Atiba de Souza.Start a conversation
WritingMarketing

You Replied to Every Comment. That Is Why Nobody Is Talking.

By Atiba de Souza

An editorial cover image for a piece on You Replied to Every Comment. That Is Why Nobody Is Talking. The cover reads: MARKETING You Replied to Every Comment. That Is Why Nobody Is Talking.

The reply that looks like engagement

You know the moment. A notification lands. Somebody took the time to say something about your post, or ask a real question, or tell you their version of what you wrote about.

You type "Thanks!" You move on.

It feels like engagement. Look at your response rate. It is perfect. Nobody can accuse you of ignoring your people.

Here is what actually happened. You closed a door you had already paid to open.

A comment is not applause

A comment is not applause you are supposed to accept. It is somebody taking a turn.

They set something down in front of you. "Thanks!" is what you say while walking away with it. It is polite. That is the problem. It is polite enough that you never notice you left them standing there.

Every conversation that becomes a relationship starts the same way. Somebody hands the turn back.

The Volley: a reply that hands the exchange back instead of ending it. You never let the conversation end on your side.

The Volley System (CRQ)

The framework is three moves, in order.

C, Compliment. Say the thing you actually think is good. Not "great post." The specific thing. This move costs real attention. You cannot fake specificity and you cannot outsource it, because the whole point is proof that you were actually there. If you cannot find the specific thing, you did not read it, and the next two moves will show that.

R, Response. Refer directly to what the post was about. Read the caption first. Every time. The response has to be customized to that post, not to posts in general. It can also be the answer to the question the post raised. This one costs you the read. You have to go back and look at what you actually wrote and what they actually said, and that is the exact step the "Thanks!" reflex exists to spare you.

Q, Question. This is the move. Always end on an open-ended question. Open means it invites a full answer, not a yes or a no. Start from why, from how, from what if. This one costs the most, and I will say exactly what it costs, because it is the move everybody skips and nobody admits they skip.

If you only get one of the three right, make it the question. A specific compliment is worth something, but it does not move anything. The question is the only move that requires the other person to come back. The framework calls it the most important move and it is not close. Volleying in a conversation is what creates meaningful, and profitable, relationships.

stop here

swap the question for a pitch

They comment

C: Compliment

R: Response

Q: Open question

They answer

Dead end: the exchange closes on your side

Cold ask: the ask lands on top of the thread

The order in the reply is C, R, Q. The order in your head, the first time, is Q first. More on that in a second.

What the question actually costs

Here is where most people nod and change nothing. So let me say the uncomfortable part out loud.

You are not skipping the question because you do not understand it. You are skipping it because a question costs something "Thanks!" does not.

"Thanks!" lets you exit. That is what it is for. A question does not. A question puts a turn in somebody else's hands and waits, and you do not get to choose what comes back. They might say something you have to think about. They might say something that turns into work. They might say nothing at all, and now your question sits in a public thread with no reply under it, which is a small public nothing you have to be okay with.

You are not protecting your time. You are protecting the exit.

There is a second cost, and it is quieter. To write a real open question, you have to have read the post and the comment closely enough to ask something that is not generic. That means being present in a way "Thanks!" is designed to avoid. The habit is not laziness. It is armor.

So here is the rewiring, and it is smaller than it sounds. When the "Thanks!" reflex hits, do not fight it. Just write C, R, and Q before you send anything. Write the question first, even though it lands last in the reply. The reflex loses the moment the question exists on the page.

The decision, on the next comment

Not someday. The next comment you get.

Before you type a word, answer one question: is this exchange going to end on my side, or theirs?

If it is a volley, write the actual why, or how, or what if. Not a placeholder. The real one, the one you would ask if you and this person were the only two people in the room. Write it before you reply, then build C and R around it.

If it is not a volley, take the gate below and walk away with a clear conscience. Both are decisions. What is not a decision is "Thanks!"

Quick test before you send. If your reply ignores the caption and ignores the comment, and it could be copied onto any other post without changing a word, it fails R and it fails Q. It is a stamp, not a reply.

When not to volley

The rule is that you never let an exchange end on your side. That is not the same as writing three sentences at a bot that tagged you at 2am.

The Volley System is for people reaching toward you. Spam gets nothing. Somebody who came to fight gets nothing, because volleying with a person who is swinging only makes a longer fight. A comment with nothing in it does not owe you a manufactured question. Judgment is part of the system, not an exception to it.

Rank the three moves, too, because you cannot give all of them equal weight all day. The compliment is the cheapest and the least important. The response is the proof. The question is the whole game. If your time runs out, spend it on the question.

The ask that lands on top of the conversation

Now the part that ties this to the thing you actually want, which is business.

At some point a comment thread is where you meet someone worth talking to, and at some point you want to make an ask. A call. A DM. A next step. Most people drop the ask into the thread like an invitation out of nowhere. Hi, by the way, here is my link. Check this out.

It lands badly. Not because the ask is bad. Because of where you put it.

Before you make the ask, tie it back to whatever the two of you were just talking about, so the ask arrives inside the conversation instead of on top of it. An ask that ignores the conversation it sits in reads as an interruption, no matter how good the ask is. It is not the ask that is wrong, it is the placement of the ask. And here is the rule that governs the whole thing: social engagement earns you the right to ask only while you stay on the subject the other person opened.

That is the difference between a relationship and a pitch. The pitch resets the subject to yours. The relationship stays on theirs long enough to earn the change.

The comparison, so you can see it

The "Thanks!" reflexThe Volley System
Ends the exchange on your sideEnds the exchange on their turn
Measures your response rateMeasures whether a thread, the artifact, was actually built
Could be copied onto any comment unchangedFails unless it names the caption and the comment
The ask ignores the conversation it sits inThe ask arrives inside the conversation
Protects the exitSpends the turn

The failure that was already working

Now the part that should change how you look at your notifications.

