Generally speaking, most people want a lower bounce rate. There are only two ways to do that:
- Change the definition of a bounce.
- Change use behavior on your site.
Let’s start with the first one.
Changing the Definition of a Bounce
A bounce is a single-engagement session. It’s when someone visits your site and doesn’t interact in a meaningful way, and then leaves. There are different nuances depending on the system (such as with Google Analytics), but that’s the gist of it.
In most cases, you shouldn’t use an analytics system’s default definition of bounce rate. You should define what meaningful actions occur on your site, and how you can track those so they reflect user behavior accurately.
In terms of Google Analytics, this means thinking in terms of events.
Are there videos on the page to be played, forms to be filled, interactive elements like live chat to be clicked? Maybe you even want to track page scroll depth as an interactive event, especially if the page is primarily for content consumption.
Factoring any of things into your bounce rate using interactive events is what people mean when they say “adjusted bounce rate.”
Since we already discussed this comprehensively in other sections of this guide, let’s move onto the second way you can lower bounce rate: changing use behavior.
Changing User Behavior to Lower Bounce Rate
We’ve discussed before that a high bounce rate is not always bad. In fact, sometimes it’s a good thing. If someone comes to a help page and gets their answer, they leave. The page served its purpose.
So everything is contextual.
In addition, depending on the traffic makeup of the page, the bounce rate could vary wildly. Same goes for seasonal abnormalities; if you get Techcrunch’d or hit the front page of Hacker News, you’re going to get a spike of traffic and a spike in bounce rate.
So when you’re analyzing landing pages to optimize, specifically in regard bounce rate, you’ll want to focus on pages that are:
- Highly homogeneous (similar in purpose, such a series of product pages)
- Supposed to have an engagement hit or conversions happening
- Highly targeted with regards to traffic. If you’re sending paid traffic especially.
- High traffic pages. You want to focus on the highest opportunities possible.
So when you open up Google Analytics and look at bounce rates (Behavior > Site Content > Landing Pages), try to cluster pages by purpose. It shouldn’t be the case that a homepage has the same bounce rate as a landing page with only paid traffic. It’s all about context.
For example, if I were analyzing the performance of CXL Institute pages, I’d probably want to organize them by intent. So I’d want to filter out blog posts, help pages, and probably the home page as well since that’s where users go to log in (or I could use an advanced segment to filter out users, but that’s another discussion).
In this case, I’m looking simply at live course sales pages, giving me a much more purposeful way to analyze page performance:
This way, when I’m optimizing traffic acquisition or on page CRO, I can prioritize the highest traffic pages with the the highest bounce rates. These will be the most likely to see a positive effect with some work.
The Top Four Ways to Lower Bounce Rate
Now, before we dive into ways to lower bounce rate, one caveat: bounce rate is not the most important metric.
Don’t optimize only for lowering bounce rate. Some things that lower your bounce rate will lower your conversions and revenue. How? My first tip will be about popups, but as you know, popups can work wonders for list building, and in turn, revenue generation. It’s all about maximizing business value and ROI, so don’t miss the forest for the trees.
That said, here are some ways to lower bounce rate:
1. Think About How You’re Using Popups (and other disruptive features)
Popups are annoying. They consistently top the list of website experience that annoy users the most.
But of course, on the flip side, they undoubtedly do their job at collecting emails – and emails, in turn, make you money.
What’s the solution?
Craft any popups (or any experience for that matter) with the user in mind. Think: could a popup actually improve user experience? Why not? As long as they’re adding value (or at least minimizing disruption), I don’t see them as being a threat to a pleasurable internet experience.
I think WaitButWhy does a good job. It’s funny (self-aware), the popup is small, and when it is triggered, it doesn’t ruin the whole experience or take you from the page.
Even better, after reading this article for about 20 minutes, I finally got this scroll-triggered popup:
It did nothing to ruin my enjoyment of the article, and it’s written in the same self-aware and humorous style that Tim Urban uses in his blog posts.
2. Do Conversion Optimization Work on Your Landing Pages
Since you’re reading a conversion optimization blog, this isn’t a surprising suggestion. If a landing page has an enormous bounce rate, perhaps it’s because the on-page user experience is horrible. Therefore, run through the process of optimizing that page and improving it.
