Your company has values, and they are probably good ones. Ownership. Customer first. Carefully worded, properly launched, and everyone in the building can recite them, most of them in order.
Then evaluation season arrives, and something interesting happens.
Nothing.
The review form asks about tasks. What was delivered, whether it worked, how it went. The values play no part in any of it. They stay on the wall, exactly where they have always been.
Everyone quietly understands the arrangement. The values are for speeches. The form is for careers.
So the words become background music. Pleasant, familiar, and disconnected from anything the organization actually measures.
The wall is not the problem
When leaders notice the gap, the conversation turns to people. They don’t really live the values. The culture is weak. Maybe another workshop would help.
I don’t buy that story.
People are reading the setup correctly. The setup says: recite these words, then go get scored on your task list. Anyone paying attention concludes that the words are ceremony.
That is not cynicism. That is comprehension.
Here is the test I actually trust. The test of a value is not whether people can recite it. The test is whether it can move a score in a review.
If you cannot ask a question about it, it is decoration.
Turn every value into a question
The exercise I run is simple to describe and uncomfortable to do.
Take each stated value and try to turn it into a question about last month. Not about beliefs or intentions. About actual behavior, something observable, with a scoring rubric attached.
Take “ownership.” What would you ask about last month that shows whether someone practiced it? What does a strong answer look like, and a weak one? If you can write that down, you have a question you could put in a review and defend.
Two things happen when a leadership team tries this, and both are useful.
Some values convert cleanly. There is real behavior underneath them, and you can point at last month and score it.
Some values refuse. You turn them every way you can and no question comes out. Nothing observable, nothing you could score, nothing anyone could actually be asked.
That refusal is information.
These are usually the words that were chosen for how they sound, not for any behavior anyone had in mind. The exercise does not kill them. It reveals that they were never alive. They were decoration, and now you know which of your values are which.
Put the question in the review
The lever is one change. The questions that converted go onto the review form, next to the task questions.
Not a poster refresh. Not a louder launch. A question, with a rubric, inside the document that decides how people are evaluated.
And the value starts to steer.
Part of it is consequence. People behave differently once a value can cost them something or earn them something specific. Not because they suddenly believe harder. Because the setup finally points in the same direction as the words.
Part of it is clarity. A question with a rubric shows what the value looks like in practice, at the scale of an ordinary month. Before, people were guessing what it meant. Now they know.
The mood shifts with it. Reciting words nobody is measured on is tiring, and everyone feels the hollowness even when nobody names it. Doing work the words actually describe, and being recognized for it, is a different experience entirely.
Behavior and energy move together here, because they were never separate levers. Both were waiting on the same setup.
So run the test on your own values. Take each one and try to ask a question about last month with it. If the question comes, put it in the review. If it will not come, no rewording will save it and no workshop will either.
Retire it, rebuild it, or admit it is wall art. But stop grading people on tasks while asking them to live by words that cannot touch a score.
None of this required different people. The people were reading the setup correctly all along.
Change the setup, not the people.