Does a bounce make a sound?

This week, I've been thinking about that old zen koan:

"If someone visits a single page on your website without triggering an event, will their visit impact your bounce rate?"

Bounce rate is a metric that comes up pretty often in my conversations with clients. People often think that a high bounce rate is bad and a low bounce rate is good. But it’s not that simple.

First, let’s define the term.

A bounce is when a visitor comes to a page on your website and then leaves without triggering an event. An event can be triggered when someone goes on to a view a second page on the site, but an event can also be triggered by playing a video or submitting a form. So, if someone lands on your contact page, submits a form, and then leaves — that’s a single page visit, but it’s not a bounce. Or someone could visit a blog post on your site, read the whole thing twice, copy the URL and paste it into an email to everyone they’ve ever met, close the window, and their visit would count as a bounce.

Each page on your site has different goals and opportunities for users, so it’s important to assess bounce rate on a page-by-page basis. Your website’s overall bounce rate is only meaningful when viewed over time.

Imagine calculating the average temperature of every major city in the US on January 30th and then saying, “The United States is 42 degrees.” That's not a very helpful assessment. That average could be interesting if you looked at it over time. You might spot a trend then, but you wouldn’t conclude that the United States is a cold country based on that sample. Keep that in mind when someone cites your website’s overall bounce rate is too high and uses that as evidence that it’s time to “REDESIGN EVERYTHING!!!”

Let’s look at some more examples.

If you have landing pages that are meant to get people on your mailing list, a high bounce rate would be bad. But a high bounce rate on a page where someone is coming just to get a phone number or some support content could be a good thing. In that case, the customer searched, found what they were looking for, and left happy.

The quality of traffic to a page is another thing to consider. Imagine you’ve created an in-depth guide on your website that has valuable information for a certain type of visitor, but it has an 80% bounce rate. You might think that you need to somehow “fix” the content, but it could be that the guide is poorly labeled in your navigation. In that case, there’s nothing wrong with the content — you’re just bringing the wrong people to it or the right people at the wrong time. (Not everyone is ready to read an in-depth guide before they understand something about your organization, for example.)

The comparison function in Google Analytics is a great way to uncover many content issues. If you sort by sessions and compare pages against the average bounce rate, you might find something like this:


The screenshot above is from the Google Merchandise Store. You can see that lots of people are visiting the Youtube portion of the site. That page has the second highest number of page views over the past few months, but the page’s bounce rate is 50% higher than that of other pages.

You might see that and think something is wrong with the content. But when you compare by revenue, you see that the page is performing slightly above average for the site as a whole:


In this case, it might make sense to begin segmenting by traffic source to see if a particular kind of visitor is contributing to the high bounce rate. Another thing to check: Is a popular item on this page very expensive? Maybe the conversion rate is low for the page, but a single expensive item is keeping revenue high overall …

The point is that bounce rate is a valuable secondary metric, but it should be viewed as a piece of a larger puzzle.

So what is a “good” bounce rate?

It depends.

But if you’re seeing a single digit bounce rate for your website overall, that’s not good. It probably means your analytics configuration is broken. I’ve seen sites with analytics code in multiple places, which caused each visit to trigger two page views, so the bounce rate was something like 3%.

Conversely, a 95% bounce rate may mean that events aren’t triggering properly on a page.

When looking at bounce rate, always be asking: What is the business goal for this page? What is the next thing we want the visitor to do? Not every visit is going to result in a transaction because not everyone is ready to convert — content that assists in a future conversion is valuable, too. Then, try to discern the visitor’s goal in visiting that particular page. (On-page surveys and funnel analysis can help.)

It’s tempting to look at a metric like bounce rate and draw quick conclusions, but context is everything — a bounce rate by itself is almost meaningless.

Thanks for reading,


Kyle Bowen