If there’s one thing my clients love talking about, it’s bounce rate.
Clients getting excited about anything in analytics is great, unfortunately, I tend to see a bit of over-emphasis on this one metric.
So let’s dive into when you should focus on it, when you should not, and, finally, yes we will talk about what a good bounce rate actually looks like (warning: you won’t like my answer).
What is bounce rate?
Makes sense to start with the basics, right?
In very basic terms, bounce rate is the proportion of visitors who land on a page, then leave without clicking on something important.
For example, if I click through to your home page from Google, scroll around for a bit then decide this isn’t really what I’m looking for, then leave, then that is counted as a bounce.
Alternatively, if I land on your home page, click through to your product page and then realise that your products are not actually what I was after, then leave, then that would not be considered a bounce (because I made a click to the product page).
Ok, got it. So that means that a lower bounce rate is better, right?
As a general rule of thumb, you obviously want more people interacting with your site rather than leaving as soon as they arrive. So yes, lower is generally better.
But that is not a hard and fast rule.
For example, we can probably agree that a bounce rate of 40% is better than a bounce rate of 65%. But what if you saw you had a bounce rate of 2%? Does that mean that you’ve hit the website jackpot with a perfect website that nobody can resist?
More likely it’s just too good to be true.
I think that Rocket Fuel summed it up perfectly with this graphic:
In particular, I want you to pay attention to the extremes on this graph: the very low and the very high, both of which are described as “something is broken”.
So we shouldn’t be aiming for a bounce rate of 0?
Simply put, web analytics that are very close to 0 (or very close to 100%) usually indicate that either your website or your analytics tracking code are not working properly.
A bounce rate sitting close to 0 usually means that there is a problem with your Google Analytics tracking code. The most common instance of this is when you have two instances of the tracking code installed on your website.
This is very common, for example, if you have just launched a redesign for your website and in the process re-applied the tracking code that was already there. Or perhaps you upgraded your tracking code to a Universal Analytics code, but forgot to delete the old one.
Either way, having two instances of the tracking code means that as soon as a user lands on your site, the first code will fire, then the second will fire and register as an interaction on your page. In other words, it’s almost impossible for a user to bounce since they always register an interaction as soon as they arrive.
Bounce rates near 100% are less common, usually coming about from simple one-page websites, or sites with broken architecture that prevent users from clicking anywhere else.
In general, I tend to view any bounce rate below 20% as indicative of a tracking code problem, and any bounce rate that is over 85% as indicative of a website problem.
Ok, so what is a good bounce rate? What should I be aiming for?
Remember how I warned you that you are not going to like my answer to this question? That’s because:
I know, I know, that’s not a helpful answer. So let me explain.
When you think about bounce rate, you need to think about the purpose of your page. Because like it or not, some pages are supposed to have high bounce rates.
Once you understand what you are actually trying to achieve with your site, you can better assess whether or not its bounce rate is too high.
Let me take you through some of the things to keep in mind. I promise it’s not too technical.
Lesson 1: stop focusing on your website’s overall bounce rate
There are two different types of bounce rate that you might come across, even though it might not be immediately clear that they are different.
The first is what I’m going to call an aggregate bounce rate. This is the bounce rate of multiple pages taken together. Aggregate bounce rates are bad.
You will see this in most of the reports in Google Analytics. For example, if you look at your Channels report, there is a column that shows the bounce rate of each channel:
But the most egregious example of this is the site-level bounce rate, which you can see at the top of the highlighted column here, and also on what many people (wrongly) use as the front page of Google Analytics:
This little fella is the bane of my work life. It shows up when clients with that dangerous little bit of knowledge log in to Google Analytics and see this on their Overview dashboard. It shows up in poorly thought-out reports developed by SEO agencies who don’t think about anything beyond vanity metrics. It shows up in top-level website summary reports given to the boss. It’s everywhere.
Why is this so problematic? Well, think about it. What does the bounce rate of an entire website actually mean? And what does it tell you about the site?
Well, it literally means that if you take all of the sessions on every page of your website and average them together, 51% of those sessions (in this case) bounced. Which means that to fix our website we need to do…what?
