Bounce rate is an awesome concept, but just doesn’t tell us much these days

Bounce rate is the rate of which someone arrives at your site and then leaves without visiting another page. When sites were geared toward desktops, providing information “above the fold” and logically dividing information by page, the bounce rate was this awesome metric that gave great insight as to how engaging your concept was, as well as offering insight into the quality of your traffic.


faling bounce rates

According to the data above, the site is having big issues. The overall bounce rate has gone from 12.3 percent to 45 percent! That has to be bad right?

For several reasons bounce rate is not as straightforward as it was in years past:

  1. Pages today have grown in length (related to responsive design and users preference to scroll over click). Sites are just not as geared toward “the next click” and users can, and should, get the information they require on their first visit, on the page they land on. Google is also helping people find the most relevant content on your site, making this more and more likely for both organic and paid search channels.
  2. Mobile usage can lead to higher bounce rates due to the lesser amount of time people have looking at a site when they are on mobile, as well as them being more sensitive to page speeds.
  3. Mobile bookmarking also grows in popularity and understanding, making it more likely to have people come back directly to the page they want (compounds the mobile shift issue).
  4. As a site is optimized, often key information is propagated to other pages that makes bounce rate go up.

For Example:

When Out of Stock status first goes on gallery pages, those pages may see an increase in bounce since the gallery page may answer the inventory question that previously could only be answered on a product detail page. Pssst. This is what happened in this case, among other improvements.

It’s easy to see in this case, as often happens with optimizing a site, the bounce rate being lower due to users having to click to get key information, not because the site was better.

We have to recognize that bounce rates will change mainly due to four factors:

  1. On-site experience change (this is what we think of typically)
  2. Off-site experience change (most easily seen with affiliate traffic)
  3. Traffic source mix (increase high bouncing display traffic and see an overall increase)
  4. Device mix (as mobile adoption increases, expect bounce to go up)

Because of these four factors, the best time to look at bounce rates is before and after a change period. Outside that, there is typically too much change happening to compare them across longer time periods.

The New Way

Today, we really need to look beyond bounce rate at a variety of engagement metrics. Below are the metrics you can look at to get a better sense of user engagement, than with bounce rate alone. The following is a custom report with one custom metric, “return visit rate,” which is explained below.

engagement report

The Metrics

Users: Before you look at any two date periods, you need to orient yourself as to the number of users. Above we see that the number of users increased more than 11 percent between March and June. This is a big number, and if the increase came from a highly engaged segment of users (i.e., email to existing customers), then that has to be factored into the rest of the report.

User Conversion Rate: It is also important to orient yourself about conversion rates between two periods. If the user conversion rate has gone up, then things are likely good (again, know where your traffic is coming from, as an increase in “good” traffic will lead to an increase in overall conversion rates).

Revenue Per User: Like user conversion rate, you want to know what your eCommerce site is doing in terms of revenue per visitor just to orient yourself to the overall date periods and how they may compare.

Page Value: This is a fantastic metric to use to measure engagement! As you optimize your site, you should be driving people down paths with the best experience, which will reduce the number of times they hit pages that offer them no value. As you do this, the overall page value will increase since fewer users are touching less pages that influence a positive outcome.

Return Visit Rate: OK, this does not exist out of the box, but rather is a custom metric. This will tell you if people are coming back to your site more often. This is the best indicator that your site is engaging, as few people buy on their first visit. Even if they do, they are likely to return to the site to check on shipping status, look at other items and click on emails from you.  If you see your return visit rate increases between two time periods, it’s either because you are seeing better traffic or have created a better experience. This is a great metric to look at while A/B testing.

The Setup

To calculate this yourself, go into the Google Analytics Admin area and add the following calculated metric.

Calculation metrics

Average Session Duration: This is a hard one to get wrong. You can improve your site, while at the same time decrease pageviews per session, but I’ve rarely seen an eCommerce site improve and see a reduction in its average session duration.

Bounce Rate: While not the harbinger of yesteryear, it does add context to the user’s experience and helps understand the user’s experience a bit better, as well as help you ask better questions. Expect this to go up, especially after a site has been improved, so people don’t have to click around to get all the info they need to make a purchase.

In Summary

If you know your site well, no change in bounce rate will bother you, because you should understand when and why it changes. BUT if you see a change in return visit rate or page value, you’d better roll up those sleeves and dig in.