I think the research is great and it's really helpful having more data to use but I feel the article direction has been let down somewhat by the usually excellent Smashing Mag team here because to my mind comparing a "marketing carousel" and a "product image gallery carousel" are incredibly different.
I'd expect low interaction as mentioned in the article and also the Notre Dame research that marketing carousels don't do that well and to be honest I'm completely unsurprised that photo galleries have vastly higher interaction stats.
The data and research are great but I felt the comparison of what I'd consider two completely different types of thing is a bit misleading in places.
Came to say the same thing. It's the "marketing" carousels that don't work - the ones that you see on the home page, not product image galleries.
I wanted to explore whether the underlying pattern was broken or if it was just a problem with marketing banner carousels. I think what this research shows is that the pattern itself is fine — and maybe everyone knew that already. That said, I think marketing carousels can work. They just need to be designed well for their goal.
Thanks for reading! I think you're right — it's hard to compare the two directly. My goal was to look at the carousel pattern itself to see if it was ineffective. What I found was that there is nothing wrong with the pattern itself — which was surprising seeing how in every article decrying carousels, they attack the whole pattern.
The language is never couched in "marketing banner carousels" versus "product image gallery carousels". It's just "carousels". I think this research shows that "carousels" can actually be really effective and we can use that knowledge to spend the time designing more effective marketing banner carousels instead of throwing the pattern out wholesale.
Hey guys, I wrote this article — it was a long time coming. We end up using carousels a fair bit in different situations so I wanted to see how they were doing. I'd love to hear your thoughts — good or bad!
Hi Kyle, thanks for the well thought-out article!
Auto advancement is bad. If you have a carousel, do not auto-advance
I think I assumed auto advancement was what made a carousel a carousel (hence the name). In your research, is there a difference between a carousel and an "image gallery"?
we only use carousels in one place on our websites: as an image gallery on product detail pages
And good thing, too. Most of the anti-carousel writing seems intent on preventing stakeholders from trying to stuff as much pet content "above the fold" as possible. So just to clarify, this article is not saying homepage carousels are a good idea :)
Thanks for the feedback!
In your research, is there a difference between a carousel and an "image gallery"
In most of the research I did before writing, a carousel was referred to as something similar to the Kodak Slide Carousel. Usually interchangeable with the term slider. I think probably that auto-advancement has been compounded with carousels these days but it's not a necessary feature to be called a "carousel".
So just to clarify, this article is not saying homepage carousels are a good idea :)
I'm also not saying it's a bad idea. I think homepage carousels could probably be very effective if people spent the time to try to design them well. A good example of that is the Amazon marketing banner carousel I linked to in the Considerations section of the article. It very clearly sells the user on why they should click on the next slide by telling them exactly what will be on it.
Every article out there that was against carousels said that "Carousels are bad". My goal with this article was to say: "Hey! Carousels aren't bad. Just some implementations of them are. They can actually be really engaging if you design them well!"