At eight in the morning, the night team hands over to the day team.

It takes only a few minutes, but a remarkable amount depends on getting those few minutes right. Everything the incoming team needs to know has to move from one group of clinicians to another clearly enough for the work to continue safely.

There is a rule inside a good clinical handover that I think leaders should borrow. You do not start with what you think. You start with what you can see.

Not, "I am worried about the man in bay four."

But, "His respiratory rate was 16 at midnight. It was 24 by six this morning. His blood pressure has dropped twenty points."

Now we have somewhere to start.

We might disagree about the diagnosis. We might disagree about what needs to happen next. But we can both look at the same information and work from there.

That distinction matters far beyond medicine, because many difficult leadership conversations go wrong before the difficult part even begins.

You have probably had it several times already. Just not with the person who actually needs to hear it.

You have rehearsed it in the car. You have replayed the last incident in your head and worked out exactly what you are going to say. By the time you walk into the meeting, the conversation feels almost finished.

Then they sit down opposite you, and somehow you hear yourself saying:

"How are things? I just wanted to check in."

Twenty minutes later, the meeting is over. They leave reassured. You leave frustrated. And the thing you arranged the meeting to discuss is still sitting there between you, completely untouched.

Why do intelligent leaders do this?

Usually, it is not because they lack courage. They are trying to protect something. They are trying to protect the relationship, the other person's confidence, perhaps even their own reputation as a reasonable leader.

So they soften the opening. Then they soften it again. Somewhere along the way, the sentence becomes so safe that it no longer says anything.

The irony is that avoiding discomfort today usually creates a harder conversation tomorrow. The other person now has evidence that everything is fine. You had the meeting. You smiled. You asked how things were going. Nobody said there was a problem.

The next time you raise it, you are not starting from zero. You are starting by undoing the reassurance you gave them last time.

This is why SAID begins with Situation.

Before impact, before explanations and certainly before solutions, answer one question:

What is actually happening?

It sounds simple. It isn't.

By the time most leaders finally decide to have a difficult conversation, they have usually spent weeks interpreting the other person's behaviour.

"She isn't committed anymore."

"He has a bad attitude."

"She has disengaged."

"He doesn't respect deadlines."

Those sentences feel like facts because you have been thinking them for so long. But they are not facts. They are conclusions.

Strip them back.

Instead of, "Your commitment seems to have dropped," try, "The last three reports came in after the deadline."

Instead of, "You have disengaged from the team," try, "You haven't spoken in the last four team meetings."

Instead of, "You don't respect other people's views," try, "You interrupted three people while they were speaking in yesterday's meeting."

The difference looks small on the page. In the room, it changes everything.

The first version tells someone what you have concluded about them. The second gives both of you something specific to examine together.

Here is the simplest test I know:

Could a camera have recorded it?

"Your attitude has changed." A camera could not record that.

"You arrived after the agreed start time for our last three meetings." It could record that.

"You don't seem interested anymore." No.

"You haven't responded to the last four project emails." Yes.

A camera does not know whether somebody is committed, disengaged or carrying a bad attitude. It only records what happened.

That is your starting point.

Not because your interpretation is necessarily wrong. You may eventually discover that you were exactly right. But if you begin with the interpretation, you ask the other person to defend themselves before you have given them a chance to help you understand what happened.

Start with what happened.

Then find out what it means.

There are three patterns I see repeatedly.

"Your attitude has changed."

"You're not committed."

"You've disengaged."

Think about what the other person now has to do. They have to defend themselves. Of course they do. You have not asked them to examine something that happened; you have told them something about who they are.

That is why specificity matters. The more observable your opening sentence is, the less energy the conversation has to spend arguing about the premise.

"I just wanted to touch base."

"It's probably nothing."

"I thought we could have a quick chat."

"Don't worry, this isn't a big deal."

Notice what is happening. Before you have even reached the issue, you are teaching the other person how seriously to take it.

Not very seriously.

This is often done in the name of kindness, but kindness and vagueness are not the same thing. You can be warm without being unclear. You can protect someone's dignity without pretending there is nothing to discuss.

Sometimes the kindest thing you can do is make sure the other person leaves the room understanding what you meant.

"The reports have been late, so you need to reprioritise."

Maybe.

But perhaps workload is not the problem. Perhaps the reporting process is broken. Perhaps another executive keeps changing the requirements. Perhaps there is something happening outside work that you know nothing about.

