Article 9: What does a conversation teach someone about having the next one?

We tend to judge difficult conversations by what happens in the moment. Was the issue resolved? Did both people remain calm? Was the feedback understood? Did the employee get the support they needed? Was an agreement reached about what would happen next?

Those outcomes matter, of course, but I think they give us a surprisingly narrow view of whether a difficult conversation has gone well.

There is another outcome that may be just as important, particularly over time: what did this conversation teach the person about having another one?

Every time somebody raises a concern, disagrees with a decision, asks for support, discloses something personal or tells another person something they may not want to hear, they learn something from the response they receive. They learn whether speaking was worthwhile, whether the relationship could tolerate discomfort and whether the organisation was able to respond constructively. They may also learn something much less encouraging: that next time, keeping quiet would probably be easier.

That behavioural legacy can extend far beyond the issue being discussed.

Conversations create evidence

Imagine an employee raises a relatively small concern with their manager. Perhaps they disagree with how a piece of work is being handled. The manager listens, asks questions, explains their perspective and, although they ultimately decide not to change the approach, the employee leaves feeling that their concern was taken seriously.

The employee has learned something useful. They have not learned that speaking always gets them what they want, because that would be unrealistic. They have learned that disagreement is survivable within this relationship.

Now imagine the same concern receives a different response. The manager becomes visibly irritated, explains why the employee has misunderstood the situation and closes the discussion quickly. Nothing dramatic happens afterwards. There is no retaliation, no formal consequence and perhaps the manager has forgotten the interaction by the following day.

The employee may not.

They have acquired information about what happens when they challenge this person.

The next time something concerns them, that information becomes part of the decision about whether to say anything.

This is why I think individual conversations can have consequences that are easily missed. The immediate issue might have been trivial, but the interaction itself has become evidence.

Psychological safety is built through these small pieces of evidence

We sometimes talk about psychological safety as though it is an organisational characteristic that can be created through the right culture statement, leadership programme or set of values. Those things may contribute to it, but people's sense of whether it is safe to speak is also shaped through repeated everyday experiences.

What happened when I disagreed last time? What happened when somebody admitted they had made a mistake? What happened when my colleague questioned something in a meeting? What happened when somebody said they were struggling? What happened when an employee disclosed a disability or asked for an adjustment?

People are continually gathering this information, often without consciously thinking about it as psychological safety. They are simply learning how the environment works.

Importantly, they do not only learn from their own conversations. We watch what happens to other people too. If a colleague challenges a senior leader and is treated respectfully, that tells us something about what might happen if we do the same. If somebody raises a concern and subsequently becomes labelled as difficult, we notice that as well.

Organisational culture is partly communicated through what happens to people after they say something uncomfortable.

A successful conversation does not require agreement

This is where I think our definition of a “good” difficult conversation sometimes becomes too focused on resolution. We understandably want people to reach agreement, solve problems and leave conversations with positive outcomes, but some workplace disagreements cannot be resolved in a way that satisfies everybody.

A manager may not be able to approve a request. A colleague may still disagree with another person's interpretation of events. An organisational decision may remain unchanged despite an employee raising legitimate concerns. Sometimes there simply is not a neat solution available.

That does not necessarily mean the conversation has failed.

Someone can leave disappointed with the outcome while still feeling that speaking was worthwhile. They can disagree with the decision while believing that their perspective was heard. They can receive an answer they did not want without feeling punished for asking the question.

That distinction matters because the longer-term behavioural message can still be positive: I can raise difficult things here, even when the answer is not necessarily the one I want.

Perhaps that is a much more realistic standard for healthy workplace dialogue than expecting every difficult conversation to end in consensus.

The opposite can also happen

A conversation can produce the technically correct outcome and still teach somebody not to speak again.

An employee might receive the adjustment they requested but have to disclose far more personal information than felt necessary to obtain it. A concern might eventually be addressed, but only after the person raising it has repeatedly chased for action and defended why it mattered. A manager might apologise for something they did while spending most of the conversation explaining why they did not intend any harm.

On paper, the issue has been resolved.

