Your internal understanding may be much more developed than your explanation.

You may see the history, the incentives, the second-order consequences, the contradictions, the hidden risk, the alternative interpretation, and the long-term implication all at once.

The other person usually does not.

Start where the other person can enter the problem, then add depth as they need it.

Select the important part, put it in order, and check what the other person received.

Before speaking, decide which of these you are trying to achieve:

1. Inform: I want them to know something.
2. Understand: I want to learn how they see the situation.
3. Clarify: I want us to establish what is happening.
4. Decide: I want a choice to be made.
5. Persuade: I want them to change a belief or action.
6. Negotiate: I want us to reach terms.
7. Coordinate: I want everyone to know what happens next.
8. Support: I want the person to feel understood.
9. Correct: I need to address behaviour or an error.
10. Protect: I need to establish a boundary.
11. Repair: something between us has been damaged.
12. Connect: the relationship itself is the objective.

Name the job before an important conversation. One person may be trying to solve while the other wants to be understood, or one may be trying to discuss while the other wants to decide.

Before explaining anything important, silently ask:

What does this person need from me to understand this?

Start with what they need to receive. The rest can wait.

Use five levels.

Level 1: Headline

One sentence.

“I think we should postpone the launch by two weeks.”

Level 2: Reason

One or two reasons.

“The product is technically ready, but our monitoring and customer support are not.”

Level 3: Evidence

Give the strongest evidence.

“Three unresolved failures can create customer-facing outages, and support coverage is still incomplete.”

Level 4: Implication

Explain what happens if nothing changes.

“Launching now saves two weeks but exposes us to a much larger reputational downside.”

Level 5: Detail

Only now provide technical depth.

Most people need Levels 1–3. Specialists may want Level 5. Do not begin at Level 5 merely because that is where your mind lives.

Whenever someone asks, “What do you think?” do not begin with the entire history.

Use: Conclusion → reason → implication.

“I wouldn't do the deal yet. The economics are attractive, but we still don't understand why the seller is exiting. I want that resolved before we commit.”

Then stop. Let them ask for depth.

BLUF = Bottom Line Up Front.

Use it whenever the audience is senior, busy, impatient, deciding, or under pressure.

Format:

Bottom line: what I recommend.
Why: the two most important reasons.
Need from you: the decision or action.

Example:

“Bottom line: I recommend option B. It costs more initially, but it removes the dependency that creates most of our operational risk. I need your approval by Wednesday.”

Put the recommendation before the supporting detail.

Structure important explanations like this:

Top

The answer.

Middle

Two to four reasons.

Bottom

Evidence underneath each reason.

Example:

We should enter Market A.

Because:
1. customer demand is stronger;
2. competition is weaker;
3. our current capabilities transfer well.

Then provide evidence underneath those three branches.

This preserves the structure without making the listener search through every detail.

Use when explaining a complicated problem.

Situation

What is normally true?

Complication

What changed?

Question

What decision or problem now exists?

Answer

What should we do?

Example:

“We originally planned to launch nationally. Customer testing now shows one region behaves very differently. So the question is whether we still launch everywhere at once. I recommend a regional rollout first.”

Use it for strategy, proposals, investor discussions, and management updates.

Use when speaking spontaneously.

Point

State your view.

Reason

Explain why.

Example

Give one concrete example.

Point

Repeat the conclusion.

Example:

“I think we should hire a senior operator rather than another junior analyst. Our bottleneck is execution, not analysis. Last quarter three projects were researched properly but delayed in implementation. So I would put the next hire into operations.”

This prevents you from constructing the argument while speaking.

When you see seven reasons, organise them into a few named categories.

Instead of: “There are eleven problems.”

Say: “There are really three problems: economics, execution, and governance.”

Then unpack them.

Your mind may see complexity. Your communication should reveal structure.

After every important statement, silently ask: So what?

Bad:

“Customer acquisition cost increased 18%.”

Better:

“Customer acquisition cost increased 18%, which means the current growth plan becomes unprofitable if the trend continues.”

Data without implication is incomplete communication.

