Sparked by Claire Alvis’s essay about a sister whose childhood talent for drawing was steered by the adults around her into a career as an architect, this conversation looks at how the same dynamic plays out at work.
We explore the “verdicts” that people make all the time, especially at work: the judgments that leaders form early, often on thin evidence, and then unconsciously act on until they become self-fulfilling. Things like “Oh you know Mark, he’s no good with numbers”.
Dennis opened with a confession. Years ago he decided a new hire, “Jeff”, wasn’t creative because he stayed quiet in a few brainstorming meetings.
Having reached that verdict, Dennis started working around him, leaving him out and leaning on his team, gathering only the evidence that confirmed the original judgment.
Jeff eventually left. With hindsight Dennis asked himself: did I discover a weakness, or help create one?
The central idea is what Dennis calls the signal gap. We transmit two things at once: our words, and a second, powerful signal made up of our behaviour, mood and past treatment of a person. When the two conflict, people believe the behaviour every time.
Drawing on Nobel Laureate Daniel Kahneman's System 1 and System 2, Dennis notes that most of these verdicts are formed as effortless snap judgments and rarely revisited, which makes them both unintentional and durable.
The conversation then ranged into related territory: Carol Dweck on praising effort rather than the person (and Dennis’s point that even the “right” praise fails if your subsequent behaviour contradicts it), giving genuine ownership through the Prince2 idea of managing by exception and the recurring theme that miscommunication usually comes from failing to speak the other person’s language.
In our conversation, we also acknowledged that sometimes a snap judgment is simply correct, and that the appropriate response is a fair assessment, training, and occasionally the kindness of parting ways if the fit between person and role really is as poor as it looks.
Dennis closed with four tests for whether a verdict about someone is fair:
Does it have a recent date?
Does it rest on evidence you’ve actually gathered rather than just your opinion?
Does the person know the verdict exists?
Have you told them clearly how they could change it?
His practical suggestion is to pick one person, examine the verdict you hold about them, and run it through those four questions. The uncomfortable part is actually acknowledging we might have been wrong about them the whole time.
Enjoyed this piece? Become a paid member and get my Consulting with AI prompt library, the full Asking Questions Like a Pro course ($199 standalone), practical templates from a decade of fieldwork, and discounts on many other resources. 🔗 See everything included in the paid tier.
