They didn’t get it.
So you explained it again, better this time. More carefully. You added the bit of context you’d left out, gave an example, walked through the reasoning properly.
And they got it less.
You could see it happening. Somewhere in your second pass the expression changed from mild confusion to something flatter, and by the end they said “yeah, no, that makes sense” in the tone that means the conversation is over.
Then they went away and did the thing you’d asked them not to do.
Here’s what happened, and it’s not that they weren’t listening.
What Goes Wrong
Re-explaining an unclear point commonly reduces comprehension rather than improving it, for four reasons. Explaining “more clearly” usually means explaining more elaborately: additional context, qualifications and examples increase the number of components the listener must hold, when the original difficulty was already one of holding too many. Repetition without structural change supplies no new information, so a listener who did not follow the first version has nothing further to work with. Re-explanation carries a status implication — it communicates that the speaker considers the listener not to have understood — which frequently produces a premature signal of comprehension to end the exchange. And the speaker typically re-explains the whole idea rather than locating the specific point at which the listener lost the thread, so the effort is distributed across material that was already understood. The effective response is diagnostic rather than expository: identifying where comprehension broke, then supplying a distinction rather than more detail.
Four mechanisms. The fourth is the fix.
One: Clearer Usually Means Longer
Watch what you actually do when you explain something a second time.
You add. Almost everybody adds.
You supply the background you’d assumed. You put in the caveat that makes it precise. You give an analogy. You mention the exception so they don’t get caught by it later.
Every one of those is a good-faith attempt at clarity, and every one of them adds a component.
Now consider what the listener’s problem was. They were already holding more pieces than they could manage — that’s what not following feels like from the inside. The pieces stopped connecting.
You’ve just handed them four more.
So the second explanation is objectively better as an explanation and worse as a transmission, and those two things are not the same variable. Precision and comprehensibility diverge past a certain point, and thorough people reliably optimise for the first while believing they’re improving the second.
Two: Repetition Supplies Nothing
Second, and it’s the version most people default to under mild frustration.
You say it again. Slightly slower. With more emphasis on the important words.
The content is identical, so there’s nothing new to work with. If the arrangement of ideas didn’t land the first time, the same arrangement delivered more emphatically doesn’t give the listener a different handhold — it just tells them you think the problem is on their end.
And there’s a signal in the repetition itself: nothing further is coming. You’ve now demonstrated that this is the explanation, that you don’t have another one, and that continuing to be confused is going to be awkward.
Which leads directly to the third thing.
Three: They’ll Say They Understand to End It
Third, and this is the one that costs real money.
Being re-explained to is a small status event. It communicates, unavoidably, that you consider them not to have grasped something you consider graspable.
Nobody enjoys that, and the exit is free: say you understand.
So you get the nod, and “no, that’s clear,” and the conversation closes — and you leave believing you resolved it, and they leave with the same gap they started with plus a reason not to raise it again.
This is why the misunderstanding surfaces two weeks later as a piece of work built on the wrong premise, and why it’s so genuinely surprising when it does. You have a specific memory of them confirming they understood.
They did confirm it. They were ending a mildly uncomfortable exchange, which is a different act.
The practical implication: a confirmation of understanding obtained after a re-explanation carries almost no information. If you need to know whether it landed, you have to get the evidence some other way.
Four: You’re Explaining, Not Diagnosing
Fourth, and this is where the fix lives.
When someone doesn’t follow, they didn’t fail to understand the whole thing. They lost it at a specific point, and everything after that point arrived without a foundation.
You don’t know where that point was. You’ve never asked.
So you re-explain from the top, which means the majority of your second attempt is spent re-delivering material that was already understood — and the listener sits through it, getting the signal that you think they missed all of it, waiting for the part they actually need.
That’s why the second explanation so often feels worse to both of you. It’s long, it’s mostly redundant, and it still doesn’t reach the break.
The move is to stop explaining and locate the break.
What Actually Works
Three things, in order of usefulness.
Ask them to say it back. Not “does that make sense” — that question has one polite answer and you’ll get it. Instead: “what would you do first?” or “how would you describe this to your team?”
The point where their version diverges from yours is the break. It’s usually much earlier than you’d have guessed, and it’s frequently a single term you used that means something different to them.
