Insights
Nov 19, 20182 min read

Where should you conduct your A/B-tests?

by Klarna.com

Calls to Action

Are your calls to action prominent enough? Do they motivate customers, or do your customers find them to be too tacky? Test different wordings, presentations and placements until you figure out what works for your target audience. Note that what works for one type of customer may not work for another, so you cannot rely on data collected from one website as a guide for how you will arrange a different website.

Pricing

Do your users prefer having shipping and handling costs included in the price or being presented with them at the end? With A/B testing you can find out for sure by showing some customers the all-in price and others the base price. You may be surprised at the results. Experiment with different kinds of discounts and special offers too. Don’t make the mistake of assuming that lower prices are always better – some consumers prefer to pay a little more, because it increases the perceived value of the product.

Copywriting

Is your copy compelling enough? Does your eCommerce content really sell, or does it bore readers to the point that they gloss over it? Are your headlines clear, and are you using the best possible fonts, font sizes and colours? Are your product descriptions easy to understand and enticing enough to convert visitors into customers? Offer multiple landing pages with different kinds of copy, and see how your customers respond.

Images

Do your images load quickly? Are they eye-catching enough? Do your users actually stop to look at them, or do they scroll right past them (heatmaps can help you to determine this)? Are your users conflating your images with navigation options? Try out some new images, and see how your users react.

Forms

Are your forms working as well as they could be? Are you losing people because they are too long-winded or confusing? Your analytics tool should be able to help you here. Refine your forms until they are as easy as possible for your users to complete.
Do not make the mistake of running a test and then forgetting about testing for a long time. You should test your website regularly. Ideally, after finding the ‘optimal’ layout and content you should start the next round of tests to see if you can make even more incremental improvements. Monitor your analytics on an ongoing basis, and keep checking to see what sort of improvements you could make. Your competitors are always looking to move forward, so you should be too.