Do Search Engines Consider Bounce Rate A Ranking Signal?

The bounce rate of your website might have a significant impact on your search engine rankings. If your site has a high bounce rate, you may end up having lower search engine rankings as a result. On the other hand, a lower bounce rate could end up giving your site’s ranking a boost (If you aren’t sure what bounce rate is, don’t worry we will be defining it below.)

And even if it isn’t a huge ranking signal today, Google and friends are getting smarter all the time and my prediction is that it will most certainly play a role.

Don’t agree with me? Fine, but SEO Black Hat recently conducted a study that shows that a website’s ranking was significantly impacted due to large bounce rate changes. Of course it’s possible that the study’s data could have been impacted by other factors occurring simultaneously, so it does mean that the study is only corollary at best.

Even if this isn’t true (yet), I speculate that it will be a ranking signal soon enough. The fact is, search engines want their search results to list quality sites. High bounce rates can be a good indicator that the site provides a poor experience for searchers, or even worse, that the user’s search query and the site content are a complete mismatch. This offers plenty of motivation for bounce rate to be looked at as a significant SEO factor, and gives webmasters strong incentive to attempt to reduce bounce rates on their sites to hopefully increase their rankings in the search results.

However, high bounce rate isn’t always a low-quality indicator. There are plenty of sites that are delivered to me in the SERPs that take me to the page of a website containing the content that I am looking for. If this content doesn’t take long to consume (such as a weather forecast), then me going back to the SERPs to continue browsing the world wide web for other things I’m interested is not at all an indicator that I didn’t get what I was looking for.

Definition Of Bounce Rate

So what is bounce rate anyway? A bonce is defined by Google Analytics to be any visit where only one page of a site is viewed by a visitor before the individual takes another action. So what happens next? Several things are possible, including the visitor clicking on a link that goes to another website, hitting the back button and leaving the site, typing in a new URL, closing an open tab or window, or a session timeout where the visitor doesn’t take another action. There is still some fuzziness involved due to how analytics packages define sessions. Analytics software relying on Javascript tags just know when somebody loads one of your website pages (the Javascript runs).

So, this type of analytics package has a hard time knowing what occurs in the meantime. In the first example, user A visits your website, looks at one page, and then goes off to lunch. He comes back and then visits 10 other of your site’s pages. Due to how these analytic packages work, it will be viewed as two separate visits. The first one will get recorded as a bounce (one page visit).

There are some other possibilities of how search engines define a bounce. A bounce might be defined as a user typing in a search query, visiting your website, going back to the search results, and then clicking on a different result on the page. Another potential definition could be the user typing in a search query, visiting your website, and then going back to the page of search results in under x seconds.

The major search engines, of course, have other data they can use also. Their toolbars can be full of data that can be used to track user actions. Also, data is licensed by the search engines from major ISPs. Additional data is collected to track where a visitor goes. So the possibilities extend well beyond what can be done with an analytics package.

Again, there are a few potential issues with using this approach to determine bounce rate, and then using it to define the quality of a website based on it. What if the searcher is looking for one small bit of information, like the birthplace of James Hetfield. The user could click on one of the search results and be taken to a good reference site that provides the answer right away, so the visitor gets his answer and is then is done with his query. He could still return to the search results and look for something else, or click on a different result from his original one. Although it was a satisfactory outcome, most analytics packages will record it as a bounce.

This kind of scenario is prevalent with searchers looking for a simple answer to a question and getting their answers from sites with almanac-like information. The key is to try to also factor in comparative data in how bounce rate is used to influence rankings. As an example, an almanac site’s bounce rate might be higher than an e-commerce site’s bounce rate, which most likely will be higher than a directory site’s bounce rate (where there is a good chance that a user will visit another website).

My best guess is the search engines are looking (or will be looking soon) to determine how the bounce rate of a site compares to other comparable sites, or other websites that it is considering to return for a particular search query. In the second scenario, a run-time adjustment can be imagined where a search engine returns with “traditional” results for a searcher’s query, and then a bounce rate adjustment is made.

Bottom Line

It is very possible that bounce rate significantly factors into ranking at the present time. Even if it doesn’t right now, I believe it will become a significant factor very soon. Skynet is getting smarter every day, so even if it isn’t a factor currently and doesn’t become one in the near future, there are still several reasons for taking a close look at bounce rate. It is an indicator of your site’s conversion potential, so this is definitely a good enough reason to review it.

Comments