Before entering detail, ask:

“Would it be more useful if I give you the conclusion, the reasoning, or the numbers?”

This avoids guessing their preferred level.

Avoid:

“Well, there are several ways of looking at it…”

followed by four minutes of analysis.

Instead:

“My current view is X. There are two reasons I could still change my mind.”

Now the listener knows where you stand.

Because you naturally see uncertainty, you may hedge excessively.

Use three levels.

High confidence

“I'm confident that…”

Moderate confidence

“My current view is…”

Low confidence

“My working hypothesis is…”

Then say what would change your mind.

Example:

“My current view is that demand is real. The thing that would change my mind is seeing retention remain weak after onboarding improves.”

This communicates both conviction and intellectual honesty.

Avoid:

“Maybe, perhaps, I could be wrong, but I sort of think…”

Say:

“My current view is X, with moderate confidence.”

One qualifier is enough.

Use this explicitly when situations are sensitive.

Fact: revenue declined 12%.

Interpretation: I think the main cause is customer churn rather than acquisition.

Recommendation: therefore I would prioritise retention before increasing marketing.

People can disagree with one layer without rejecting the others.

Here's what I think.
Here's why.
Here's what I'm uncertain about.
Here's what I recommend.
Here's what I need from you.

Waiting until you fully understand a problem can make silence look like agreement or permission to continue.

“I'm not ready to give a final view, but I do have a concern.”

Name the concern before you can explain the whole theory.

If something feels wrong, say something while the concern is still 20% formed.

Example:

“I can't fully explain it yet, but I'm uncomfortable with the dependency we're creating. Let me test the concern and come back tomorrow.”

This preserves your position.

Otherwise your pattern may become:

silence → analysis → certainty → strong rejection.

To other people, the rejection appears sudden. To you, it has been developing for weeks. Make the middle visible.

In a reversible discussion, speak once you can name the concern and how you will test it.

Demand much stronger evidence for irreversible commitments, legal statements, public accusations, major financial decisions, and claims where error carries serious consequences.

When your position is unfinished, tell people.

“I'm thinking out loud rather than making a recommendation.”

Now you can explore without being mistaken for making a final commitment.

Before explaining, assess four things.

A. What do they already know?

Skip the introduction if they already know it.

B. What do they care about?

A CFO hears economics. An engineer hears reliability. A customer hears consequences. A partner hears trust.

C. What are they worried about?

Address the hidden concern.

D. What action can they take?

Explain why the information matters to the action available to them.

The same proposal may need four versions.

To a CEO

“This increases speed without increasing organisational risk.”

To finance

“This reduces fixed cost exposure and gives us a clearer payback period.”

To engineering

“This removes the dependency that's causing most of the failure cases.”

To employees

“This should reduce duplicated work and make ownership clearer.”

Same idea. Different doorway.

Your mind may naturally prefer abstractions. Translate them.

Instead of:

“We have an incentive-alignment problem.”

Say:

“Sales gets rewarded for closing deals even when operations cannot deliver them.”

Instead of:

“The relationship lacks reciprocity.”

Say:

“I am usually the one initiating, planning, and repairing things.”

Concrete language is harder to misunderstand.

When something is difficult to explain, use:

Example → pattern → principle.

Example:

“Three projects this quarter had the same problem: nobody knew who had final approval. So I think we have an ownership problem, not three separate project problems.”

This lets the listener discover the abstraction with you.

Be able to explain every important idea in:

• 10 seconds;
• 30 seconds;
• 2 minutes;
• 10 minutes.

If you cannot compress an idea, you may understand many details without yet understanding its core.

Listen

Listen actively for both the words and the feeling beneath them. Do not prepare your answer while they speak.

Affirm

“I appreciate you telling me directly.”

Affirm the person's experience or contribution without pretending to agree.

Respond

“I understand why that would be frustrating.”

Respond to what they actually said and answer questions they asked.

Add information

“May I add one thing that might change how we see the delay?”

Only now introduce missing context, a correction, or another perspective. LARA is designed to keep disagreement inside a dialogue rather than turning it into parallel arguments.

Ask before offering analysis, advice, counterexamples, solutions, or perspective.

