What Is Dwell Time?

Dwell time is defined as the time that passes between a user clicking a result on the search engine results page (SERP) and returning to that same results page. The logic behind the concept is easy to follow: if someone clicks a link, spends a good while on the page, then goes back to Google to search for something else, the content probably matched what they were looking for. If they come back within a few seconds, that suggests the page didn't meet their expectations.

Here's the point that needs to be clear from the start: Google doesn't publish dwell time as a figure, and there's no way to look it up in any standard analytics or Search Console tool. Google Analytics doesn't collect it, because Analytics only sees what happens on your own site: it has no way of knowing whether a user arrived from a Google search, or whether, after leaving, they returned to the SERP or simply closed the tab. Search Console doesn't measure it either: its reports cover clicks, impressions, CTR, and average position, none of which touch what a user does after clicking.

That's why dwell time is better described as a theoretical SEO concept than as a metric: it describes an idea about how a search engine might behave, not a number that exists in any system a marketer can access. Any article that shows a "dwell time figure," or promises to "increase your dwell time with these tactics," is at minimum oversimplifying.

The difference isn't a minor technicality. A user can spend four minutes reading your article and then open ten more tabs without ever going back to Google: that produces high time on page and, at the same time, no reconstructible dwell time, because the user never returned to the SERP. The only one measurable with your own tools is the former.

The term dwell time didn't originate at Google. It was popularized by Duane Forrester, then head of webmaster communications at Bing, in a 2011 blog post explaining how a search engine might use the time between a click and a return to the results, alongside other behavioral signals like the "long click," as a quality signal [verified, corroborated independently by Ahrefs and Semrush]. That's the source of the recurring confusion: many people credit the concept to Google when it originally described how Bing might work.

On Google's actual use of dwell time, the spokesperson who has addressed it most directly is Gary Illyes, Trends Analyst for Webmasters at Google. On November 6, 2019, he responded to a direct question about the metric with "we're not using such metrics" [verified, the quote appears independently on both Ahrefs and Semrush]. He's also been credited with a more informal public remark calling dwell time and CTR "generally made up crap," made while commenting on unconfirmed SEO theories [reported by only one source I could access, Semrush; I wasn't able to check it against the original recording or a transcript, so I'm treating it as not independently verified].

Martin Splitt, from Google's Search Relations team, is also reported to have made similar comments in 2019, denying that Google uses this kind of engagement metric for ranking [reported by only one source, Semrush; also not verified against an original recording]. I didn't find a direct, sourced quote from John Mueller specifically about dwell time. What multiple Google spokespeople have repeated over the years is the broader position that Google Analytics behavioral data isn't shared with Search's ranking systems, because they're separate products.

Google's own guide to its ranking systems doesn't list dwell time among the documented systems either, though the document itself clarifies it only covers "some" of the more notable systems, not all of them. So there's no public confirmation that Google uses it, but also no permanent, absolute denial, given that Google doesn't typically disclose the full internal workings of its algorithm.

What you can look up is GA4's engagement time, or "average engagement time," together with the share of engaged sessions. As explained in detail in the article on Bounce Rate, GA4 treats a session as engaged when it lasts longer than 10 seconds, includes two or more page views, or fires at least one key event; that figure measures activity on your own site, not whether the user later returned to Google.

In Search Console, you can cross-reference a page's CTR and average position against its impressions and clicks to spot indirect patterns: a page that ranks well but has a very low CTR for its position points to a title or description problem, not dwell time. A position drop after several weeks of solid traffic can have dozens of causes, and pinning it on dwell time without more evidence is an oversimplification the available data doesn't support.

As additional proxies, some teams use heatmaps, session recordings, or scroll-depth percentages to approximate how genuinely interested users are. These are useful signals for improving experience and content, but none of them literally measures dwell time as Forrester defined it in 2011: they remain limited to what happens within your own domain.

What matters more than any single snapshot is the trend over time. If a page's engagement time rises noticeably after a content rewrite and its share of engaged sessions stays consistently high, that's a more solid signal of improved content quality than one value compared against an industry average. That relative comparison, before versus after on the same page, produces more useful conclusions than trying to guess an absolute dwell time figure that isn't retrievable anywhere in the first place.

• Treat GA4's "engagement time" as a content quality signal, not a literal stand-in for dwell time.
• When reporting this metric to a client, always clarify that it measures activity on your site, not the return to the SERP.
• Combine engagement time with CTR and position from Search Console before drawing conclusions about a specific page.
• Use heatmaps or session recordings to understand why a page holds or loses users, rather than trying to infer it from an average alone.
• When citing Google on this topic, link to the original source of the statement instead of repeating a secondhand quote.
• Prioritize content improvements that reduce real user friction, like a clear answer in the opening paragraphs, over tactics designed purely to "hold" time.

• Presenting dwell time as a confirmed Google ranking factor when no official statement supports that.
• Confusing dwell time with GA4's time on page, assuming they're the same figure.
• Promising a client that dwell time can be "measured and improved" with standard analytics tools.
• Attributing a ranking increase or drop to dwell time without first ruling out other documented factors, like technical changes, competitor content, or algorithm updates.
• Repeating secondhand quotes from Google spokespeople without checking the original source or the context they were said in.