The Coronavirus pandemic has put statistics and data at the forefront of the public debate.
We’ve heard ‘R-Rate’ talked about a lot, and it’s been used as one of the key measures of where and when restrictions are put in place. Keeping the R-Rate below 1 has become something of a national obsession.
The problem with R-Rate, though, is that there are so many variables and outliers which can sway the number one way of the other.
At one point England’s R-Rate was the highest it had ever been while new cases were the lowest they’d been since the outbreak.
The number can be inflated by an isolated spike in infections in a small community, not giving a true reflection of the situation nationally. As an arbitrary figure, it’s not reliable enough to provide us with enough context to make informed decisions.
That data is absolutely meaningless without context.
So what’s that got to do with SEO and digital marketing? It’s a nice little segue into something which has been bugging me for a while – our obsession with bounce rate and the constant striving to lower our bounce rate stats.
This is where much of the confusion lies with bounce rate and why we get so hung up on it.
Having spoken with many clients on the topic before, it seems that a lot of people think of bounce rate as the percentage of visitors who land on a page and immediately ‘bounce’ away from it. In this regard, you can understand why it’s a metric that website owners would want to lower as much as possible.
But this isn’t the case.
Bounce rate measures the percentage of users who have a single-page session on your website. They land on the page and exit without visiting another page during that session – but the time they spend on that page isn’t taken into account.
It could be 5 seconds or 5 hours and still count as a bounce if it’s the only page the user visits.
In that sense, a more accurate name for bounce rate might be ‘single-page sessions’, which would certainly soften some of the edges and help with our comprehension of exactly what’s happening on our sites.
Before I started writing this article, I did some research into bounce rate and looked at a few studies to make sure I wasn’t about to write absolute bollocks. I’ve probably managed to do that anyway, but let’s crack on.
I opened up a few tabs on the topic from sites which are generally considered to be authoritative in the SEO industry.
I spent a lot of time reading them, digesting and absorbing the information.
I found them all incredibly useful.
I closed all the tabs down once I was finished with them. Session over, one page visited.
A lot of other users landing on that content have probably done the same thing, as well. But the content did everything it was supposed to do:
But for the webmaster at the other end looking at their bounce rate statistics, none of this will be taken into account.
This is where the most crucial element of any bounce rate analysis comes in – context and purpose.
Here we go. The SEO’s mantra. The inscription we’ll all have on our gravestones.
The answer here really is “it depends”. It depends on which channel the user came from (PPC, paid social, SEO etc). It depends on the search query they used to land on your page. Above all, it depends on what the purpose of the page is.
If you have a high bounce rate on your homepage, that’s probably not good. It’s the gateway to the rest of your site, the funnel that brings users through to your money pages and quality content.
If you’re an eCommerce site and have a high bounce on your category and product pages, that’s probably not good either. These are pages where we want users to take action – add a product to the cart, view related products or save to a wishlist. All of these would prevent the session from being recorded as a bounce.
But if you’ve built content designed to educate and inform then a high bounce rate isn’t necessarily a bad thing, for exactly the reasons I highlighted earlier on.
Of course, there are plenty of things we can do within this kind of content to encourage users to take further actions on the site:
Ultimately, though, when the page is built primarily to inform, we really shouldn’t get too downhearted if the bounce rate remains high.
It’s important to remember the impact that we can’t directly measure when we create this kind of content: brand exposure, authority and market placement.
Going back to the example of my research for this article, although I didn’t explore any more of the site in that first session, I have kept a few bookmarks.
I spent a long time on the page, but it doesn’t matter if I spend 2 seconds or 2 hours on a page – so long as it’s the only one I visit in that session, Google records it as a bounce.
I’ll go back to them again, and if there are any other similar topics I need to research I’ll go back to those same sites.
The more I go back to those same sites, the more likely it is I’ll become a paying customer of their tools or software someday (as opposed to other providers who aren’t giving me authoritative content).
Many of those sessions I spend on the site will involve just one page and count as a bounce.
With each touchpoint, though, I inch another step close to being a paid customer.
It’s a long game to play, but it’s one that proves very profitable for a lot who invest in proper content marketing strategies.
You might think I’m being a bit harsh on those who want to reduce bounce rates.
Let’s get one thing clear; reducing bounce rate is generally a good idea, but it all depends on the context and purpose.
Let’s first ask why we want to reduce bounce rate and what impact that will actually have on the bottom line.
Reducing bounce rate for the sake of it is never a good idea. We should look to reduce bounce rate if it’s an indicator of poor user experience or low conversion rates.
Going back to the example of my research for this article, there are all sorts of things that these sites could do (and do do) to encourage further action. Pop-ups, flashy boxes, big banners.
Most do it subtly without impacting the quality of the content, but others try and stuff as many CRO gimmicks into their content as possible. That’s just a massive put-off, and it’s not going to make me want to come back to the site again.
Some sites even artificially manufacture another click to bring bounce rate down by getting you to enter an email address to read further or including content in an iframe.
Then there’s the old 0% bounce rate trick – putting duplicate Google analytics tracking codes on your site so a double session is recorded.
Ok, so it brings bounce rate down, but does it keep the user coming back for more in the future? Does it lead to sustainable sales over the long term? Probably not.
All it does is massage another vanity metric without any tangible results.
In short, when looking to bring bounce rate down, ask yourself first what the purpose of the page is.
Is the page designed to convert and make sales?
Then yes, by all means, you should be looking at reducing bounce rate.
Is the page designed to lead users deeper into your site to pages which convert and make sales?
Yeah, a lower bounce rate is desirable.
Is the page designed to educate and inform, build trust in your brand and authority in your industry?
Well, bounce rate should be quite far down your list of priorities. Do what you can to take users to your money pages in a natural and organic way, but don’t compromise on the quality of your content to do so.
Yep, that’s right. In quite a lot of cases, actually,
A high bounce rate can be an indication that you’ve provided the user with exactly what they were looking for. On the flip side, a low bounce rate can suggest that the user didn’t land on a page that satisfied their query, and they’ve had to go elsewhere on your site to find what they wanted.
Recipe sites are a perfect example of this.
Let’s say I search for a ‘vegetable lasagne recipe’. Among the top search results is a page titled simply ‘lasagne recipe’ which looks good enough, so I click through. But to my disgust, as I land on the page, I find it’s actually a recipe for a meat lasagne.
I have to click through to another section of the site to find the page I was really looking for, wasting my precious time and browsing energy. And I’m hungry. Infuriating.
Perversely, if I had landed directly on the veggie lasagne recipe page, followed the instructions, popped my delicious lasagne in the oven and closed the tab, Google would have recorded a bounce.
In this case:
Google says it isn’t. But as any tired and frazzled SEO trying to do things ‘the right way’ knows, you take anything Google says with a pinch of salt.
While user behaviour undoubtedly plays a role in Google’s understanding of web pages, there’s limited evidence to suggest that bounce rate itself is a solid ranking factor.
Let’s consider the fact that Google is only able to record these user metrics if the site has integrated its analytics tracking code. BuiltWith estimates that about 60% of sites in the UK use Google Analytics, so if bounce rate was a ranking factor then Google wouldn’t be able to factor it into their algorithms on nearly half of all sites in its index.
There probably is a close correlation between low bounce rate and high rankings in certain industries and for certain pages, but as we often find out the hard way, correlation doesn’t always mean causation.
I’ve seen pages with a bounce rate of over 90% rank well for high-volume keywords before.
It all depends on whether that page is providing the user with what they were looking for.