This chapter introduces A/B testing as the simplest yet powerful experimental design. It explains how comparing two conditions—control (A) and treatment (B)—can isolate the impact of a single architectural variable. Examples, such as testing different lighting schemes or signage, demonstrate how random assignment ensures unbiased results. The chapter covers statistical underpinnings like hypothesis testing and average treatment effects, showing how data-driven decisions emerge from clear comparisons. Practical guidance addresses challenges unique to architecture, such as the permanence of built interventions and adaptation over time. By leveraging A/B testing, architects can refine designs based on measurable outcomes rather than intuition alone.