Time on page vs. session duration?

Session duration measures the whole visit from entry to the last interaction, while time on page is one page's slice of that visit. Summing the page times of a session usually does not match its duration, because the exit page contributes nothing.

What is a good time on page?

No reference number exists, because the page's purpose decides: a short support answer should be read fast and a low number there is success, while on a long article the same number means the content went unread. The right baseline is a page against its own history and against comparable pages.

Time on Page

Beginner

Updated Jul 11, 2026 Shahin editorial 3 min read

PronunciationEN/taɪm ɑːn peɪdʒ/

Time on page is how long a user stays on a page, classically derived from the gap between that page loading and the next tracked hit in the same session.

That mechanism creates the main limitation: with no following hit there is nothing to subtract, so the last page of every session and every single-page session is zeroed or dropped, biasing the average upward. Newer implementations fix this by measuring foreground time directly, through periodic heartbeats or the pagehide and visibilitychange events, and GA4 reports engagement time instead of time on page, which is why two tools can give very different numbers for the same page. In the classic method an abandoned background tab still accrues time even though nobody read anything. Interpretation is not one-directional either: a high number can mean interest, or it can mean a user who cannot find the answer.

How it is calculated

Time on page = timestamp of the next hit − timestamp of the page load

The calculation only works for views that have a following hit in the same session; exit pages fall out of the average and the unit is seconds.

FAQ

Sources:MDN — Document: visibilitychange event