The point is that bounce rate measures the activity of a user on one page. Averaging bounce rate across multiple pages doesn’t tell us anything at all, it’s just muddying the data. It’s like trying to take the average hair colour of the world’s population—you can do it, but it just doesn’t give any useful information.
Ok, so what bounce rates should I be focusing on?
The bounce rates for individual pages.
To find this, you want to click the Behaviour menu on the left of Google Analytics, then Site Content, then either All Pages or Landing Pages.
Now we’re talking! The first column will give you all of the pages on your website (I’ve censored them in the example above, just use your imagination), and the Bounce Rate column will give you the bounce rate for that particular page! Woohoo!
(Just remember to ignore the number at the top of the column. That’s an evil aggregate bounce rate!)
Ok, great, so I see lots of different bounce rates for all my pages. I still don’t know what’s good and what’s bad.
That’s ok, we’re getting there, I promise.
Now that we’re at least looking at the right thing, let’s have a think about why the bounce rates on those pages might be different.
Different pages have different purposes
Let’s consider three different pages on our website:
- Home page
- Product page
- Blog article
Each of these pages has a different purpose for our users. We might describe those as:
- Home page: introduce user to our brand, our purpose, and our products
- Product page: give an engaged user more information about our products, guide them towards a sale/conversion
- Blog article: provide our user with relevant information
Naturally, each of these purposes is going to have a different impact on bounce rate:
- Home page: medium bounce rate. Since this is a very general introduction, some users will be interested and want to click to other parts of our site, but others will bounce.
- Product page: low bounce rate. Users on this page have already shown that they are interested in your site and your products, so it should be fairly easy to get them to make one more click.
- Blog article: high bounce rate. Since the purpose of this page is to get users to stay on the page and consume the content, there is no reason to expect they will click anything else, and we should expect most users to bounce from these pages.
The most important thing I can emphasise here is that none of this is bad. Yes, even that blog article with a high bounce rate. It is simply the natural behaviour for traffic on that page, given how we have designed the page’s purpose.
You still haven’t told me what a good bounce rate actually looks like
Well, I’ll tell you now. Are you ready?
It is all relative to your website’s baseline.
Given what we just discussed, we can expect each page on our site to have a better or worse baseline than, say, our home page.
For example, let’s say our home page has a bounce rate of 55%. Then I would assume that pages that feature my products and services would have a bounce rate of less than 55%, while pages that contain information for my users (including things like my contact page) would have a bounce rate higher than 55%.
Now, when I go into Google Analytics, let’s say I discover that my core product page has a bounce rate of 63%! Well, obviously something has gone wrong with this page. My users should not be bouncing off this page more than my home page. So perhaps I need to revisit the content, design, or structure of this page in order to rectify this issue.
Looking at each page relative to your own site’s baseline is much more informative than if I just told you “aim for a baseline of 40%”.
Are you sure you can’t just give me a number to aim for?
You see, the problem is that each website is unique.
For example, every industry has a different baseline bounce rate.
Take a look at this, courtesy of Google Benchmarks 2017:
For example, this tells us that websites about “autos and vehicles” have an average bounce rate of 46%, while news websites have an average bounce rate of 65%.
Great! So we can just look up our industry in that table and there’s our number! …Right?
I’m afraid not.
Each website also has its own traffic profile that contributes greatly to its bounce rate.
For example, Paid traffic (from your Adwords PPC campaigns) almost always has a much higher bounce rate than Organic traffic. Referral traffic has a widely varied bounce rate depending on the referral source. And so on.
That means that if you take two websites in the exact some industry, but one relies heavily on Paid traffic, while the other relies on Organic traffic and Referral traffic from an article on a popular news site, they will have completely different bounce rate profiles.
So there’s no point comparing my website’s bounce rate with any other?
If you take nothing else away from this, remember that your website is not comparable with any other, and every page on your website is different.
Try to always keep in mind the purpose of the individual page, never the website as a whole, and you can always figure out whether your pages are performing better or worse than they should in terms of bounces.