A/B testing is a comparative method to determine which design user prefer in terms of usability, aesthetic, or just overall preference. Implementing this type of testing early in the design process quickly can show which option works best for users and guiding designers towards the most successful design option.
Data from A/B testing can be gathered either in an observational setting or through a post-experiment survey. In an observational setting, a user experiences both option A and option B and decide which design they prefer, voicing their opinions and thought process for their choices. A survey targets specific questions that ask about a users comparative experience with option A and option B.