This one question replaces the entire second explanation, and it takes ten seconds.
Supply a distinction, not more detail. Once you’ve found the break, the effective repair is almost never more information. It’s a boundary: this, not that.
“It’s a review, not an approval.” “This is for the client, not for internal.” “You’re checking whether it’s true, not whether it’s good.”
That’s the instruction the Warring States persuaders gave for the listener who isn’t following: lead with distinction. Not with breadth, not with volume — with edges. They separated that listener from the merely uninformed one two thousand years ago, and the separation still holds. The nine types are here.
Take something out. If you must re-explain, make the second version shorter than the first. Cut the caveat. Drop the exception — you can add it back once the core has landed. Remove the analogy if it introduced a second domain.
The instinct is to add. The correct move is nearly always to subtract, and it feels like doing a worse job.
What This Is Not
It is not a claim that the listener is at fault. The opposite — the argument is that the failure is almost entirely on the transmission side, and that the standard repair makes it worse. Nothing here requires the other person to be slow.
It is not an argument against precision. In writing, precision is the whole job and every caveat earns its place, because a reader can re-read at their own pace. Live conversation has a working-memory constraint that a document doesn’t, and the same content behaves completely differently in the two channels.
It is not the same as dumbing something down. Distinction isn’t simplification. “It’s a review, not an approval” is more precise than a paragraph of context, not less — it just carries the precision in a form the listener can hold.
The Cost
Here’s the price, and it’s specific to competent people.
You have to accept that your best explanation was the problem.
The version you’re proudest of — the one with the context, the caveat, and the good analogy — is frequently the one that failed, and it failed because of the qualities you value in it.
That’s a difficult thing to hold if explaining well is part of how you understand yourself professionally. You’ll be asked to deliver something that feels sloppy, incomplete, and slightly imprecise, and watch it work better.
Second cost: asking someone to say it back feels like a test, and it makes some people uncomfortable, including you. There’s a version of the question that lands badly if the framing is off — “tell me what you understood” reads as an examination, while “what would you do first?” reads as planning. The wording matters more here than almost anywhere else in this material.
Common Questions
Why does explaining something more clearly sometimes make it worse?
Because “more clearly” usually means more elaborately — additional context, caveats and examples all add components. If the listener’s problem was already holding too many pieces, the improved explanation increases the load that caused the failure.
Why doesn’t repeating my explanation help?
Because the same arrangement of ideas supplies no new handhold, and the repetition itself signals that nothing further is coming. That combination tends to produce a premature “I’ve got it” rather than comprehension.
Why do people say they understand when they don’t?
Because being re-explained to carries a status implication, and agreeing is the cheapest exit. A confirmation obtained immediately after a second explanation carries almost no information about whether it landed.
What should I do instead of explaining again?
Find the break. Ask what they’d do first, or how they’d describe it to someone else, and listen for where their version diverges from yours. Then supply a distinction at that point rather than more detail.
Isn’t asking someone to repeat it back patronising?
It can be, depending entirely on framing. “Tell me what you understood” reads as a test. “What would you do first?” reads as planning and gets you the same information without the charge.
Should my second explanation be longer or shorter?
Shorter, nearly always. The instinct to add is the mechanism producing the failure. Cut the caveat, drop the exception, remove the analogy — and add them back later once the core has landed.
Where to Start
One conversation. This week. When someone doesn’t follow you, do not explain it again.
Instead, ask this:
“What would you do first?”
Then be quiet and listen for the divergence.
You’re not checking whether they understood. You’re locating a specific point, and the point will be earlier and smaller than you expect — usually a single word you used in a sense they don’t share.
Then repair that one thing with a distinction. This, not that. One line.
Don’t rebuild the explanation. Don’t go back to the top. Fix the break and stop.
It takes about a fifth of the time your second explanation would have taken, and it works, and it will feel to you like you barely explained anything at all.
I’ve spent twenty-five years working on how ordinary people develop capabilities that look impossible from the outside, and the person who can make anything understood to anyone is almost never the one with the best explanations.
They’re the one who worked out that explaining is what you do after you’ve found out where the other person stopped following, and that almost everybody skips that step and then wonders why the room went flat.