What does this process look like?
Though there are different ways you could manage it, generally you seek to find opportunity areas for testing as well as possible solutions to problems through qualitative and quantitative research. These get turned into hypotheses, which get turned into A/B tests, and you learn from those experiments.
We’ve written a ton on conversion optimization and A/B testing. Here are a few of the most comprehensive resources:
- Conversion Optimization Research Guide
- Beginner’s Guide to CRO
- A/B Testing Guide
- A/B Testing Statistics
- Landing Page Optimization Guide
3. Optimize Your Traffic Acquisition
It could be that the page is well-designed and you really couldn’t optimize it any further, at least to any significant value. Perhaps it is poor traffic that you’re bringing to the page? Unqualified, irrelevant, or premature traffic won’t convert like qualified traffic will, and your bounce rate will likely be higher.
How do you know you have mismatched traffic? The answer, too, lies in your data.
If, upon doing a comparison of conversion rates and bounce rates on your landing pages, you find that similarly designed pages (especially if they are templatized) have far different bounce rates or landing pages, you can assume it’s a traffic optimization problem.
It’s not necessarily a bad thing if you’re getting irrelevant traffic to landing pages (it doesn’t really hurt your revenue), but you should do some persona research and think about your marketing efforts in this case.
This traffic mismatch problem is often forgotten about in landing page optimization, but it can bring surprising results.
There’s no doubt that site testing is a good way to improve online marketing, but when it comes to “bang for your buck,” traffic optimization is often the logical first choice.
The benefit here is that you’re treating marketing as an exercise in frugality – how can I buy a dollar for fifty cents, as Warren Buffett might say. You’re chopping out irrelevant traffic, and therefore saving money as well as getting cleaner data, as opposed to dumping more traffic on top just to see if it works.
There’s a lot to this topic, to be sure. If you’re interested in learning more, we published an extensive article on how to optimize your traffic acquisition here.
4. Work on Site Performance
Site performance is super important to conversions and user experience. Mere millisecond delays have quantifiable effects on revenue.
An easy way to prove this, and to evangelize the idea of working on site speed in your organization, is to use Bounce Session and Non-Bounce Session segments in Google Analytics (both default segments) and look at Behavior > Site Speed report.
It’s almost always the case that page load times are much slower for bounced sessions. You can do further analysis on the effects of page load, either historically in Google Analytics by creating advanced segments based on page load time and comparing conversions, or simply by improving site performance and watching key metrics improve.
There’s a whole art and science to this of course. If you want to learn more, here’s an article on some low hanging fruit for increasing site speed, improving UX, and increasing conversions.
There’s more to it, of course, but that article is a good place to start.
A lower bounce rate is not always a good thing. It’s always contextual, so think about the purpose of your page and how you can improve the value of the page to the user experience – not just some proxy metric for success like bounce rate.
When it comes to lowering bounce rate, though, you can do it in two ways:
- Change your definition of a bounce.
- Change user behavior.
The first is very strategic and has to do with what events are important on your website. There’s no checklist that can tell you how to set up tracking to adequately explain user engagement, it’s something you have to have an organizational conversion over.
The second mainly has to do with conversion optimization and traffic optimization. Bring the right traffic to your site – relevant, motivation, and qualified traffic. Then make sure the page is relevant, without distractions, clear, and motivates the user to take meaningful actions (convert, or whatever).
A lower bounce rate isn’t (or shouldn’t be) a goal in itself, but as a measurement, it can help you decide if you’re moving in the right direction.
Since Google Analytics is far and away the most popular web analytics tool, it pays to know exactly how bounce rate is calculated in Google Analytics.
"What's a good bounce rate?"...common question, not a super clearcut answer. This piece will explore bounce rate benchmarks and give you a good baseline understanding of where you may stand.
While most people think of bounce rates in terms of website analytics, email bounce rates are a thing, too.
Now for the meat of the guide: how do we lower bounce rates?
Does bounce rate affect SEO? Some say yes, some say no, but this article spits the truth of the matter.
Unfortunately, the two metrics are often misunderstood and used interchangeably, when it is their subtle differences that make them useful metrics.
Summary and wrap up.