Do users really interact with carousels?
Carousels, rotating banners, sliders – whatever you call them, in web design they’re often one of the first areas of a website’s homepage. But how effective are they? Do they increase engagement or do they just give developers an unjustifiable headache? Over the past 7 years, our team have built over a thousand WordPress websites and over 90% of those include some type of carousel or slider. In earlier years, it used to be a key feature in a website which often was used to up-sell to clients. Some may argue that it’s one of the most effective ways of promoting news, updates and services or offers, but do they really work? Erik Runyon completed some research to find out if users actually interact with carousels. He added tracking facilities to the main feature on ND.edu as well as four other websites with carousels, three of which were static and another one which automatically rotated the features of the website.
The experiment focused on three main areas to identify engagement:
- The number of times which the feature was switched by users
- The total amount of clicks on the feature
- The total amount of clicks per slide position
The experiment also included further interactions such as how many times the feature was switched using the different pagination options (dots and arrows), but that was not the focus.
The experiment was run over approximately three months from 15th October 2012 to 22nd January 2013. The percentages are based on the number of clicks recorded through Google Analytics Events and compared to the number of visitors to the homepage. “Feature” refers to the individual calls-to-action that are either manually or automatically rotated in and out of view.
Approximately 1% of visitors click on a feature. There was a total of 28,928 clicks on features for this time period. The feature was manually “switched/rotated” a total of 315,665 times. Of these clicks, 84% were on stories in position 1 with the rest split fairly evenly between the other four (~4% each).
The same tracking was applied to three other websites which had varying amounts of traffic and features. The average amount of feature clicks were between 1.7% and 2.3%, which were higher than the ND.edu example. This is likely down to the fact that the websites were in different niches and had different target audiences. The ND.edu website had very general traffic and the others were aimed at specific demographics. The main area of these sites averages around 48-62% of the total clicks, with the other features splitting the traffic distribution evenly. The website which showed the most balanced distribution of traffic was interestingly the one with the least features (3 in total).
This then poses the next question, is there any change in interaction if the carousel is automatically rotating?
From the experiment, there was only one website which had an auto-rotate function. This site has the highest number of clicks with the homepage boasting an 8.8% click-through rate from a visitor clicking on a slider. The first slide feature averaged just over 40%. Unsurprisingly, the click through rate of the other features reduced heavily with the second slide’s click through rate being 18% and the last slide being 11%.
It’s plausible to say that the first slide is always going to generate a higher click through rate than the subsequent slides. But what does this mean for websites designed in 2018 and how should you advise clients? Firstly, it would be strong advice to avoid a carousel – they’re often harder to respond and can often deliver unexpected results on older devices and browsers. As we move through 2018, Google has also made it quite clear that focus should be on speed and boosting user experience through ease of use. Carousels definitely don’t fit into that plan. It’s likely that many customers will still have their own judgements on the effectiveness of carousels and sliders on websites and as such as website designers, we’d always advise to keep the number of slides to a minimum and ensure that they include valuable, compelling content.
The final suggestion would be to take into consideration the niche which the website is in. If you look at the top five feature stories based on click-throughs for ND.edu during the experiment, the difference between feature one is 6,331 clicks vs. 1,509 for the fifth feature. ECommerce website design is a separate domain entirely as visitors will often be looking for exclusive discounts and offers which are traditionally found in carousels on the homepage.
What should you replace them with?
The big question – if you’re going to take the plunge and remove or discontinue using carousels, what should you place there instead? The answer to this question largely varies depending on the type of industry which the website is in. The simple solution would be to include a single static banner which has a single feature and perhaps some smaller features underneath. This allows the visitor to view the information they’re looking for much quicker and also will reduce page load times. For eCommerce websites, a popular choice is now to include a Load More option with ajax functionality so there is no delay in loading.