Synclio
Look Bigger & Handle CallsMore Efficiently
If you’ve just opened a Google Analytics account and eagerly got everything set up only to find a 100% bounce rate figure staring you in the face, don’t toss in the gloves just yet. In reality, bounce rate (at least as reported in Google Analytics) is not quite as straight cut as, say, French Fries – but let’s not get ahead of ourselves...
According to Google Analytics:
“Bounce rate is the percentage of single-page visits or visits in which the person left your site from the entrance (landing) page.”
Google goes on to say that you should use this metric to measure visit quality – but this begs the question: do single-page visits always necessarily indicate poor quality?
The reasoning behind Google’s definition is that if a visitor landed on your page and then jumped back without engaging any further (filling in a web-form, exploring other pages, etc), it’s most likely because he didn’t find the page relevant to his query.
But there’s something missing from this assumption...
Not all web pages are the same.
In the case of a landing page that offers a free product trial, for instance, Google’s estimation may very well be sound: the goal here is conversion, and if the visitor clicked on the ‘Back’ button instead of filling in the web-form, he most likely wasn’t interested in what you had to offer. So the bounce – or single-page visit – may indeed indicate a problematic area to explore (perhaps your keywords are misleading, or your offer isn’t compelling enough, or your design is too confusing, etc).
However, consider a blog for instance: an internet user searches for some specific item of information, comes across your site, and clicks through. He reads the article, finds the answer he was looking for, and carries on with his day.
As a blogger (or any other online content provider), your goal is to be read and (preferably) to establish repeat readership. The fact that the visitor left your site after reading only the one page he landed on doesn’t necessarily indicate that he didn’t find your content relevant—to the contrary, it might be that your page answered his question so completely that he had no need of searching any further.
Some may argue that if he did like your page he would have proceeded to explore other pages, but – while it's certainly a reasonable proposition– in reality not all users have the time for random browsing. Many of us often seek to answer a specific question or looking for some specific item of information. In either case, we might find what we need without having the time (or desire) to engage any further – at least not during that same visit.
But we might still return at a later date, in which case both your goals will have been accomplished.
So with Google’s definition of Bounce Rate (and the same pretty much applies to other web analytics vendors across the board), the metric should be understood to indicate the number of single-page visits rather than the number of visitors who didn’t care for your page.
That's an important distinction.
Perhaps a more accurate definition – at least where pages that don’t aim to get an immediate conversion are concerned – would take into account the amount of time a visitor has spent on the page in question. That’s assuming Bounce Rate as a concept attempts to measure lack of relevance or quality.
Unfortunately, as aptly explained by Analytics expert Avinash Kaushik, Time on Page and Time on Site metrics give only a crude picture at best of what happened in reality. As it relates to our discussion, when it comes to single-page visits (or bounces), analytics registers a grand total of 0 seconds regardless of how much time the visitor actually spent on your page – that’s right, ZERO, zip, nada!
But what if you wished to find out not only how many single-page visits you got (which may not concern you as much if you run a one-page website or a blog), but how much time the visitor really spent consuming the content on your page (which would be of the utmost concern)?
Fortunately, where there’s a will, there’s a way – and where there’s a crack, there’s usually a hack!
While there may not be an official way to do this, Claudiu Murariu of PadiCode has posted a short snippet of code that utilizes Event Tracking to create an event every time a visitor stays on your page past a set period of time. This in effect accomplishes the very thing, as you'll learn how many visitors dropped out before they could have reasonably consumed your content.
Generally less than 5 seconds is considered a bona-fide bounce (i.e. the page was completely irrelevant), though you can set it higher if you wish to get a sense not only of the perceived relevance, but also the perceived quality and value of your content.
When looking at your page's stats under the Content section, Bounce Rate is grouped right along with Pageviews and Unique Page Views – but again, this can be misleading. You might think the bounce rate percentage figure is calculated based on the number of pageviews or of unique page views – but neither would be correct...
As explained by Claudiu Murariu, the bounce rate is calculated based on visitors that landed on a page (for example, by following a link from a search engine or from a referring site). However, it does not take into account visitors that came from another page on your site.
So let’s assume your dashboard shows 300 pageviews, 200 unique page views, and a bounce rate of 50% – you might take this to mean that 150 (or at least 100, if going by unique views) of your visitors bounced. However, if you check how many Entrances you got for the page under the Top Landing Pages tab, you may find only 80 or so. So in reality, 40 visitors bounced.
While this doesn’t necessarily alter the significance, it does change the perspective. And at any rate, it’s a matter of understanding what you’re looking at. (As a case in point, when the folks at PitStop Media ran a one-question survey to see how many people understood this fact, only 23% of responds got it right!)
Wrong – sorry... Google needs measurable parameters to evaluate the perceived ‘quality’ of a page, and Bounce Rate happens to be one of them. So having high bounce rate might eventually affect your search engine ranking. No one knows exactly how much it factors in the overall computation, and hopefully your other on page and off-page SEO efforts will offset its impact, but that’s certainly something to be aware of.
Moreover, just because single-page visits aren’t necessarily a bad thing doesn’t automatically make them ideal. You do want your content to encourage further engagement, and if visitors consistently fail to explore more of your site, there may very well be something lacking. Look under Top Landing Pages and consider the ratio of Bounces to Entrances for individual pages. If certain pages consistently get a high ratio, you'd want to revisit them with a scrupulous eye.
Good luck!
posted by Maty Grosman @ 8:37 AM