Trust the science or not?

As a manager, you are constantly told to “trust the science” and to rely on scientific evidence to help them take decisions. But evidence-based management is only as good as the evidence given. Consequently, insufficient research can result in misguided decisions and poor outcomes. Like an example? A recent investigation showed that “nudge” interventions perform much worse in the field than academic papers suggest. If we believe this claim of effective intervention (e.g., personalising a reminder to enrol in a savings programme), applying it at scale yields a performance that is six times worse on average.

Sure, the scientific process includes quality control, but there could be many reasons why individual findings might be inaccurate or inapplicable in a specific business context. Think about questionable research practices, mistakes in how data is handled or interpreted, and neglected differences between tests and applications. Unfortunately, these problems are both frequent and difficult to for practitioners/managers to inspect. Then you find yourself in an uncomfortable position: how do you separate the sound, actionable evidence from the faulty, inapplicable rubbish? When can you trust that the evidence is solid enough for managerial action?