Designing a Causal Test for a New Airline Pricing Engine:
In this case, you will step into the role of decision scientists at a low-cost European airline facing a high-stakes question: does a new pricing engine actually increase total revenue, or does it only appear to? A simple comparison between the old and new systems is not enough, because passenger search behavior, connecting or multi-flight itineraries, and competitor reactions can all cloud the true effect.
Your challenge is to separate signal from noise. You would think through how the airline should collect data, define treatment and comparison groups, and estimate the causal impact of the new engine. Along the way, you will use concepts such as treatment effects, counterfactuals, experimental design, spillovers, and identification assumptions. The case invites you to build and defend a credible approach rather than search for one perfect answer.