How Nike A/B tests their recommendations

Recommendations can be a useful first step in personalizing your users’ experience. Encouraging users to share their interests or favorite activities allows you to build curated lists of items for them.

Using broad categories that already exist for your content or products is a simple place to start. For example, Nike’s app customizes the home screen based on the user’s gender.

Why we think this is a good A/B test:

To try personalizing based on user’s inputs to deliver an app experience based on the categories they’re interested in or simply based on their gender to start. Test the engagement and downstream effects these recommendations to understand whether the presentation of the product categories is relevant to them based on engagement and add to cart events.