The moment you attach the solution to your opening sentence, you have stopped diagnosing. You have decided what the problem is before hearing the other half of the information.

Leaders do this far more often than they realise. We see a symptom, decide what caused it, prescribe the treatment and then invite the other person into a conversation whose conclusion has already been written.

That is not diagnosis.

It is confirmation.

Before your next difficult conversation, ask yourself three questions.

If not, you are probably describing your interpretation rather than the situation. Find the observable fact underneath it.

Behaviour can be discussed and changed. Character invites defence. Keep the conversation anchored to what happened rather than what you think it says about the person.

This one matters most. If you walk into the room already certain about the diagnosis, you will spend the conversation looking for evidence that confirms it and discounting anything that does not.

That is not a conversation.

It is a verdict with questions attached.

Think about the conversation you have been postponing. There is probably one.

Do not plan the whole conversation. Just write the first sentence exactly as you would say it.

Now look at it again. Remove the judgments, assumptions and softeners. Keep the dates, numbers, specific incidents and words that were actually said.

What remains will probably be shorter and plainer. It may also feel more uncomfortable to say because there is nowhere left to hide the issue.

Say that sentence first.

Then stop talking.

Do not rush to explain what you think it means. Do not immediately tell them how to fix it. And do not fill the silence simply because it feels uncomfortable.

Give the other person a chance to look at the same situation with you.

There will be time for impact. There will be time for their perspective. There will be time to decide what happens next.

But first, you both need to be looking at the same thing.

A director had the same conversation with the same person four times in one year.

Each conversation ended warmly. Each time, she came away believing the message had landed.

Nothing changed.

She thought she had been clear. She had certainly been kind.

Those are not always the same thing.

By the time she understood the difference, eighteen months had passed and the organisation had lost someone it should have kept.

At first glance, this looked like a performance problem. It wasn't. That was the presenting symptom.

The real diagnosis was a conversation that everyone thought had happened, but never really had.

Inside this week's Diagnosis Room, I unpack the case, show where those conversations went wrong, and give the prescription I would use to stop the pattern repeating.

14-day free trial. No card required.

Over the next four weeks, I am taking SAID apart one letter at a time. Not as theory, but as something you can use in the next difficult conversation you have.

S - Situation: What is actually happening?

A - Acknowledging Impact: What is it costing?

I - Inviting Perspective: What am I missing?

D - Deciding Together: What happens next?

The order matters.

If you cannot name the situation clearly, you cannot talk honestly about its impact. If you have already decided what caused it, you cannot genuinely invite the other person's perspective. And if you skip both of those steps, "deciding together" becomes another way of telling someone what you have already decided.

So this week, stay with S.

What happened?

Nothing more.

Next week, we move to Acknowledging Impact.

Last week on LinkedIn, I wrote that nobody pays you exactly what you are worth.

Read the post and join the conversation →

Difficult Conversations - Douglas Stone, Bruce Patton and Sheila Heen

One of the most useful ideas in the book is that a difficult conversation often contains several conversations at once. We are arguing about what happened, dealing with how we feel about it, and somewhere underneath both, wondering what all of this says about who we are.

Separating those conversations is much of the work.

It is another reason to begin with what can actually be observed before deciding what it means.

Find it here →

A good clinical handover does not begin with a verdict. It begins with what can be observed.

Two clinicians start with the same information and then work out what it means.

A difficult leadership conversation should work the same way. But we often reverse the order. We arrive with the meaning already decided and tell the other person what their behaviour proves about their attitude, commitment or character.

Then we wonder why they defend themselves.

And naming something does not mean announcing your conclusion about it. It means being willing to say the plain thing first.

"This happened."

"This was said."

"This deadline was missed."

"This has happened three times."

Then stop.

There will be time for meaning. There will be time for impact. There will be time for solutions.

First, make sure you are both looking at the same thing.

So think about the conversation you have been rehearsing.

What is the observable fact you keep softening?

Start there.

Lead honestly. Lead clearly. Lead where it matters.

Dr Ikechukwu Okoh, The Leadership Diagnostician

Dr Ikechukwu Okoh is a physician executive, leadership advisor, and author of Leadership Pulse, where he helps leaders make better decisions under pressure. He is also the host of The Fireplace Conversations podcast. Learn more at ikonmd.org.

New here? Start with MISDIAGNOSED, my free executive briefing on why smart leaders treat the wrong problem, and how to diagnose it before they act.

Forward this to one leader who has been rehearsing the same conversation for a month.