The Values Exercise That Made the Best People Leave Faster
%201.avif)

The workshop was two days. A facilitator, a set of prompts, dot voting on words. By the end of Friday afternoon they had six values. Candor. Ownership. Curiosity. Respect. Craft. Speed.
The company printed them on the walls of the main office. The CEO put them on the second slide of every all-hands. Every offer letter now included them.
Within four months the company had lost its Head of Product, its VP of Engineering, and its Head of Design. All three of them had voted for the values on Friday afternoon. All three had left because of them.
What a values exercise does to the room
The workshop had done a specific thing. It had made every person in the company articulate a picture of the company they wanted to work for. It had asked them to voice the picture in front of their peers. It had led them to a shared vocabulary for that picture.
This is normally described as alignment. What it actually is, neurologically, is priming. Once a person has publicly named what a healthy company looks like, their nervous system starts running a comparison every day between the picture and the reality. The gap becomes visible in a way it was not visible before.
If the reality moves toward the picture, the priming builds loyalty. If the reality does not move, the priming builds resentment. There is no middle path. The values on the wall become a daily inventory of what the company is failing to deliver.
The three senior leaders who resigned had all voted for candor. They had also all been in the same meeting the week before Friday's workshop when the CEO had shut down a hard question from the CFO. The CFO had absorbed it. The three of them had watched. The word "candor" on the wall a week later did not feel like a value. It felt like a lie the wall was telling on behalf of the CEO.
How much is one unresolved conflict really costing your company?
What the exercise should have done first
The exercise should have started with the room the CEO was running, not with the words the company wanted on the wall.
Values that get named without the behavior underneath them will drive out the exact people whose belief in the company was highest, because those people are the most sensitive to the gap. The lower-belief employees will be fine. They never expected the wall to mean anything. Their nervous systems are calibrated to ignore corporate language.
The high-belief employees are the ones who read the words and expected them to be true. When the CEO's behavior contradicts them the following Tuesday, the amygdala registers it as a specific violation. The person feels foolish for having believed. They begin, quietly, to update their view of the company.
If a leadership team wants to run a values exercise safely, it has to install the behavior before it prints the words. The behavior is not a slogan. It is a specific move that leaders make when the room becomes hard. Doug Noll's work centers on one such move, affect labeling, which is the practice of naming an emotion in the room accurately before responding to it. That move produces candor. The word "candor" does not.
Doug Noll's new book Empathy Leadership: The Powerful Skill That Drives Winning Results lays out the behaviors leaders install before they print the words on the wall. See it here.
What the company did the second time
After the three resignations, the CEO did not run another values exercise. He did something quieter.
He spent six months practicing one thing in every leadership meeting. When someone raised a hard concern, he paused for two seconds. Then he named what he thought that person was feeling. Then he waited for a correction. Then he responded to the content.
Nobody wrote a value on the wall about it. Nobody made a slide. The behavior spread anyway, because his direct reports started doing it in their own teams within three months, and their teams did it within six.
Two years later the company had a culture people used the word candor to describe. The word had never been printed. The behavior had been practiced.


