typically 30-40% shorter than desktop.
Typical average session duration ranges by vertical and device
| Vertical | Desktop | Mobile | What's a red flag |
|---|---|---|---|
| Apparel & accessories | 2:10 – 3:30 | 1:20 – 2:20 | Below 0:45 site-wide |
| Beauty & personal care | 1:40 – 2:50 | 1:00 – 1:50 | Below 0:40 site-wide |
| Home & furniture | 3:00 – 5:00 | 2:00 – 3:20 | Below 1:15 site-wide |
| Consumer electronics | 3:30 – 5:30 | 2:10 – 3:30 | Below 1:30 site-wide |
| Food & beverage (DTC) | 1:20 – 2:20 | 0:50 – 1:30 | Below 0:35 site-wide |
The fastest way to make the metric useful is to stop reading the site-wide number. Segment by traffic source, landing page, and device, and compare converting vs non-converting sessions in the same segment. That comparison — not the absolute value — is where the diagnosis lives.
Frequently asked questions
GA4 sums the elapsed time between the first and last event in each session, then divides by the total number of sessions. Sessions with only one event still count in the denominator, which keeps the average lower than many people expect.
Three common causes: a high share of paid social or display traffic that bounces fast, single-page landers where no second event fires, or tracking that doesn't emit engagement events. Segment by source and landing page before assuming the site is the problem.
No. Long sessions can mean genuine interest or genuine confusion — a shopper stuck on shipping or size guide pages. Pair the metric with conversion rate and exit pages in the same segment to know which one you're looking at.
Most online stores fall between 1:30 and 3:30 site-wide. Higher-consideration verticals like furniture and electronics run longer; impulse categories like beauty consumables run shorter. Compare against your own historical baseline before chasing an industry number.
Average engagement time (GA4) only counts time when the tab is in focus and the user is actively engaged. Average session duration counts elapsed time between events, including idle periods. Engagement time is the stricter, more behavioural number.
Yes — in GA4, a non-engaged single-event session has a duration near zero, which pulls the average down. If you've recently improved engagement event tracking, expect duration to rise even though user behaviour hasn't actually changed.
Not as an absolute number. Set targets at the segment level — for example, 'paid social mobile landing pages above 45 seconds' — and tie them to a downstream metric like add-to-cart rate so you're not optimising a vanity number.
Mobile sessions are genuinely shorter — smaller screens, faster scanning, more interruptions — and mobile traffic skews more to paid social, which has higher bounce rates. A 30-40% gap between mobile and desktop is normal.
Improve content relevance on entry pages, fix slow product pages, and surface internal links to related products and reviews. But don't optimise duration in isolation — pair every change with a conversion-rate check, because longer sessions on confused users hurt revenue.
It's a diagnostic layer between traffic metrics (sessions, source/medium) and outcome metrics (conversion rate, AOV, revenue per session). Use it to explain movement in those outcome metrics, not as a primary KPI.
Get an AI expert review of your site
Paste your URL — Metricuno's AI runs the same heuristic checks a senior CRO consultant would, scoring your page and prioritising the fixes that'll move conversion fastest.