Many thanks for your thoughtful response, Starling.
This chasm between in vitro and in vivo is indeed one of the biggest challenges in studying behaviour. In the real world it’s much harder to control for the many factors that might also influence behaviour. It’s tough enough to even figure out what it is you’d need to control for. (Arguably this kind of issue is implicated in the replication crisis that has been affecting a sizeable chunk of psychological research.)
In a sense this is at the same time a source of frustration, and a source of fascination. Experimenting on small populations in organizations is inevitably very limited compared to large-scale consumer or well-controlled lab studies. but it means you have to be inventive and creative to find the evidence that an intervention works, and that can be seriously cool!