Yes, LinkedIn comments work — and for most people they work harder per minute than posting does. Not because of a hidden algorithm trick, but because of where a comment puts you. When you publish, you are asking your own network to show up. When you comment on someone else's post, you are stepping into a room that has already filled up. The attention is spent; you are borrowing it. That is a structural advantage, and it is available to anyone with 500 followers. What follows is the mechanism, the difference between a comment that earns a profile visit and one that gets scrolled past, and the parts nobody sells you — the ceilings, the pods, and why comment automation quietly makes you worse.
Posting asks your network to assemble. Commenting joins a crowd that already has. A substantive comment gets read by people who came for the topic, and the next thing they do is check who said it — so your profile is the actual conversion page, not your post. What separates a working comment from noise is knowing something the author didn't say. The limits are real: it doesn't scale, engagement pods violate LinkedIn's user agreement, and automated comments read as exactly what they are.
Why LinkedIn comments outperform posts per minute of effort
A post is a cold start. Every time you publish, you are asking LinkedIn to find people who might care, from scratch, and you are competing with everything else in their feed. Some days that works. Most days it is a coin flip you do not control.
A comment skips the cold start. Somebody else did the hard part: they wrote something, it caught, and a few hundred or few thousand people are now reading it with the topic already loaded in their heads. You are not interrupting them. You are adding to a thing they chose.
LinkedIn does not publish how its ranking works, so I will not pretend to know the weights. What I can describe is the mechanism you can observe yourself: comments appear underneath the post, sorted with the ones that got replies and reactions near the top; a comment that people engage with stays visible longer to more of that post's readers; and your name and headline sit next to it the entire time. Nothing about that requires a theory of the algorithm. It is just where the eyeballs already are.
There is a second, slower effect. Comment thoughtfully under the same three or four people for a month and their audience starts recognizing your name before they have ever read a word you published. Recognition is cheaper to earn in someone else's comments than in your own feed.
The funnel: comment, profile visit, follow, conversation
Here is the part people skip. A comment is not the conversion. It is the first step of a four-stage path, and each stage has a different job.
The failure mode is almost always stage two. Someone reads a sharp comment, taps your name out of curiosity, lands on a headline that says "Helping businesses unlock growth" — and leaves. You did the hard part and lost them on the easy part.
| Stage | What happens | What it depends on |
|---|---|---|
| Comment | You show up in a thread the reader already chose to read | Having something specific to add |
| Profile visit | They tap your name to see who said that | The comment being good enough to raise a question about you |
| Follow | They decide you are worth hearing from again | Your headline, About section, and last three posts agreeing on one thing |
| Conversation | They message you, or reply, or reference you months later | Repetition, and a profile that makes the next step obvious |
What each stage is actually for
Your profile is the landing page, not your post
If commenting is the traffic source, the profile is where the traffic goes. Treat it like a landing page and it behaves like one. Treat it like a resume and it converts like a resume.
Three things do most of the work. The headline should say what you help with and for whom, in plain words, not in achievement language. The About section should read like the first paragraph of a good post — a specific claim, not a career summary. And your featured or pinned content should match the topic you comment about, because a visitor arriving from a thread about pricing does not want to land on a post about your company retreat.
If you want a second opinion on yours, we built a free LinkedIn profile checker that reads it the way a stranger arriving from a comment would. It is blunt. That is the point.
The order matters more than the effort. Fixing the profile takes an afternoon and improves every comment you leave afterward. Leaving 200 comments against a weak profile improves nothing.
What separates a comment that works from "Great post!"
Here is the test I use, and it is the only one I trust: could someone who had not read the post, and did not know what I know, have written this comment? If yes, it is noise. Delete it.
Agreement is noise. Summary is noise. Praise is noise — pleasant, forgettable noise. Compliments plus a question are still noise wearing a hat. What is not noise is information: a counterexample, a number from your own work, a case where the author's advice breaks, a piece of context they left out, or a disagreement stated without theatrics.
Length is not the variable. A two-line comment that names a specific exception beats six paragraphs of agreement. But the two lines have to cost you something to write.
- Add the exception. "This holds for B2B SaaS. It inverts for agencies, because the buyer is the same person who does the work." You just gave the audience something the post didn't.
- Bring the receipt. Not a borrowed statistic — your own experience, stated at whatever confidence it deserves. "We tried this for four months and it didn't move anything" is worth more than a citation.
- Extend one step. Take the author's idea and follow it to the consequence they stopped short of. This is generous and it makes you look like the person who has thought about it longest.
- Disagree cleanly. Name what you think is wrong, say why, stop. No "respectfully" throat-clearing, no dunking. Clean disagreement is the fastest way to be remembered.
