Some historyHITS was the number of times a web server was hit, but if one page had, say, 8 images on it, the web server would receive 8 hits. Not such a useful metric if you want to track commercial success. You could manufacture success by dumping a load more images on a page, and lo and behold you had more hits. That’s good, right? No. You just had more images, not more people. HITS = How Idiots Track Success. I’d contend that Hits has been replaced as a mis-used and mis-quoted measure by Bounce Rate. More times than I can recall, I’ve heard references to a site-wide bounce rate being “bad”, or “worse than normal”. What does that mean, and should you be worried? And what should you think about the person saying it?
Bounce RateLet’s look at Bounce Rate. It’s a frequently cited measure of the combined effectiveness of Marketing and e-Commerce. As a start point, what is the standard definition of the Google Analytics Bounce Rate:
A bounce is a single-page session on your site. In Analytics, a bounce is calculated specifically as a session that triggers only a single request to the Analytics server, such as when a user opens a single page on your site and then exits without triggering any other requests to the Analytics server during that session. Bounce rate is single-page sessions divided by all sessions, or the percentage of all sessions on your site in which users viewed only a single page and triggered only a single request to the Analytics server. These single-page sessions have a session duration of 0 seconds since there are no subsequent hits after the first one that would let Analytics calculate the length of the session.A high bounce rate is a bad thing. Your traffic is untargeted. Your content is poor. You’re wasting your budget. Your site is failing to meet the needs of your users. Not so fast. Let’s look at a few scenarios, and examine why a site-wide bounce rate is largely meaningless, and anyone quoting it has a limited understanding of digital.
People use the internet including your website as a phone book. If they arrive at your homepage looking for your phone number, which is displayed prominently, and they then leave, have you failed as a website? You could bury the phone number. That would improve the bounce rate. Maybe not such a good idea.
If you run Display activity, you may well serve millions of impressions, and generate substantial numbers of clicks to your site. Of course you’d like these clicks to convert to a deeper level of engagement, but that’s not always what Display is best at. Depending on your ad, where it is served and the message it delivers, simple brand awareness may be the objective. It’s a given that Display creative should have a strong message, directed to a landing page that invites engagement.
If you have website pages where the only CTA is to call you, removing these pages will certainly reduce the bounce rate, and with it any down-stream revenue you might derive via the telephone.
If you run Social activities, with an intention of awareness driving, you may well see a high bounce rate. Mis-directed clicks are not great, but if you’re paying on a Cost Per Mille basis, it’s not the end of the world. You might have competition Ts and Cs attached to a competition where the entry mechanic is not on your site. This would have a high bounce rate.
If you have a resource centre (link bait, or SEO fodder), such as FAQs, you may attract traffic from people who have searched for an answer to a specific question. Having find their answer, they may well leave, with your brand in mind. They may well return later. Informational pages will almost always have a Call To Action into your e-Commerce funnel, but direct revenue is not their core purpose. This is true for many other pages such as “Contact Us”, “About Us” etc.
If you bid on a high volume, high cost generic term, such as “car insurance”, you will certainly tailor your PPC activity to reduce irrelevant clicks to every extent possible, to make it clear who you are, and what you offer, but it costs the user nothing to click on your £20 per click PPC ad, just to find you’re not the company for them. The bounce rate for “car insurance” will always exceed a phrase such as “young driver insurance”. Do you work hard to optimise the “car insurance” PPC ad, or switch off PPC for this term in order to reduce your bounce rate, and with it the income?
Bounce Rate is absolutely an important metric to be aware of and track, but it needs context, and can only be used in conjunction with other metrics such as time on site, traffic source and page. There can be no complacency when a PPC click costs £20, but digital requires an holistic view, including Attribution modelling.
What is good?
So, what is a “good” site-wide bounce rate? 0%? 10%? 25%? 50%
The answer lies in your marketing and website objectives. I would contend that there is no “good” measure. It’s about Objectives and Context.
Before activating any marketing activities or building a website, consider your commercial objectives and the behaviours that you wish/ expect to see. It’s easy to review data after the event, and to interpret it in a way that supports your “hope”.
Use Bounce Rate in conjunction with other metrics that provide a more balanced view.
Consider structuring your website and/ or analytics in a way that will allow you to analyse your traffic by objective (Informational, Transactional etc.)
Consider the creation of additional, campaign or channel specific pages.
Ensure that all Digital acquisition activities are tagged up in a way that will allow you to segment your reporting.
As with all web analytics, look at trends in behaviour. Focus this analysis where your spend is directed.
Not so long ago, I was in a meeting with a supplier who cited my site-wide bounce rate as being a significant cause for worry, and higher than average. This is a headline grabbing statement that can (and does) get attention at a senior level.
If you meet a supplier who makes an un-supported statement such as this, make your loved ones safe, keep your eye on your possessions and look for another supplier. Or ask them to explain and substantiate their reasoning for the statement.