To the Manager Who Softened the Finding

You told her it was excellent work.

It was. That part was true. She framed the problem correctly. She gathered the right data. She built the model with appropriate care and presented the uncertainty honestly and delivered a recommendation that was specific and defensible and, as it turns out, correct. By every measure the organization uses to evaluate analytical work, she did her job well.

Then you told her you wanted to talk about how to position this.

We understand why. The recommendation contradicted a direction the organization had already committed to in a previous meeting. The people who would need to act on it had already told their stakeholders something different. The cost of the recommendation being right was a conversation nobody wanted to have, and you were the person who would have had to have it, and the cost of softening the recommendation was much lower and distributed across time in a way that made it hard to see clearly from where you were standing.

So the confidence interval was removed. The framing was adjusted. The recommendation was repositioned. The final deliverable supported the direction that had already been chosen, and it was presented in a meeting she was not invited to, and the people in that meeting received it as confirmation rather than analysis, and the decision was made, and the consequences followed.

The consequences were as she predicted.

You may not have connected those two things. The gap between the recommendation and the outcome was long enough that the chain of causation was easy to miss, and you were busy with other projects by then, and the postmortem, if there was one, looked at execution and adoption and market conditions and not at the brief that was handed to an analyst eight months earlier and the recommendation she returned and the conversation you had with her about positioning.

She left four months after that conversation. You may not have known why. People leave for many reasons and she was professional about it and the exit interview, if she gave one, probably did not name the specific meeting on the specific afternoon in which she understood something about the building she could not unlearn.

Here is what she understood: that the work was not the point. That the work was the ritual, the performance of analysis, the appearance of rigor, and that the actual decision had been made before she started and would have been made without her and her job was not to find the right answer but to produce a document that looked like the right answer had been found. She understood this precisely and completely in the moment you said let’s talk about how to position this, because those words, in that order, mean exactly one thing.

She was good at the work. The discipline lost her.

You did not intend this. We believe that. The specific harm of softening is that it does not require intent. It requires only the reasonable calculation that a hard conversation today costs more than a soft document this afternoon, and that calculation is almost always correct in the short term, and the people who make it are not villains, they are people inside systems that reward the soft document and do not punish, or even notice, the loss of the practitioner who would not produce one.

We are not asking you to be brave every time. We know what the buildings are.

We are asking you to know what you are doing when you do it.

The work was excellent. Tell her that.

Then let the work stand.