“Do you want me to listen, help you think through it, or help solve it?”

You can say:

“I understand why you saw it that way.”

without saying:

“You were correct.”

Use:

“I can see how that landed badly.”

“I understand why you're angry.”

“That makes sense from the information you had.”

Validation acknowledges experience. It does not surrender your judgment.

Use Situation → Behaviour → Impact.

Avoid attacking personality.

Bad:

“You're unreliable.”

Better:

“When the report came in two days late without warning, I had to delay the client review. I need earlier notice if the deadline is at risk.”

Specific behaviour can change. Character accusations usually create defence.

For firmer boundaries:

Describe

State what happened.

Express

Explain the effect.

Specify

State what needs to change.

Consequence

Explain what happens otherwise.

Example:

“This is the third time the scope changed after approval. It makes staffing impossible. From now on, changes after approval need a revised timeline. Otherwise I won't commit the team to the original date.”

When personal conflict becomes emotional:

“When plans change at the last minute without telling me, I feel unimportant. I need more predictability. Can you tell me as soon as you know something is changing?”

Avoid:

“You never respect my time.”

One describes experience. The other starts a trial.

Your likely dangerous pattern:

small problem → tolerate → rationalise → tolerate → internal certainty → final decision.

Replace it with:

small problem → mention → observe response → clarify → decide.

Use the 48-hour rule: If something still bothers you after 48 hours, address it.

Do not accumulate a private legal case against someone.

If someone says, “That hurt me,” do not immediately respond:

“But I didn't mean that.”

Say:

“That wasn't my intention, but I understand that was the effect.”

Then investigate. Intent matters. Impact also matters.

Avoid:

“That makes no sense.”

Use:

“I think we're making different assumptions.”

“I see the logic. The place I disagree is…”

“I agree with the objective. I'm less convinced by the mechanism.”

Separate the person from the idea from the objective.

Before criticizing an important position, summarise it in a form the other person would accept.

“Let me check that I understand your argument before I disagree.”

Then explain it. Ask:

“Is that fair?”

Only then challenge it.

This is especially valuable for you because you naturally see weaknesses quickly. Being able to find a flaw does not mean you have demonstrated understanding.

Use this order:

1. Ask how they arrived there.
2. Identify the assumption.
3. Confirm the shared objective.
4. Introduce contradictory evidence.
5. Offer an alternative.
6. Invite correction.

Example:

“I think we're both trying to reduce churn. My concern is that we're assuming price is the cause. The exit interviews point more strongly to onboarding. I'd test that before discounting. What am I missing?”

Persuasion fails when you begin with why you care.

Their objective

What do they want?

Their obstacle

What prevents it?

Your proposal

How does your idea help?

Evidence

Why should they believe you?

Cost

What does your solution require?

Next step

What small commitment should happen now?

Persuasion often requires a sequence.

Instead of:

“Agree with my entire strategy.”

Aim for:

“Can we agree this assumption should be tested?”

Then:

“Can we agree the result changes the decision?”

Move one bridge at a time.

Before negotiating, know:

What will I do if we do not reach agreement?

This is your alternative. Define it before an important negotiation so you know which terms you can reject.

Before entering, know:

1. ideal;
2. target;
3. minimum acceptable;
4. walk-away.

Do not invent these while sitting across from someone.

Say your position. Give one reason. Stop.

Bad:

“I'd really need 20%, although I understand budgets are difficult, so maybe 15%, or…”

You just negotiated yourself downward.

Better:

“I'd need 20%. The scope is materially larger than the role we originally discussed.”

Silence.

Position:

“I need Friday.”

Interest:

“I need the work before Monday's board meeting.”

Now alternatives become possible.

Ask:

“What does Friday allow you to do?”

Understanding the real need expands solutions.

Instead of:

“I can't.”

Use:

“I can do A by Friday or A+B by Tuesday. Which matters more?”

This is especially important if you tend to become over-responsible.

Do not silently absorb impossible scope. Make trade-offs visible.

“I can take that on if we move X. Which should have priority?”

This makes the workload trade-off visible.

