Automatic carousels have come under increasing fire this year from this blog, and UX experts like Jacob Nielsen and conversion pros like Peep Laja

The conclusion, which all three of us share, is that unadulterated carousels are bad for business!

In this post I’ll show you three ways to improve the relevancy of content that occupies your carousel ‘hero’ locations and the contribution it makes to your conversion rate.

Carousels say ‘We couldn’t figure out or agree on exactly what you need so we decided to show you a few options instead’.

In my experience there are three main reasons why carousels find their way into production:

  1. Stakeholders can’t agree on the purpose of a page and the content that is best suited to support the most valuable user action on that page.
  2. Stakeholders believe that it doesn’t matter where their content is displayed just so long as it’s there somewhere.
  3. They are an out-dated technological fashion that has been left unchallenged for too long and (wrongly) have become a ‘landing page standard’. 

Your carousel is likely to be hurting, hindering or not contributing to your conversion rate

Qualitative research conducted by Norman Nielsen Group revealed that auto-forwarding carousels present a number of barriers to conversion such as:

  • Poor accessibility. 
  • Banner blindness.
  • Distraction.
  • Lack of control. 
  • Lack of relevancy. 

If you don’t believe these findings then take a look at your analytics data. My guess is you will quickly discover that the vast majority of users click on the first slide (if at all) as described in this report by Erik Runyon.

Ultimately, this means that the vast majority of people ignore your carousel. If they do click on it it’s likely that they will click on the first slide just because it is the first slide, not because it represents a considered selection of all available options. 

This phenomenon is the ‘carousel version’ of what Herbert Simon defined as 'satisficing'. This mashup of ‘to satisfy’ and ‘to suffice’ describes a cognitive process with which we expedite the process of choice by forgoing consideration of all available options and rapidly selecting the first thing we see which looks reasonable.

Start by deciding what the primary goal of the page is

The first step towards rationalizing the contribution your carousel makes to conversions is to decide exactly what your users are trying to achieve on the page and how you can aide them in moving to the next stage in your conversion funnel.

This means you need to understand your users’ problems, needs, barriers and persuaders.  What goals do they have? What questions do they need answered?  What will help them arrive at the decision to take action?

It is likely that you’ll discover there are a range of actions you want your visitors to perform on your landing pages. The trick here is to prioritise all actions based on the impact each action has on your conversion funnel and the value that goal represents to your business.

Using your new conversion hierarchy take a moment to consider how focused the page is on persuading users to perform Action #1.

Three ideas for improving your carousel conversion rates 

Idea #1: test removing it!

Before optimizing something you should first establish a benchmark of current performance. What contribution does the page or piece of content make to your conversion rate at the moment? 

In the case of your carousel, the easiest way to find this out is by running a simple inclusion/exclusion test.

Earlier this year I performed a test like this for a financial services client.  The test proved that removing the carousel increased sales by 23%.

Example exclusion ab test

I concluded that removing the carousel reduced distraction while increasing the visibility and clarity of content that was required to support the core conversion task.

In another experiment (this time for BSkyB) my carousel inclusion/exclusion test proved that the conversion rate was the same whether the carousel was present or not.

This test provided a business case for bold ideas to replace the carousel entirely.

BSkyB Carousel Test

Idea #2: replace your carousel with a targeted static banner

The purpose of using targeting is to improve the relevance of content displayed to individual users.

By making inferences about what a user needs or is likely to be interested in you can improve your chances of increasing conversions by removing generic content and displaying more relevant content instead.

Another Adobe Target client (Tui Travel PLC) uses this technique on the homepage of

When a visitor lands on the site, Adobe Target checks if they have previously viewed the SplashWorld or Holiday Villages pages.  If not, then the hero location displays generic content relating to summer holidays for 2014.

If the behavioural profile on the visitor’s cookie shows they have previously visited pages relating to SplashWorld or Holiday Villages then the homepage displays targeted hero content instead: 

Last touch targeting example

Idea #3: Use imagery to reinforce visual clues

Visual clues include imagery and design elements which aide in communicating meaning, purpose and the hierarchical flow through a digital experience.

They can help answer the following critical questions held by all new visitors:

  • Is this page what I was expecting?
  • Does it look like it will help me attain my goal on this visit?

Imagine you’re searching for a new laptop.  You land on site A and see a carousel displaying deals on printer cartridges, a click and collect promotion and a call to action relating to the latest type of wifi router. 

How sure are you that this site sells laptops? Is it worth spending any more time on this site or should you hit the back button?

Imagine now that you land on site B to continue your search for a new laptop. This time you see a static banner with an image of a laptop (possibly even the one you searched for). 

The products displayed beneath the hero banner are all laptops and the primary call to action label says ‘View all laptop deals’.  How would you feel now?  Would you feel more confident that you’re in the right place to attain your goal?

You can create the experience delivered by site B through the use of targeting.  Here are some sources you can use to infer user intent:

  • Search keywords: you can dynamically assemble a customized experience for visitors in response to specific search terms which lead them to your site.
  • PPC terms: use targeting to display content that relates specifically to each individual ad or ad group.
  • Display and Affiliate ads: use targeting to display content on your landing pages that matches and reinforces the content in specific offsite advertising campaigns.


Unadulterated carousels can be a waste of screen real estate that distracts users from completing the most valuable conversion action on a page.

