We used to run a rotating carousel on the front page of our .com site. Rotating carousels are very useful for solving political disputes about who gets real estate on the front page, but not for much else.

One day I tested to see what percentage of clicks landed on each banner. The results were extremely unbalanced. Over 90% of clicks were on the first banner, then a few percent on the next, and a very small number on the third. Few visitors even saw past the first banner, much less the subsequent ones.

Users come to your site to perform a very specific task. (I suppose you might work at Amazon or something like that where someone could conceivably just be browsing, but for most sites, someone will be there do something in particular.). That includes searching for information.

They develop blindness to anything that isn’t on the way to completing that task. If it’s information, your users are following information scent. If it’s a specific action, they’re looking for whatever pieces they need to be get that task done. “Don’t make me think” is the advice, and it’s really good advice, because users need to be shown very specifically want to do in order to get what they want.

In doing so, what you’ve built and what users see are two completely different things.

For many of your users, the advanced features in your app just don’t exist, because they’re not relevant to the task at hand. Similarly, on your website, some of the calls to action and some of your copy just isn’t there, because it’s ignored or, in the case of banner blindness, not even seen in the first place.

And some of the best A/B testing results I’ve gotten have been the result of simplifying, to remove obstacles to what users are trying to do or learn:

  • Remove text from some onboarding emails, add bulleted lists instead of paragraphs where appropriate, and repeat the call to action: 30% increase in clicks

  • Change “Our software, hosted on the infrastructure of your choice” to “One-click setup”, and remove a diagram: 50% increase in clicks

  • Make the next step on a page sticky, so that it always occupies the same position in the sidebar even as the user scrolls down the article: 500% increase in clicks

  • Remove the menu bar from our landing pages: 20% increase in successful form completions

In example (2), we decided not to talk about a salient feature of the product. That’s OK. It didn’t exist for the user, anyway. They simplified it in their mind. We can re-introduce it later when it’s appropriate for the task at hand.

You have to simplify, or else your user will do it for you. The major risk in letting the user do the work is that they’ll simplify it wrong, because they don’t understand it. In simplifying, they may even end up with nothing.

Instead, do your user the service of removing anything that’s not completely necessary for the task at hand.