Say:

“I don't want to answer too quickly. Let me think and come back by 3 PM tomorrow.”

Time requests become trustworthy when they contain a return point.

Say:

“I don't know yet.”

Then add:

“Here's how I'll find out.”

And:

“I'll come back by X.”

Give uncertainty a process and a return time.

Use four sentences.

“I got X wrong.”

“The cause was Y.”

“The impact is Z.”

“I'm doing A to correct it and B to prevent recurrence.”

Stop before explanation turns into defence or blame.

Ask:

“What happened?”

before:

“Why did you do that?”

“What happened?” invites information. “Why did you do that?” often sounds accusatory even when unintended.

As authority increases, you must communicate priorities, decision rights, standards, context, trade-offs, and expectations.

Rank competing priorities instead of leaving competent people to guess which one wins.

For any important initiative, state:

Why

Why are we doing this?

What

What exactly are we trying to achieve?

What not

What are we deliberately not doing?

Who

Who owns what?

When

What are the deadlines?

Decision rights

Who decides if disagreement remains?

Success

How will we know it worked?

Clarity reduces unnecessary management.

Preserving every complication can make an explanation look sophisticated and leave the audience lost.

Make the difficult part understandable without making it false.

Do not say:

“Handle this.”

Communicate:

1. objective;
2. constraints;
3. authority;
4. deadline;
5. quality threshold;
6. check-in point.

Example:

“I need three viable vendors by Friday. Budget is under $80,000. You can contact anyone except our current partner. Choose the shortlist yourself; I'll make the final selection Monday.”

Use:

Continue → Change → Why → Next

Example:

“Keep the analysis depth. Change the presentation length; the main recommendation appears too late. The audience needs to decide quickly. Next time put the recommendation on slide one.”

Do not make people reverse-engineer your standard.

Ask:

“What specifically should I do differently next time?”

If feedback is vague:

“Can you give me one example?”

If feedback conflicts:

“Which outcome are you optimising for?”

Turn judgment into usable information.

Do not build presentations around everything you discovered.

Build them around what the audience needs to decide.

Ideal order:

1. question;
2. answer;
3. three reasons;
4. evidence;
5. risk;
6. recommendation;
7. next action.

Your research process and your presentation structure should not be identical.

Every slide should answer one question.

The title should ideally contain the conclusion.

Bad title:

“Customer Data”

Better:

“Retention, not acquisition, is now the main growth constraint.”

A person should understand your argument by reading only the slide titles.

Before important meetings, write:

Three things they must leave knowing.

Three things you need to learn.

One decision you want.

One objection you expect.

One sentence you must say even if the meeting becomes chaotic.

This makes your performance less dependent on improvisation.

If discussion becomes diffuse:

“Can we clarify what decision we're trying to make?”

If people are talking past one another:

“I think we agree on the objective but disagree on the assumption.”

If something is unresolved:

“What information would actually settle this?”

If everybody agrees too quickly:

“Before we close, what's the strongest argument against this?”

If discussion is endless:

“What would prevent us from deciding today?”

Say:

“Let me check the outcome.”

Then summarise the decision, owner, deadline, unresolved issue, and next checkpoint. If any are unclear, settle them before people leave.

Your writing should not reproduce your entire internal reasoning.

Subject or headline

What this is about.

Conclusion

What matters.

Reason

Why.

Action

What happens next.

Example:

Subject: Recommend delaying launch to 18 October

I recommend moving launch by two weeks. Monitoring and customer-support readiness remain below the threshold we agreed. Engineering can close the major gaps within that window. Please confirm by Thursday whether we should revise the public date.

Use:

1. Current status.
2. What changed.
3. Why it matters.
4. What happens next.
5. What you need from them.

This should be your default professional update.

Never send six paragraphs and end with:

“By the way, could you approve this?”

Put the ask early.

“I need approval for X by Thursday. Context below.”

Written brevity can accidentally sound cold.

When the relationship matters, add relational context.

Instead of:

“Can't tonight.”

Use:

“I can't tonight. I'm wiped out and need a quiet evening. I do want to see you though. Saturday?”

Same boundary. Much less ambiguity.