You will likely improve your conversion rate by displaying imagery, headline copy and call to actions which match the inferred intent of your visitors.

Follow these steps to discover the contribution your carousels are making to your conversion rate:

  1. Review your analytics to see how people are currently interacting with your carousel.
  2. Perform and inclusion/exclusion test to challenge the contribution your carousel makes to your conversion rate. 

Finally, give some consideration to how targeting can be used to improve content relevancy on your landing pages.

Think about the example from First Choice and imagine how content in other areas of your website could be dynamically adjusted to strengthen visual clues, help users attain their goals and ultimately improve your conversion rate. 

Beamish Steam gallopers image credit: ralph&dot via Flickr

Blair Keen

Published 13 May, 2013 by Blair Keen

Blair Keen is Optimisation Manager at Adobe and a contributor to Econsultancy. You can follow Blair on Twitter and Google Plus or connect via LinkedIn

4 more posts from this author

You might be interested in

Comments (7)

Save or Cancel

Pascal Ly

Hey Blair - we used to have carousel on our homepage 2 years ago and it had been taken down because indeed people were mainly clicking on first banner and conversion was poor.
However I felt like the carousel can be perfoming and converting well and also help reduce bounce rate if it's correctly set up.
What I mean by correctly set up is basically: are we providing content relevant to the audience base on their maturity level in the funnel.

So I asked to have the carousel introduced again on our homepage with the following principle:
1. Clear offer
2. No more than 1 offer per category(product, event, customer reference, etc...).
3. Content relevant to audience
4. Auto play of the carousel
5. Attractive design
6. No more than 5 slides in the carousel

Guess what? It's the 3rd slide in the carousel performing the best, followed by all 1st,5th,2nd slide all performing equally, with only the 4th one really behind in performance.

2 lessons learned here:
1. We now know event better what our prospect are looking after on our site
2. Build the content around your user's expectations

Our next step:
--> figure out if 5 slide in the carousel is too much. We are going to test 4 slide and see how it performs.

about 5 years ago

Blair Keen

Blair Keen, Optimisation Manager at Adobe

Hi Pascal,

Thank you for taking the time to leave such a good response.

I agree with you. The best way to increase conversions is to provide content that meets your users expectations and actually helps them attain their goal.

So actually I think you and I are aligned: only use a carousel if you are sure that the content contained on each slide will be relevant to each individual visitor.

For anyone else reading this thread, I would encourage you to visitor the carousel Pascal is describing

Take note of the 3 static calls to action which appear consistently on each of the 5 carousel slides.

This is an important feature which you don't often see with 'standard carousels'. It increases the likelihood that this particular carousel will generate a reasonable conversion rate.

Pascal - as you're using Adobe Target on your homepage I'd love to continue the conversation with you offline to talk more about your carousel tests and your audience segmentation/profiling.


about 5 years ago

Albie Attias

Albie Attias, Ecommerce Director at King of Servers Ltd

You make some valid points Blair. Unadulterated, poorly implemented carousels are clearly a bad idea but there's a place for a well designed, user friendly carousel too I think. They seem to work for John Lewis for example who've used them for ages.

Design is key - don't have it set to auto scroll at ridiculous speeds, replace those tiny, meaningless 'frame advance' buttons with something more user friendly and perhaps most importantly, don't make the content look like a banner to hopefully avoid the classic 'banner blindness' syndrome.

about 5 years ago


Deri Jones, CEO at SciVisum Ltd


I'm confused - you say you use content relevant to audience maturity in funnel; - so there are many different carousels you're showing.

But then you say No3 performs the best.
So, no matter which set of carousel mages being shown, No 3 was the best?

Are you saying that there is something good in general about being the 3rd slide: maybe that viewers have digested the main page, and come back to the top and by then it's showing slide 3 ?

Or that, in all your carousel versions, it happened that slide 3 was the best of the 5 ? That coincidentally slide 3s all had the best attractiveness, or best offer, or best call to action or something?

about 5 years ago


Deri Jones, CEO at SciVisum Ltd


> Take note of the 3 static calls to action which appear consistently on each of the 5 carousel slides.

Content that is static is not really 'carousel' anymore, is it?

The UI has laid out those static calls to action 'as if they are in the carousel; but could equally have been a static box to the right of the carousel.

Albie, I think that is a fundamental issue you've mentioned there.
> don't make the content look like a banner to hopefully avoid the classic 'banner blindness' syndrome.

about 5 years ago


Pascal Ly

Hi Blair,

Thanks for your reply, and happy to discuss further offline.


about 5 years ago


Pascal Ly

Hi Deri,

What it says is that in our context, the content on carousel number 3 was the one attracting the most number of interest. Which proofs that people do not just click on the first banner but will be looking for content relevant to them.

Just to give a little bit more details, we had few weeks ago swaped the positioning of our offer of slide number 3 to be in 4th position, and currently we are seeing that having swaped the positioning doesn't effect the fact that people are still clicking on that slide the most.

If you go to our current site, it's the banner saying "Grow your business with the World's #1 Sales App" - it's in 4th position, while it was in 3rd position at the launch few month ago.

A thing to remember: test, test and retest to see what is attracting your audience.

about 5 years ago

Save or Cancel

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Digital Pulse newsletter. You will receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.