Brian Cray ··· Home > Blog > Analytics > Measure what is relevant: A critical look at bounce rates

Measure what is relevant: A critical look at bounce rates

Every time I see "web analytics" I inevitably find "bounce rate" trailing close behind, being thrown around as if it's some kind of silver bullet for measuring user engagement. "Get your bounce rate low, and life is good." Wrong.

Bounce rate's simplicity is both its greatest strength and its greatest weakness. Simplicity gives it the strength to be easy to read and measure over time. Its weakness is that it makes people lazy.

So lazy in fact that people are failing to ask the most important strategic question about measurement: Are my methods of measurement relevant?

Is bounce rate relevant?

In order to "knock bounce rate down a peg or two" we need to first clearly define it: Bounce rate is the percentage of users that leave on the first page they visit.

"The first page they visit." Therein lies the problem.

Consider these 3 questions in which the bounce rate is actually counter-productive:

1. What if user engagement has nothing to do with multiple page views?

Let's use a blog as a case in point. About a month ago I used Survey Monkey to survey my readers. The survey revealed that approximately 90% of my visitors came for one thing: to read an article. While this may not be the same in every case, I don't think far from the center of the proverbial bell curve.

If users come to read an article, doesn't it make sense that the time they spend reading the article is the correct form of measuring engagement? How many times do you read more than one article?

Sure, ideally content pulls visitors further into your site, but pull them in by understanding and designing to their intentions first. Otherwise you're firing at the wrong target.

2. What if your conversion leads them to another site?

Many small to midsize ecommerce and micropayment sites use PayPal or Google Checkout to handle payments. Most web analytics software—including Google Analytics—only measure additional page views on your site, but isn't checking out with PayPal a conversion?

Or what if you have a microsite designed for a specific campaign and you want visitors to jump to your brand's main website. That jump, while a desirable effect on the surface, is reporting as a bounce! Oh no! You've been fooled!

3. What if you're successfully converting users on the first page they visit?

Online support systems still rely on phones at the least as a last resort support function. Would you rather pay the support cost to take a phone call or lose a customer? Hopefully you're in the latter group and make it easy for people to call you for support. And if your support is good, you've just built brand loyalty. But your bounce rate will tell you you've just lost a user. Oh no! Fooled again!

There are more examples of online calls to offline actions. What if you want your users to print out a coupon and/or visit your brick and mortar?

How to prevent being mislead by the bounce rate

Don't let the bounce rate tell the story for you.

First, consider your website objectives and what your users intend to do. Hopefully those two compliment each other. Second, ask yourself how users will actually fulfill your objectives through a website. Finally, find metrics that measure user behavior that is relevant.