Shot, Neglected, Trained: What Happens to Messengers

Published August 25, 2026

Part 2 of The Generative Culture Series

Someone walks into a room with a probabilistic forecast: “The likelihood of us getting this done by this date is basically zero.” For months, project leaders have heard optimistic predictions built without history or data. The person who brings the grounded forecast has made the risk visible. To everyone accustomed to the good news, that person can still look like the bad guy.

I have carried that kind of news often enough to recognize the pattern. Sometimes I was speaking to a client about delays, dependencies, or problems. Sometimes I was challenging an internal plan. The forecast changed the conversation, and the way people received it told me more about the culture than the official values did.

Every organization has messengers. Ron Westrum gave leaders a plain way to watch what happens after one of them carries unwelcome information into the room. In his 2004 typology of organizational cultures, messengers are “shot/neglected/trained.” Those three responses describe what people learn about speaking up.

Westrum built the typology studying information flow in hospitals, aviation, nuclear safety, and NASA, and he was candid about its evidence: plentiful case studies, mixed systematic tests. He offered leaders a diagnostic, and never promised a universal causal law. (Ron Westrum, “A typology of organisational cultures,” Quality and Safety in Health Care 13, Suppl. II, 2004.)

Shot

In a pathological culture, the person who surfaces a problem becomes part of the problem. Leaders blame the messenger “even if all you're doing is pointing it out.” They can treat an unwelcome forecast as evidence that the person carrying it is disloyal or difficult. The question shifts quickly from whether the warning is true to what is wrong with the person who brought it.

That response reaches everyone who watches. They see a colleague raise a risk and lose standing for it. The next person with an early signal now has more to calculate: how certain the evidence needs to be, and how much personal risk comes with saying something. People may keep working the issue quietly while they wait for proof that feels safer to share.

The messenger can look especially threatening after leaders have repeated an optimistic story for months. A contrary forecast exposes the risk and the weakness of the earlier predictions in the same conversation. Blaming the person who made the conflict visible lets people protect the story a little longer. The underlying condition remains.

In DORA's later survey for technology organizations, people respond to this item: “Messengers are not punished when they deliver news of failures or other bad news.” A leader can use that question to see the pathological response directly. Leaders learn less from it about the bureaucratic middle, where people can neglect a warning and change a messenger's reputation without recording a punishment. (DORA, “Generative organizational culture.”)

Neglected

Bureaucratic leaders often receive the same forecast more politely. “That's nice,” someone says. “That's interesting. How did you come up with that? Did you follow our prediction approach?” The project manager followed the sanctioned method and reached the optimistic date. The messenger used a different approach and reached a conflicting answer. People begin by doubting the forecast that came through the unfamiliar route.

The response sounds responsible because leaders really do need controls. In regulated work, they may also need an auditable method and clear accountability for who produced an official forecast. Those requirements still leave room to inspect evidence that contradicts the official view. Trouble starts when leaders require every inconvenient fact to arrive through the sanctioned process before they will consider whether it is true.

Nobody needs to accuse the messenger of anything. People ask about the method, confirm that the project manager followed the rules, and move on. The information receives no hearing on its own terms. A real risk can remain unexamined while everyone involved can truthfully say that they followed the process.

This is why neglect deserves more attention than a simple punished-or-safe test gives it. The messenger may face no discipline and still pay a quiet reputational cost. People begin to see that person as negative, difficult, or insufficiently aligned. The warning gets ignored or overlooked. The next time the person sees something, speaking carries the memory of what happened before.

Leaders may miss their own role because the exchange stayed civil. Nobody raised a voice. Nobody documented retaliation. The status meeting ended on time. Yet everyone watching saw which information counted and which method a person had to use before leaders would listen.

Pay attention to the first test you apply to unwelcome information. You can examine how someone produced a forecast after you understand what the forecast says and what evidence supports it. When procedural legitimacy comes first, people learn that a weak signal has little chance unless an approved system already agrees with them. In those rooms, the first question was still, “Did you follow our prediction approach?”

Trained

In a generative culture, leaders take the news at face value and move toward action. The first response sounds more like, “Thanks for surfacing this. What needs to happen next?” Leaders keep the messenger involved. They gather what that person knows before they revisit the timing or quality of earlier reports.

