

This is the sixth article in a series on the human capabilities that AI cannot replicate.
The tools available to leaders have never been more powerful. Dashboards, real-time analytics, AI-generated summaries, performance tracking systems. And yet the leaders I sit with are not clearer about what to do. They are more confused.
Why? Statistician and forecaster Nate Silver points to a reason that’s simple but devastating: Most information is noise. And that noise buries the signal: the meaningful patterns worth acting on. The more data available, the easier it becomes to find patterns that aren’t actually there and overlook the ones that matter.
AI amplifies this problem. It can surface more data, faster. It cannot tell you which data matters. That’s why today’s most important leadership challenge is making that distinction and acting on it with enough precision that the noise doesn’t win.
In my work with organizational cultures, I define signal and noise in precise terms.
The signal is always behavioral. It is the set of actions, behavioral norms, traditions, rituals and decisions that define the culture you are building. These inform the choices you make as a leader. Your job is to provide rich soil for those behaviors to take root and spread.
The noise is everything that pulls your attention away from that: gossip, assumptions about who someone is based on how they performed under a previous leader, misaligned employees, and the organizational pressure to act quickly before you have understood what you are actually seeing. That last one is the most dangerous.
Discernment is the discipline to hold that distinction clearly, under pressure, and act on the signal without being pulled off course by the noise.
Discernment is not a philosophical virtue. It is a culture protection strategy. I often see discernment (or its lack) surface in personnel decisions. Leaders must be able to tell the difference between someone who needs more time and someone who needs to leave.
I worked with a CEO, “David,” who was doing everything right. Clear vision. Defined behavioral norms. A new cultural operating system being carefully seeded across a 300-person organization. The work was difficult. But it was working.
Until two senior leaders started pulling against it.
The first was “Maria,” a longtime senior director who had been a top performer in the previous culture. In the new culture, she was quieter. Slower to volunteer. She still defaulted, in pressure moments, to the old patterns David was trying to move the company past. He wasn’t sure how to read her. Was she resisting? Or just adjusting on her own clock?
The second was “Greg,” a VP who was also a long-tenured star under the old culture. Greg’s behavior pattern differed from Maria’s. In team meetings, he subtly dismissed the behavioral standards David was working to establish. In private conversations, he planted seeds of doubt to gain support for keeping the status quo. He was good at looking aligned, but even better at covertly undermining alignment.
Before David, Maria and Greg worked in what I call a culture of personality, shaped entirely around one leader’s preferences and reactions, with no standardized norms or shared behavioral expectations. They had both thrived in it. That was part of what made the situation hard to read.
When I asked David when he had first noticed issues with the two of them, he paused.
“About a year ago,” he said. “Maybe longer.”
By the time we addressed it, three high performers had left and recruiting had become more difficult because the tension in the organization was palpable, even during interviews and walk-throughs. Others were waiting to see what David would do.
This is where most leaders overcorrect. The instinct is to cut both—Maria and Greg, anyone else who looks like a problem—and remove the ambiguity at its source.
But that instinct is wrong. The Marias and the Gregs of an organization are not the same problem, and treating them as the same problem creates fear, uncertainty and anger in those who remain. It also costs you the Marias, who often turn out to be exactly the people you most needed to keep. Speed without discernment is not a solution. It is a different kind of noise.
What David had to do was tell the two apart. Maria got coaching, clarity and time. Within months, she was one of the strongest internal advocates for the new norms. Greg got an exit. By then, the work to remove him was straightforward, because the pattern was undeniable.
(The story of David, Greg and Maria is a composite case study based on real clients.)
Weeding the garden is the disciplined ability to detect behaviors that damage your culture early—the weeds—and act with enough precision to pull only those, not the fruit, flowers and vegetables you want to grow. Neither too quickly nor too slowly.
That precision is the hard part. A leader who pulls everything that looks suspicious risks losing the very people the culture needs most. A leader who pulls nothing watches the weeds take over.
The weeds begin as specific behaviors, which people have an opportunity to change. If they cannot change, or choose not to, then it becomes time to set them free to work in someone else’s culture.
Before you can separate signal from noise, you must articulate the signal. This means defining cultural norms not as values statements on a wall, but as concrete behaviors. Not “we value transparency,” but “We transparently share everyday actions with our teams and the organization so we can celebrate the good, repair the bad, and change the ugly.” When the signal is named, discernment becomes possible. When it isn’t, everything looks like noise. Or, worse, everything looks like a signal.
Not all noise is equal. The worst decisions happen when leaders don’t realize this.
A person who hasn’t yet adopted new behavioral norms might just need coaching, clarity and time. In our Culture Catalyst program, we regularly see people who, given the right environment, align more fully than anyone anticipated. Their previous behavior was shaped by a previous culture. Change the culture, and behavior often follows. Maria was one of those people.
A person actively undermining the new norms, like Greg, is something different. They are not waiting to align. They are planting doubt and cultivating weeds in your garden of culture. Misalignment is a coaching challenge. Active toxicity is an organizational weed problem.
The signal test is direct: Is this person moving toward the culture you are building, or making it harder for others to get there?
When the evidence is clear that someone is too misaligned to repair and actively spreading dysfunction, discernment demands precise action.
Precision means acting on behaviors, specific impacts and patterns over time you have actually observed —not on what you fear or what others have reported. It means communicating clearly about what wasn’t working and why. And it means attending to the people who remain.
When David finally acted on Greg, the culture began to recover faster than he expected. People were relieved. They had been waiting for evidence that the new norms would actually be protected. Maria, by then, was already among the people doing the protecting.
That is what weeding the garden with precision looks like. Not fear. Clarity.
AI can monitor behavior. It can flag patterns. It could generate a report showing David that both Greg and Maria had missed deliverables, kept quiet at two town halls and scored below the organizational average on peer engagement surveys.
But it could not tell David whether Greg and Maria were struggling or sabotaging. It could not read the room when Greg subtly dismissed a new initiative. It could not tell Greg from Maria—two people whose data might look similar but whose behavior, intent and trajectory were not the same.
Discernment is wisdom applied to incomplete information, under uncertainty, with real human stakes on both sides of the decision.
In the age of AI, that human wisdom will be worth more, not less.
In my next article, I’ll explore integration: how leaders turn information into shared meaning, and why alignment depends on more than the right data.
This article is part of a series on the human capabilities that AI cannot replicate. Previous installments have covered connection, trust, accountability and curiosity.
If you are trying to protect a culture you are building, or trying to tell the difference between what is signal and what is noise, I’d like to help you think it through. Book a 30-minute Insight Call with me and let’s talk about what you’re seeing and what to do about it.