I tracked the same interactions for three other Notre Dame sites, each with varying degrees of traffic and number of features in the carousel. The average number of feature clicks were between 1.7 and 2.3% which is higher than ND.edu and to be expected due to the different natures of the sites (ND.edu is more general traffic, these sites are specific audiences). The top slot on these sites averaged 48-62% of the total clicks, with subsequent features splitting the rest fairly evenly. The site with the most evenly distributed percentages based on position was the one with the fewest number of features (3).
We only have one site that automatically switches the feature (see Nielsen’s warning against Auto-Forwarding Carousels). This site averaged the highest number of clicks with 8.8% of homepage visitors clicking a feature. The first feature averaged 40%. The click-through percentage for subsequent features steadily declined for each feature starting with 18% for the second slot down to 11% for the last.
My question exactly. It’s safe to say that whatever is in the first slot is going to draw the most attention (naturally). But what do these numbers give us that we can take back to our clients?
First, if they’re going to insist on a carousel, they need to include compelling content that not only entices users to click, but can get their attention in the first place. Second, I might suggest keeping the number of features to a maximum of four (or better yet, three), as it appears that as the number of features increases, the click-throughs on sub-features decreases dramatically.
Finally, I’d suggest that the subject matter can make a big difference. For instance, when looking at the top five feature stories based on click-throughs for ND.edu in the time-frame mentioned above, the difference is 6,331 clicks for number one vs. 1,509 for number five. And in the case of ND.edu, the top three are sports related, and the next two are from our “Fighting For” series (which were advertised during football games).
If anyone has been tracking similar interactions on carousels, I’d be interested to know how your numbers compare.
I think perhaps we should all just take @lukew’s advice.
Brad Frost posted a follow-up with his thoughts on carousels.
I posted some updated numbers from the first six months of 2013.
I have been notified of other articles pertaining to carousels. Contact me if you find others whether they’re for or against. I’ll include them here.