- Answer the question in the replies. If someone in the thread asked something the author hasn't answered, and you know the answer, answer it. Most people are competing for the author's attention. Very few help the audience.
The honest ceiling: this does not scale, and that is the point
Every article about social selling eventually promises leverage. This one cannot, because the thing that makes a comment work is the thing that cannot be multiplied: you knowing something.
Ten good comments a day is a lot. Fifteen minutes of real reading and thinking, most days, is the realistic ceiling for someone who also has a job. If a workflow promises fifty, the comments are not good — they are just numerous.
That constraint is also the moat. The reason commenting still works while so much of LinkedIn has been flattened by generated content is that a genuinely informed comment is expensive to fake. Anything that makes it cheap makes it worthless at the same time.
The only comment worth leaving is one you could not have left without knowing something.
Three shortcuts that quietly cost you: pods, bots, and the reply-guy trap
Since the strategy is slow, an industry has grown up around pretending it isn't. All three shortcuts share a shape: they produce the metric and destroy the thing the metric measured.
Engagement pods. Groups that agree to comment on each other's posts on cue. Beyond whether they work, they are against LinkedIn's User Agreement, which prohibits arrangements to artificially inflate engagement — and accounts have been actioned for it. Worth reading LinkedIn's own terms rather than taking my word for it. There is a longer explanation in our engagement pod glossary entry. The subtler cost is that pod comments are written by people who did not read the post, which means they read exactly like what they are.
Automated commenting. Tools that watch a feed and post generated replies. I will be against my own interest here, because Liftli sits adjacent to this category: the automated end of comment tooling produces precisely the generic filler this entire article is about avoiding. A model that has read the post but knows nothing about your work can only produce plausible agreement. Plausible agreement is the noise. Assistance that helps you find threads worth your time and organize what you know is useful; generation that replaces the knowing is not.
The reply-guy trap. Commenting on everything a big account posts, fast, with nothing to say. It is visible, it feels like progress, and it builds the wrong recognition — people learn your name as the person who is always there, not the person who is worth reading. See reply guy if you want the full anatomy.
| Comment that earns a profile visit | Comment that gets scrolled past |
|---|---|
| Names a case where the post's advice fails | Restates the post's main point |
| Includes something only you would know | Includes something anyone could write |
| Readable on its own, out of context | Meaningless without the post above it |
| Written after reading the whole thing | Written after reading the hook |
| Ends where the point ends | Ends with "Thoughts?" |
Same thread, two comments
How to run this as a real linkedin lead gen habit
Commenting turns into inbound leads the slow way: someone reads you enough times that when the problem you talk about becomes their problem, you are already the name attached to it. You cannot compress that. You can make it deliberate instead of accidental.
What that looks like in practice is a short, fixed list rather than a feed. Pick fifteen to twenty accounts: five or so people your buyers already read, five actual buyers, five peers, and a couple of people whose posts reliably generate real discussion in the replies. Check them at a set time. Comment where you have something. Skip where you don't — skipping is a feature, not a failure.
- Build the list. Fifteen to twenty accounts, saved as a real list, not a vague intention to browse.
- Set one window. Ten to fifteen minutes, same time daily. Commenting expands to fill whatever you give it.
- Comment only where you have material. Three real comments beat ten reflexive ones. On days you have nothing, leave nothing.
- Come back to your own threads. Replying to the replies under your comment is where conversations actually start, and almost nobody does it.
- Move to DMs slowly. When someone engages with you repeatedly, a message referencing what you actually discussed is welcome. A pitch two hours after a first comment is not — that is the behavior that gave social selling its reputation.
Where a tool helps, and where it can't
A tool can shorten the search. Finding the threads worth your fifteen minutes, remembering which conversations you are already in, keeping your commenting pointed at the same few topics as your posts — that is administration, and administration is worth automating.
A tool cannot supply the substance. I build one, so take this with the appropriate salt: Liftli is an AI head of content for LinkedIn, X and Substack that runs inside Claude through MCP. It drafts from your own raw material in your extracted voice, and it never publishes without your one-tap approval. It has comment planning and engagement planning tools. They help you decide where to spend attention and shape what you already know. They cannot know your customers for you, and a comment planned on an empty input will be as empty as one you wrote yourself in a hurry. Free tier, paid from $29.
If you just want to see the shape of a decent comment before writing your own, our free comment generator is a reasonable starting point — with the same caveat: use it as a structure, then put something of yours in it. Founders and independent consultants can find the fuller workflow on for founders and for consultants.
The uncomfortable summary is that the highest-return activity on LinkedIn is unglamorous, uncountable, and cannot be delegated. Which is exactly why it still works.