-
Is dwell time and bounce rate a ranking factor? And how does Google measure this?
Posted by on November 24, 2023 at 8:50 pmI’ve read in many places that the time a visitor spends on your site after clicking a search result is a ranking factor, which makes sense. But technically, how does Google measure this?
Do they only measure people who visit a result and then go back to the search results?Do they count the time a visitor takes to go back to the search results (quick bounce vs slow bounce)?
What about users that visit a site and close the page quickly (but don’t go back to the search results) vs a user that spends a long time browsing the site and closes it? Since neither goes back to the search results, can Google tell the difference?
Does Google use Google Analytics information to analyze how much users spend on each page? If so is it a good (or bad idea) to use Google Analytics.
I understand that many of these things are not disclosed by Google but just wondering if there is any information out there about this.
thanks!
​
-
1 Reply
-
Neither are ranking factors there is also no such thing as dwell time and they don’t use google analytics information as lots of website don’t even use it
-
Google has all the user data it needs to figure these things out and it’s a huge ranking factor. Right below search intent and proper keywords and heeat.
-
No.
-
Officially, no. Practically, yes
-
It’s easy when you use Google search google gives you a unique id associated with your browser, ip, current browser session etc.
When you visit a website and click on search result theyll be able to measure how much time you spend before you go back and do another search result, IF you do another search or click on other websites. Everything can be recorded in database
They don’t have to rely on the website using google analytics.
-
It’s not a ranking factor.
It can be spoofed, so it’s not a ranking factor.
It was even tested years ago and made literally no difference. Some guy wrote a script to click on a test website from various IPs and instantly click back, thus giving a high bounce rate. The idea was to see if thousands of people who were not enjoying the content would make a difference. It did not.
-
No its not. Look. There are many pages without content still ranking high.
-
Bounce rate was never a ranking factor. It’s not a factor because it really isn’t a good indication of whether or not the searcher had a good or bad experience.
If I search for an emergency plumber, pick the first search result, and call that business, in the old UA that would have registered as a bounce because I didn’t interact with any other pages. However, I found exactly what I wanted, so why would Google punish that page.
In the new GA4, bounces are measured a little differently. You have to spend less than 10 seconds on a page for it to register as a bounce. It’s a little more useful metric now, but I’m still not convinced that Google is using it.
What they more likely have used, but it’s something you cannot measure, is pogoing. Pogoing is when someone clicks on a search result, hits the back button to go back to the SERP, and then selects another search result. That could be an indication that the user did not find what they wanted on the first result.
Even pogoing though is not always a good signal. If someone is searching for reviews of something or shopping for the best price on a product, visiting multiple sites would be totally normal.
As for dwell time, I never really bought into that. In the old UA system it was measured terribly. All that was measured was the time between ‘hits’, not actual time on pages. And even then, what if someone gets up to go to the bathroom and comes back 5 minutes later?
I think something like scroll depth makes a lot more sense than how much time someone spends on a page.
-
They absolutely do and admitted it in court – what evidence do you have that google does not have the user data?
-
It’s not bounce rate it’s if the search was terminated. Did user go back and click other sites or do another search. More along those lines.
-
Officially, no.
Do they? *Yes*
Is it in any way available for us to measure the same way they do? No
Log in to reply.