I want bad news as early as possible because we can do more when we hear it sooner. The content may still be unpleasant. People may still need a difficult conversation with a client or with those responsible for the work. Earlier information gives us more choices while those choices still exist.

When someone brings me bad news, I use a sequence that keeps the immediate risk separate from the trust judgment. The moves are familiar, and I adjust the questions to the person and the news.

First, I say thank you and appreciate the messenger. Raising the issue may have felt difficult or personally risky, so I acknowledge the act of speaking without deciding what caused the delay. I need the information to keep moving.

Then I ask for more: “Anything else?” and “What do you think we can do?” The messenger may have more context, an idea about what to do, or both.

Next, I separate what is within our control from what we can influence. I focus on the actions available now and the help we can request. Someone else may control some of the available decisions.

Finally, I communicate expectations. The people entitled to know the real state hear it early, and resetting expectations never means softening the news. Where disclosure obligations exist, they set the clock, and nothing about how the message is framed changes that.

This sequence keeps the current exchange from becoming an instant adjudication of the messenger. I can return to questions about reliability and judgment after the immediate risk is under control. The person who spoke remains involved and has reason to bring back the next piece of information.

In 1996, Amy Edmondson studied hospital units and found higher reported medication-error rates in units rated as better managed. She interpreted the result to mean that better teams were more able to discuss mistakes — a magnitude Westrum later called “suggestive, not definitive.” A higher count of reported problems can be evidence of greater visibility. Whether any specific receiving ritual works, this study cannot say. (Amy C. Edmondson, “Learning from Mistakes is Easier Said Than Done,” Journal of Applied Behavioral Science 32, no. 1, 1996.)

“Trained” is active. Leaders go first. They surface problems early and ask for help in the open. When someone else raises an issue early, leaders give that person positive recognition, especially when speaking carried personal risk. Over time, people see that bringing a problem forward earns a hearing and that the leader wants to know before every answer is available.

The cost of silence

Shot punishes the act of speaking. Neglect starves it. And training — the only response of the three that takes deliberate work — rewards it. Whichever lesson a culture teaches, people learn it, and leaders collect on it through the green-until-red pattern. A regulatory or audit requirement is due in six months. The updates keep coming back the same way: “We're good.” The project stays green. At two weeks out, someone finally says the work is off track. By then, the organization may have little room to prevent noncompliance, fees, penalties, or audit findings.

People delay warnings for several reasons. They may be optimistic. Some believe they can manage the issue quietly, or hope it will resolve itself. Some conceal what they know. Others fear that bringing the issue up early is going to “look bad.” These are possible explanations within a recurring pattern. A leader usually cannot determine the reason in the first minutes.

Some risks genuinely become knowable late. Nobody hid anything. Nobody missed anything. The fork applies only when a signal existed and stayed out of view. Telling the cases apart starts with one question: did anyone see this earlier and sit on it? When the answer is no, there is nothing to adjudicate. Treat the news as new information and move.

When the answer is yes, both available explanations damage trust: concealment raises a question of integrity, a missed visible risk one of capability. The leader rarely has the context to tell which fits while time is running out.

The late warning also reverses months of reassurance. Leaders are usually angrier than they would have been earlier: they face the underlying problem and the lost time, and now they doubt the reporter too. They still need the warning. If the first conversation becomes a trial of the person who carried it, other people will watch and remember.

You should turn that phrase back on your own response. Bad to whom? What happened the last time someone brought you a small concern before they had every answer? Your response history is part of the reason a team keeps reporting green. Did you test the content of the warning before testing the method that produced it?

What you have trained

Think about the last status surprise that reached you, and how long the work had been reported healthy before it arrived. One question tells you which response your organization runs: after the conversation, what happened to the messenger's standing?

I train for earlier warnings through repeated behavior. I surface problems early myself. When I need help, I ask for it. When someone else raises an issue early, I give that person positive recognition, especially when they risk their standing by speaking.

People see what I support when I recognize the person who raised the issue. I keep going first and repeat this over time. When someone speaks early, I recognize that act even while the underlying risk remains unresolved.

Back to Field Notes