Dwell time: what it is in SEO and how to measure proxies

What is dwell time?

In SEO, dwell time is the time between a searcher clicking a result and returning to the search results page. A searcher who clicks, reads for three minutes and never comes back has a long dwell time; one who clicks and bounces back after five seconds to try another result ("pogo-sticking") has a short one.

The term was popularized in SEO around 2011 by Bing's Duane Forrester. Outside SEO, "dwell time" also means time spent in a zone (retail, logistics) or, in security, how long an attacker stays undetected; this page is about search and web analytics.

Dwell time vs similar metrics

The key point: your analytics cannot measure dwell time. It sees the visitor arrive and act on your site, but not whether they went back to Google or closed the tab. Time on page and engaged sessions are the closest proxies you have.

Is dwell time a Google ranking factor?

Google has never confirmed "dwell time" as a named ranking factor, and its representatives have repeatedly said they do not use such a metric directly. Google does say it uses "aggregated and anonymized interaction data to assess whether search results are relevant to queries", which it turns into signals for its machine-learned ranking systems. In practice:

• Treat dwell time as a symptom of whether a page satisfied the query, not as a lever to pull.
• Tricks to inflate time on page (hidden content, slow-loading reveals, autoplay video) make the experience worse and do not fix the underlying mismatch.

How to estimate it with analytics proxies

You cannot see the return to the results page, but you can find landing pages where search visitors leave quickly. For each organic landing page, look at:

1. Engaged rate of sessions from search engines
2. Share of single-page sessions under 10 seconds
3. Median time on page (not the average, which a few long visits inflate)

Sample data for three organic landing pages over 30 days:

The third page attracts plenty of searches but almost half of visitors leave within ten seconds. That usually means the title promises something the first screen does not deliver, for example a single number when the page opens with a long introduction.

How to improve dwell time (by satisfying the query)

• Answer the question in the first screen. Put the definition, number or step the query asks for above the fold, then expand.
• Match intent. A "what is" query wants a definition and example; a "how to" query wants steps; a "best" query wants a comparison. See how to improve organic search.
• Make the page fast and stable. Slow loads and layout shifts send people back; see Core Web Vitals.
• Help the next step. Clear headings, a table of contents and relevant internal links keep people who want more on your site.
• Rewrite misleading titles. A click that ends in disappointment is worse than no click.

Common mistakes

• Using average session duration as dwell time. It includes all traffic sources and is skewed by long sessions.
• Comparing pages with different intents. A quick-answer page with a short visit can fully satisfy the searcher.
• Ignoring zero-duration sessions. Many tools record a duration of 0 for single-page visits without a later event, which says nothing about how long the person actually read.

In TapCub

TapCub counts a session as engaged when it lasts at least 10 seconds, has at least two pageviews, or hits a key goal event. Filter the pages report to search-engine traffic and sort entry pages by bounce rate and average time to find candidates. Session duration is the time from the first to the last event, so a single pageview with no later event and no leave signal shows as 0; read time-based metrics together with scroll depth and events rather than alone.

Sources

• Google Search. How results are automatically generated: ranking results.
• Backlinko. Dwell time in SEO: what it is and why it matters.
• Wikipedia. Dwell time.