Assume tone will be interpreted more negatively than intended when the message is very short, conflict already exists, hierarchy is involved, or timing is sensitive.

If tone matters, add one sentence of intent.

“I'm asking because I want to make sure we solve this before it becomes bigger.”

Use text to identify the issue, set the conversation, and clarify logistics.

Example:

“I think we're misunderstanding each other and I don't want to make it worse over text. Can we talk tonight?”

Do not conduct a three-hour prosecutorial exchange in messages.

Be useful, memorable, and easy to reconnect with.

Notice

Something specific about their work.

Connect

Explain why it matters to you.

Ask

One real question.

Exit

Do not trap them.

Example:

“Your point about customer incentives was interesting because I'm dealing with a similar problem in a different market. How did you determine incentives were the cause rather than product quality?”

Specific interest gives them something real to answer.

Prepare three versions of yourself.

5 seconds

Who you are.

20 seconds

What problem you work on.

60 seconds

Why it matters and what makes your approach different.

Never force yourself to invent your identity live every time.

Within 48 hours:

“Good meeting you yesterday. Your point about X stayed with me because of Y. Here's the article or project I mentioned. Would be glad to continue the conversation sometime.”

One specific follow-up gives the next conversation somewhere to begin.

Family interactions often contain old roles. Therefore say the current reality explicitly.

“I understand why you prefer that choice. I'm going to make this decision differently.”

Do not turn every boundary into a philosophical debate. You can respect someone without persuading them.

Use the broken-record technique.

“I understand your concern. I've decided to do X.”

If repeated:

“I hear you. My decision is still X.”

Do not provide ten new reasons. Every additional reason becomes another object to argue against.

Your biggest risk is expecting a partner to infer things you have processed privately.

Make internal information external.

“I'm stressed and quieter than usual. It's not about you.”

“I need two hours alone, then I want to talk.”

“I'm upset, but I'm not ready to explain it clearly yet.”

“I want reassurance right now, not advice.”

These sentences prevent unnecessary interpretation.

Do not convert needs into tests.

Bad: become distant and see whether they notice.

Better:

“I've been needing more initiative from you lately.”

Bad: wait to see if they remember.

Better:

“This is important to me. I'd like you to remember it.”

Tests create ambiguity. Requests create information.

Request:

“Would you be willing to do X?”

Boundary:

“I cannot continue Y.”

Demand:

“You must do X because I want it.”

Know which one you mean.

A useful apology can:

1. explain what went wrong;
2. acknowledge responsibility;
3. express regret;
4. offer repair;
5. say how you will prevent a repeat;
6. request forgiveness without demanding it.

Use only the parts that are sincere and relevant.

Example:

“I should have told you earlier that I was reconsidering the plan. I stayed quiet, which made the decision feel sudden. That was my mistake, and I'm sorry. I will correct the timeline today and tell you when concerns begin next time. I hope you can forgive me.”

Leave “but you also” for another conversation.

Do not decide only from words.

Watch:

acknowledgment → repair → changed behaviour.

Communication is not merely what people say. Repeated behaviour communicates too.

The Krebs crisis-message principles are:

1. communicate what is known;
2. communicate what is not yet known;
3. explain what you are doing;
4. tell people what they should do;
5. say when and where more information will appear;
6. communicate consistently and frequently;
7. use trusted messengers;
8. set the expectation that information may change.

Example:

“The system went offline at 09:20. Customer data currently appears unaffected; we do not yet know the root cause. Engineering is restoring service and preserving logs. Customers do not need to reset credentials at this time. We will update this page at 10:30, or earlier if that guidance changes.”

Do not speculate, conceal uncertainty, or promise that early information is final.

When consequences are large, slow your certainty while keeping the message clear.

“Here is what we know.”

“Here is what we don't know.”

“Here is my current recommendation.”

“Here is the downside if I'm wrong.”

“Here is what would trigger a change.”

Do not communicate the final conclusion at peak emotional intensity.

Use:

“I'm too angry to have this conversation constructively right now. I want to return to it tomorrow.”

But return. Do not use time-outs as avoidance.

