“How Did You Reach That Conclusion?” — A Decision or a Need for Evidence?

Recommendation presentation. Zofia shows a slide recommending option B. Instead of accepting the conclusion, Tomasz starts asking how the numbers were calculated.

— I recommend option B. It will improve the result fastest.
— What assumptions did you use?
— What about seasonality? And churn after the first month?

Zofia starts to feel challenged. Tomasz is using questions to decide whether he can trust the recommendation.

What really happened

Zofia presents a recommendation and expects the conversation to move toward a decision. Tomasz starts probing assumptions, seasonality, and data. For him, questions are a route to trust: if the answers hold up, he can support the conclusion with confidence. Zofia hears the same sequence as a test of her competence. The more scrutinized she feels, the easier it is to become defensive. Tomasz may then read that defensiveness as yet another reason to scrutinize. A disagreement about the quality of an argument begins to acquire a personal meaning.

What the research says

Team research distinguishes conflict about the task from conflict about people and relationships. Task conflict can be useful, but it can readily spill into relationship conflict, especially when people do not trust one another or attribute hostile intentions to questions (Simons & Peterson, 2000; Peterson & Behfar, 2003). That maps well onto this scene: “What assumptions did you use?” is, on its face, a question about the model. Zofia may nevertheless hear, “I don’t believe you know what you’re doing.” The problem is therefore not only the question itself, but the meaning assigned to it.

How to handle it

Tomasz can state the purpose before launching into a series of questions: “I want to be able to support this option, so I’m going to test three assumptions.” Zofia can separate herself from the model: “Are you challenging me, or the assumption?” The team can also create an explicit stage in important recommendations for stress-testing the argument, without treating that process as disloyalty. The general rule is: a question can be an attack, but it can also be a way of building trust; state why you are asking before the other person has to invent the reason.

How Empatyzer and Em can help

Empatyzer can help Zofia and Tomasz see that questions do not carry one universal meaning. The assessment and profile comparison may show that Tomasz builds confidence by testing assumptions and precision, while Zofia’s fast, outcome-focused style makes a stream of questions easier to experience as decision blocking or a competence test. Em can prepare Tomasz to state his intent before probing — “I want to support this option, so I’m going to check three assumptions” — and prompt Zofia to separate a question about the model from a question about her ability. If the same tension repeats across the team, group-level insights can support a shared stage for testing arguments. Micro-lessons reinforce the idea that a well-framed question can be a way of building trust, not evidence that trust is missing.