My team spent two months trying to build a community on Reddit. It was a spectacular fail. I told them to decide when to shut it down. They decided to shut it down.

That same morning, a report landed. The business had been cited fifteen times in AI search, because of the very Reddit posts the failed experiment had produced.

Two months of work on a project that missed the goal it was measured on produced fifteen citations in AI search. Nobody on the team was measuring that. Nobody knew it was there.

And here is the part that should make you sit up. Had those posts been deleted when the community came down, the citations would have gone with them, and nobody would ever have known they existed.

That project failed at the goal it was measured on and succeeded at one nobody was measuring.

You are running the same measurement error on your comment section right now.

The goal you are measuring is "did I respond to every comment." You are crushing it. Every comment has an answer under it.

The thing you are not measuring is whether anything came out of it, and it does not live in the count. It lives in the artifact. In a comment section, the artifact is the thread. The back and forth. The thing the two of you build in public. That is where the value you cannot see yet is stored. "Thanks!" is how you make sure it never gets built.

The lesson out of the Reddit shutdown is not "never shut anything down." It is: kill the goal, not the artifacts. The team was right to end the experiment. They would have been wrong to take the posts down with it. You are running the smaller version of that mistake every single day, one green checkmark at a time.

The wrong answer was the bug report

Shaina Ballesteros wanted to know whether Athena could tell her, every week, when the newsletter got no replies. She assumed Athena had no access to that data. So she did the sensible thing. She asked Stephanie instead, and Stephanie sent her to me.

When Shaina finally asked Athena directly, Athena gave her a wrong answer. It said it had no access, when the truth was that Shaina had no access. And that wrong answer was the most useful thing in the meeting, because it was the only way I could find the missing rule and fix it.

Think about what would have happened without it. Because she routed around the tool instead of asking it, a broken rule sat undiscovered while she waited two weeks on stale content, and Stephanie and I both spent time answering a question the tool should have answered. The cost was not the wrong answer. The cost was the two weeks of not asking.

When I told her to bring it to me, I said it plainly. It is good, but that is why I need you to ask her. So that when you ask her and you get this answer, then I can help her be better to answer you next time, so she can be smarter.

Now bring that back to your comment section, because it is the same shape.

The comment you least want to answer is the bug report. The one that goes quiet after your "Thanks!" is the bug report. The reply that came back awkward, or angry, or off topic is telling you a rule of yours is broken, and it is the only way you find out. You will never see it by standing at the count.

Ask the person directly before you route around them to somewhere easier. And when the exchange answers you wrongly, when it goes nowhere, hand the exact exchange to the only person who can change the rules. That is you.

The bad answer is the bug report. Do not delete it.

Your next comment, not your next week

So here is where I land, and I want it small enough that you actually do it.

The next comment you get, before you send anything, ask the one question. My side, or theirs?

Then write the question first. Why, how, or what if. Build the compliment and the response around it. Send it when the question is real. If the comment is spam, a fight, or empty, take the gate and do not reply at all.

You spent money and time to open that door. "Thanks!" is how you close it and tell yourself you did not.

Hand the turn back.

<!-- cluster-links:begin -->

Related reading

<!-- cluster-links:end -->

Questions

What people ask next

Okay, but when you've got hundreds of comments coming in a day, how is anyone actually supposed to write a real, specific question on every single one without it eating your whole day?
You don't give all three moves equal weight all day. The compliment is cheapest and least important, the response is proof, and the question is the whole game — so when time runs out, spend what's left on the question and let the compliment go thin. And not every comment earns a volley in the first place; spam, fights, and empty comments get nothing, which frees up the time for the ones that do.
What do you do when you actually ask the open question and the person just never answers? Do you keep volleying on their next comment too, or does that start to feel like you're talking to a wall?
A question doesn't let you choose what comes back, and sometimes nothing comes back — that's the cost, a small public nothing you have to be okay with. It isn't a wall, it's just a turn that went unanswered; you still make the same decision fresh on their next comment rather than writing them off or chasing them.
If I'm asking a genuine question on every comment, how do I stop it from turning into its own kind of template that people can smell the same way they smell 'Thanks!'?
Run the test the article gives for the whole reply: if it ignores the caption and ignores the comment, and could be copied onto any other post without changing a word, it fails. A real question only exists because you read closely enough to ask something that isn't generic — the real one is the question you'd ask if you and this person were the only two people in the room.
How do you train someone else on a team to make the judgment call about who's worth volleying with and who isn't, instead of that just living in your own head?
Judgment is part of the system, not an exception to it, so the categories to hand off are the ones already named: spam gets nothing, someone swinging to fight gets nothing because volleying just extends the fight, and a comment with nothing in it doesn't owe a manufactured question. Everyone reaching toward you gets the full C, R, Q.
Where exactly is the line between staying on the other person's subject long enough to earn the ask, and just never getting around to making the ask at all?
The rule the article gives is that the ask only lands well if it ties back to whatever the two of you were just talking about — so the line isn't a timer, it's whether you can still connect the ask to their subject. Once you can genuinely tie the ask to what they opened, make it; past that, you're just stalling.
If the real payoff from that Reddit effort showed up somewhere nobody was tracking, how am I supposed to know whether this whole comment-replying habit is actually working or not?
Stop measuring response rate, which is exactly what the article says it's not about — measure instead whether an actual thread, a real relationship, got built. And hold the possibility the article ends on: some of the value from doing this right may not show up where you're looking for it at all, the way the citations showed up somewhere nobody was tracking.