The surprising power of online experiments: getting the most out of A/B and other controlled tests

The authors posit that too many firms rely on subjective opinions when making key decisions, from marketing campaigns to new product features, and therefore place large investments at risk. They emphasize the importance of using A-B testing to shift decision making to hard data. A-B tests utilize two experiences, the control and the modification, to draw out key metrics that can be compared. Experimentation structure can be organized as a centralized model, a decentralized model, or a center-of-excellence model which focuses on design, process, and analysis. Firms should be wary of low-quality data and avoid assumptions regarding causality.