User Engagement Affecting Rank

How quickly could a decrease in user engagement affect rank?

[rwilson]

I have a site that recently the average time on page decreased nearly 50% due to a survey we implemented. Around the same time our SERPs dropped. Not sure if there is a direct correlation. Any thoughts on how quickly Google would react to this signal and change rank?

[fathom]

I don't believe there is any way to categorically determine if UX had any impact on ranks.

In theory, it is suggested to be a quality signal but in practical I doubt it since Googlebot doesn't mingle with patrons.

[tangor]

I think I'd be more concerned with the drop than page rank. If something "negative" happened to the page, reverse that and see if engagement returns.

From a personal perspective, if a page wants me to take a survey to continue, I just back out and go somewhere else.

[engine]

Assuming that's the kind of survey you're talking about,

... if a page wants me to take a survey to continue, I just back out and go somewhere else.

Me, too.
If that takes me straight back the SERPs that's a signal that the visitor did not find what they wanted on the site.

Of course, the drop may also be connected to other factors.

Take the survey off the site and watch to see what happens.

SEO is often about experimentation and testing.

[fathom]

Ya speculation on what Google is speculating about seems impossible to nail down and it very well could be something more tangible like a loss of link juice from two months before that.

[rwilson]

I think you're all right. I think my drop in rank is much bigger than time on page affecting user engagement. Now that Google confirmed their content quality update, I'm chasing that as I'm sure content could be better on my site. The survey is off my site now, I think I'll be able to get a better picture of user engagement now.

[engine]

Don't forget to test one thing at a time. If you change several, you don't know which was the issue.

[ken_b]

I have a site that recently the average time on page decreased nearly 50% due to a survey we implemented.

Did the time on page drop because visitors are using their back button, or does the survey take them offsite?

Is the survey on all pages? If not I guess we should ask if the time on page drop is only on the page(s) with the survey, or all pages?
.

[fathom]

Did the time on page drop because visitors are using their back button, or does the survey take them offsite?

Is the survey on all pages? If not I guess we should ask if the time on page drop is only on the page(s) with the survey, or all pages?

Ya... making decisions based on faulty observations pushes you in the wrong direction.
.