First determine whether it was intentional disrespect, incompetence, cultural difference, misunderstanding, or a repeated pattern.

Match the response to the cause and the pattern.

Reduce explanation.

“No.”

“That doesn't work for me.”

“I'm not agreeing to that.”

“I've already answered.”

People who respect boundaries may appreciate explanations. People who exploit boundaries often use explanations as negotiation material.

Do not compete in speed.

Say:

“I want to make sure I answer the substance rather than the pace. Give me a moment.”

Or:

“Let me summarise where I think we disagree.”

Control structure instead of speed.

Use:

“Let me finish this point and then I want to hear your response.”

If repeated:

“I haven't finished.”

No apology required.

Say:

“Sorry, I cut you off. Finish your thought.”

Do not pretend it did not happen.

Do not automatically repeat the same explanation louder or longer.

Ask:

“Which part isn't landing?”

Or:

“What are you hearing me say?”

Their answer identifies the transmission failure.

For important instructions:

“Just so I know I explained it clearly, can you tell me how you're planning to approach it?”

Do not ask:

“Do you understand?”

Most people say yes. Teach-back reveals actual understanding.

Use:

What is it?

Why does it matter?

How does it work?

Example.

Exception.

Do not begin with the exception. You may naturally see edge cases immediately. The listener first needs the normal case.

Give the minimum information needed to understand the normal case.

Then ask:

“Want the deeper version?”

Your tendency may be:

answer → qualify → add nuance → add exception → reopen answer.

Do not.

Once the answer is complete, stop. Silence is allowed.

When criticised, do not immediately prove why your reasoning was rational.

Ask:

“What outcome did my approach create?”

A decision can have been intellectually defensible and still produce a poor interpersonal result. Both matter.

“Do you want an idea?”

An unrequested solution can feel like criticism.

If you express care through actions, occasionally name the intention.

“I handled this because I wanted to take something off your plate.”

“I remembered this mattered to you.”

“I'm asking because I care about how this affects you.”

Avoid vague praise.

Say:

“The way you handled the client objection without becoming defensive was excellent.”

Specific praise teaches people what you value.

Prefer:

“What am I missing?”

“What would change your mind?”

“What outcome are you trying to protect?”

“What is the biggest constraint?”

“If we couldn't do X, what would we do?”

“What would make this fail?”

“Which assumption carries the most risk?”

“What decision does this information change?”

These suit your natural strategic thinking while inviting the other person into it.

When you are unsure what someone means, ask.

Instead of:

“They obviously don't value my contribution.”

Ask:

“How are you thinking about my role in this?”

Use:

Recommendation → business consequence → risk → ask.

Put judgment before supporting detail.

Give more context than senior people need.

Explain why, what good looks like, what they can decide themselves, and when to escalate.

Do not make them guess your expectations.

Make reciprocity explicit.

“I can own A. Can you own B?”

Avoid silently doing both and later resenting it.

You may go deeper, but first align definitions.

Many technical arguments are actually arguments caused by different meanings for the same word.

Ask:

“When you say X, what exactly are you including?”

Never use jargon when an ordinary word works.

If jargon is necessary:

term → plain definition → example.

The goal is not to make them speak your language. The goal is shared understanding.

Never assume silence means agreement, directness means hostility, indirectness means dishonesty, quick decisions mean confidence, or slow decisions mean weakness.

When uncertain, establish the norm explicitly.

“Would you prefer that I give direct feedback here, or discuss concerns privately?”

Failure 1: Processing privately for too long

Counter: “I have an early concern.”

Failure 2: Giving too much context

Counter: conclusion first.

Failure 3: Hedging until your position disappears

Counter: “My current recommendation is…”

Failure 4: Solving when someone wants understanding

Counter: “Listen or solve?”

Failure 5: Avoiding conflict until the issue becomes large

Counter: address persistent problems within 48 hours.

Failure 6: Accepting excessive responsibility

Counter: “If I take X, what moves?”

Failure 7: Correcting details that do not affect the outcome

Counter: ask whether the correction changes a decision.

Failure 8: Assuming intelligent people will infer your meaning

Counter: state the implication.