Psychologically, something else may have happened.

The employee may conclude that getting support requires more emotional effort than they are willing to expend again. They may decide that raising a concern technically works, but the process is exhausting. They may become more selective about what they disclose or decide that smaller problems are simply not worth the effort.

This is why measuring the outcome of a difficult conversation without understanding the experience of getting there can be misleading.

The first response may have disproportionate importance

This becomes particularly significant when somebody has spent considerable time deciding whether to speak. By the time they finally raise something, they may already have rehearsed possible reactions and worried about how the conversation will be received.

The first response therefore carries information.

If somebody discloses something personal and the immediate reaction is discomfort, they notice. If they raise a concern and the first response is an explanation of why they are wrong, they notice. If they challenge a decision and somebody becomes defensive, they notice that too.

Equally, people notice curiosity, respect and the willingness to listen.

That does not mean managers and colleagues must respond perfectly. Difficult information can be surprising and emotionally challenging for the recipient too. What matters is whether the interaction ultimately communicates that the conversation itself was legitimate, even if it was uncomfortable.

There is a significant difference between communicating, “I disagree with what you are saying” and communicating, “You should not have said it.”

Organisations that want people to speak need to become very good at maintaining that distinction.

Avoidance can be learned

This is also why I am uncomfortable when avoidance is treated primarily as an individual characteristic. We sometimes describe people as conflict avoidant, lacking confidence or reluctant to speak up without asking what experiences may have shaped that behaviour.

Perhaps someone has learned that raising concerns achieves very little. Perhaps previous disclosures have changed how they were treated. Perhaps challenging authority has carried consequences before. Perhaps they have repeatedly watched difficult conversations become defensive, personal or unresolved.

In those circumstances, avoiding the next conversation may not reflect a lack of communication skill at all. It may reflect learning.

The person has gathered evidence from previous interactions and adjusted their behaviour accordingly.

If that is true, simply teaching them to be more assertive is unlikely to address the underlying problem. We would be asking them to override what their experience has taught them without changing the environment that taught it.

Leaders are teaching people how to talk to them all the time

I think this has important implications for leadership because managers are continually teaching people what kinds of conversations they can have with them, often without realising they are doing it.

Every response to disagreement communicates something. Every reaction to bad news communicates something. Every time somebody says, “I need help,” “I don't agree,”

“Something isn't working,” or “I need to tell you something,” the response becomes part of that person's understanding of the relationship.

Leaders therefore do not create openness simply by saying that they welcome feedback. They create it through what happens when feedback is uncomfortable. They do not create psychological safety simply by encouraging employees to speak up. They create it through what employees experience after they do.

This does not require perfection. Managers will sometimes become defensive, misunderstand something or respond less skilfully than they would have liked. What matters then is whether they can recognise it and return to the conversation. Repair itself can become evidence that difficult interactions do not have to permanently damage relationships.

Perhaps that is an important part of psychologically safe workplaces too: not the absence of imperfect conversations, but confidence that imperfect conversations can be repaired.

We may need a longer-term definition of conversational success

This is why I think we should judge difficult conversations over a longer time frame.

Alongside asking whether the immediate problem was resolved, perhaps we should also ask whether the interaction increased or reduced the likelihood that somebody would speak again. Did it strengthen trust or weaken it? Did the person leave with greater confidence in the relationship, or with a clearer understanding of what they should avoid discussing in future?

Those questions matter because workplace relationships are rarely built through one significant interaction. They develop through accumulations of smaller experiences, each of which changes what people expect from the next one.

A single poor response does not inevitably destroy trust, just as one positive conversation does not create psychological safety. But patterns develop, and people learn from them.

That learning eventually becomes behaviour.

Perhaps the most important outcome of a difficult conversation, then, is not simply whether we managed today's problem. It is whether the way we handled today's problem made tomorrow's conversation more possible.

Because every difficult conversation is doing two things at once. It is dealing with whatever has brought us into the room today, while quietly teaching us something about whether we should come back into that room next time.

And organisations that genuinely want people to speak may need to pay much more attention to both.