Design of Experiments

All our ideas for new products, services, features, and indeed any changes we can introduce to make things better (more growth, increased revenue, enhanced experience, and so on) start off as a hypothesis or an assumption. In a traditional approach to planning, a team may place bets on which experiment to run based on some form of return on investment-style analysis, while making further assumptions in the process.

Design of Experiments is an alternative to this approach, in which we try to validate as many of the important ideas/hypotheses/assumptions we are making as early as possible. Some of those objects of the experiments we may want to keep open until we get some real-world proof, which can be done through some of the advanced deployment capability (such as A/B Testing) that we'll explore later in this chapter.

Design of Experiments is a practice we use to turn ideas, hypotheses, or assumptions into concrete, well-defined sets of experiments...