In Berne, can feedback distinguish progress from chance?
A manager intervenes after an unusually poor result, then observes a return to a more ordinary level. The managerial question is precise: how can this improvement be prevented from becoming supposed proof that the feedback worked?
What the literature establishes
Daniel Kahneman and Gary Klein (2009, American Psychologist) identify two important conditions for intuitive expertise: a sufficiently predictable environment and opportunities to learn its regularities. Their analysis highlights, in particular, the quality and speed of the feedback available. For managers, observing outcomes is therefore not enough: those outcomes must actually provide information about the quality of the work.
The mechanism: learning a relationship that does not exist
Hillel J. Einhorn and Robin M. Hogarth (1978, Psychological Review) analyse how confidence in judgment can persist despite inadequate validity, particularly when the information needed to assess it is incomplete. A related trap concerns interventions triggered by an extreme result: when outcomes contain a random component, the next observation may be less extreme without any effective intervention. This regression towards the mean can lead managers to overestimate the usefulness of their criticism.
The premature conclusion
The tempting conclusion would be that feedback is pointless when outcomes fluctuate. These papers instead invite a distinction between an observable correction in the work and a change in the overall result. Checking that a verification step is now being completed is not the same as demonstrating that feedback alone explains a reduction in incidents. (our executive and employee training programmes)
What the evidence cannot resolve
These two articles offer theoretical and integrative analyses; they are not a trial of feedback in organisations in Berne. In practice, case difficulty, workload and procedural changes can shift alongside outcomes. A before-and-after comparison, especially one starting with an exceptional incident, cannot isolate the effect of the conversation.
A practical check in Berne
Within SHR — Swiss Human Resources' programme « L'art du feedback », an exercise applicable in Berne is to define a verifiable behaviour before the conversation: a file check in the federal administration, protocol compliance in public health, a telecom ticket check or a component inspection in precision manufacturing. Over a period defined in advance, record whether it occurs at every comparable opportunity, before and after the conversation, while logging outcomes and contextual changes separately. The measure is the proportion of opportunities in which the expected behaviour actually occurs, verified against authorised work records rather than the manager's recollection. It documents a change in practice without turning a simple rebound in outcomes into causal proof. To go further: explore the The Art of Feedback training in Bern in the canton of Bern, or browse our executive and employee training programmes in Switzerland.
In pictures: The Art of Feedback in Bern
- the art of feedback
- Bern
- research
- training bern