Time on page measures the seconds between one pageview and the next signal from that visitor. On a 3-page visit, only 2 pages produce a measurable figure, because the last page sends no follow-up signal, so most trackers record that page as zero seconds and every published average is understated.
How Time on Page Is Calculated
Time on page is the difference between two timestamps: the moment a page loaded, and the moment the tracker next hears from that browser. Reading itself is never measured. A visitor who reads for 8 minutes and then closes the tab leaves the same trace as one who left after 2 seconds.
Every analytics tool therefore works from arrivals rather than attention, and the five engagement metrics below get used interchangeably when each one is blind to something different.
| Engagement metric | How the metric is calculated | What the metric cannot see |
|---|---|---|
| Time on page | Next signal’s timestamp minus the pageview’s timestamp | The last page of a visit |
| Average session duration | Last timestamp minus first timestamp in a visit | Single-page visits, which score zero |
| Engagement time | Seconds the tab is focused and in the foreground | Reading in a background or blurred tab |
| Scroll depth | Furthest point of the page a reader reached | Whether the reader stopped to read |
| Exit rate | Share of views of a page that ended the visit | How long the reader stayed first |
Why the Last Page of a Visit Records Zero
A pageview-based tracker learns how long a page stayed open only when the visitor does something else. On the final page of a visit there is no something else: the browser closes, the tab is discarded, or the visitor walks away, and that page’s duration never reaches the server.
The gap follows from what a pageview tracker can observe, so no tool escapes the gap entirely. An exit ping, a small beacon fired as the page is hidden or unloaded, raises coverage without completing it, because browsers drop that beacon under memory pressure, on power loss, and when a network filter blocks the request. Periodic heartbeats cost a request every few seconds and still stop when the tab dies.
A page that frequently ends visits is penalised twice: once in its bounce rate, and again in a time on page that logs a zero whenever a reader finishes there and leaves satisfied. Read exit rate alongside bounce rate before judging an exit-heavy page.
What a Good Time on Page Looks Like
No single benchmark for time on page applies across page types, because a pricing page that converts in 40 seconds is working while a 2,000-word guide read in 40 seconds is not. Compare a page against its own history and against other pages built on the same template.
Published industry averages come from sites unlike yours, so a template-level comparison, article to article and landing page to landing page, holds the variables still. Time on page is also confused with the counting metrics beneath it, and the distinction between sessions, users and pageviews decides which denominator an average gets divided by.
How SlimStat Records Time on Page
SlimStat stores two timestamps per pageview: one written when the page loads, and one written when the browser’s exit ping arrives as the page is hidden or unloaded. The Access Log prints the difference under a stopwatch icon, on the rows where that second timestamp arrived.
The Visit Duration report under Audience buckets whole visits into seven ranges, from 0-30 seconds to more than 10 minutes, and counts only visits not identified as bot traffic. A single-page visit whose exit ping never arrives lands in that 0-30 seconds bucket. Both reports ship in the free plugin, part of the wider case for a Google Analytics alternative for WordPress.
Frequently Asked Questions
These four questions cover what readers ask most often about time on page: what the metric actually measures, why a reported average looks so low, which figure counts as a good one, and how time on page differs from average session duration.
What does time on page mean in analytics?
Time on page is the number of seconds between a page loading and the next signal the tracker receives from that visitor. The signal is usually a second pageview, a tracked click, or an exit ping fired as the tab closes.
Why is my average time on page so low?
Zeros drag the average down. Every visit that ends on a page without a follow-up signal contributes a zero to that page’s average, so a site with a high bounce rate reports a figure well below what its readers actually spend.
What is a good average time on page?
No single figure applies, because the honest comparison is a page against its own history. Commonly cited benchmarks put blog articles at 2 to 3 minutes, though a 400-word answer page reaching 3 minutes would signal confusion rather than engagement.
What is the difference between time on page and session duration?
Time on page covers one page; session duration covers a whole visit, from the first timestamp to the last. A visit spanning 4 pages produces 4 time-on-page values, one of them usually zero, and a single session duration.