Failure 9: Assuming your actions communicate your intentions

Counter: occasionally explain why you did something.

Failure 10: Mistaking precision for clarity

Counter: simplify without falsifying.

• careful listening;
• structural thinking;
• detecting contradictions;
• emotional restraint under pressure;
• thoughtful writing;
• preparation;
• pattern recognition;
• willingness to revise;
• explaining causes rather than symptoms;
• seriousness about commitments.

Make those strengths visible in words another person can use.

Use plain words and expose the context the listener needs.

Write five things:

1. What outcome do I want?
2. What do they probably want?
3. What must they understand?
4. What do I need to understand?
5. What sentence must I not leave without saying?

Remember:

Listen first.

State the headline.

Give no more than three reasons.

Check whether they understood.

Ask what they think.

Surface disagreement early.

Name the decision.

Name the next step.

Ask:

• What did we decide?
• What remains unresolved?
• Who owns what?
• By when?
• Did I say what I meant?
• Did they understand the same thing I intended?

If important, send a written recap.

Thinking silence

“Give me a moment.”

Disagreement silence

Do not remain silent.

“I have a concern.”

Emotional silence

“I need some time before I can discuss this properly.”

Strategic silence

After asking or negotiating, stop talking.

Your silence should be intentional rather than ambiguous.

“Can we reset for a second? What are we trying to decide?”

Then ask:

“What do we agree on?”

“Where exactly do we disagree?”

“What information would resolve it?”

“I think we're reacting to each other rather than understanding each other. Can I tell you what I heard, and then you tell me what I missed?”

“I understand what you're asking. I'm not agreeing yet.”

This keeps urgency from becoming consent.

“Before I answer, I need to understand the trade-off.”

Slow the interaction, recover the structure, then decide.

“Let me restate the priority, owner, and deadline.”

These three facts resolve many coordination problems.

“There is something small I want to say before it becomes something big.”

“I don't know yet. My current guess is X, but I don't trust it enough to act on. I need Y information. I'll come back by Z.”

“My recommendation is X. The main reasons are A and B. The largest risk is C. I would change my mind if D happens.”

“I agree with the objective. I disagree with one assumption in the current approach. Here's the assumption, here's why I'm concerned, and here's what I'd test instead.”

“When X happened, the result was Y. From now on, I need Z.”

“That sounds difficult. Do you want me to listen, help you think, or help you solve it?”

“I can do X. If Y is also required, something else needs to move. Which is the priority?”

Before

Outcome: What needs to happen?

Audience: What do they care about?

Headline: What is my conclusion?

Three reasons: No more.

Concern: What could they misunderstand?

Ask: What do I need from them?

During

Headline first. “My view is…”

Clarify objective. “What outcome are we optimising for?”

Check level. “Do you want the short answer or deeper reasoning?”

Signal unfinished concerns. “I have an early concern.”

Buy thinking time. “Give me a moment.”

Separate assumptions. “I think we're assuming different things.”

Check understanding. “What are you hearing me say?”

Close. “Decision, owner, deadline?”

Personal conflict

“I want to say this while it's still small.”

“When X happened, I experienced Y.”

“What I need is Z.”

“How did you see it?”

“What should we do differently next time?”

Negotiation

Know your ideal, target, minimum, walk-away, and alternative. Say your position. One reason. Stop. Never negotiate against yourself.

When overwhelmed

Do not explain everything. Say: “There are three things that matter.” Then name them.

When uncertain

“My current view is X.”

“Confidence is moderate.”

“The thing that would change my mind is Y.”

When saying no

“That doesn't work for me.”

or

“I can do A or B, not both.”

When someone is emotional

“Listen, think together, or solve?”

When you made a mistake

“I got X wrong. Here's the impact. Here's the correction.”

When something feels wrong

Do not disappear into analysis.

“I don't have the complete argument yet, but I want to flag the concern now.”

Your communication becomes strongest when you stop trying to transmit your entire internal model.

The other person needs the part that matters, in the right order, at the right level of detail, with the implication made explicit, and enough room to respond.

“Create enough shared understanding for the next right action.”