Studies · 684,000 LinkedIn posts · June 2025 to September 2026
Until July 2026, “Comment KEYWORD” posts did a little better than their author's average post. Then it flipped. Within two weeks, they were down to about half the impressions of an average post.
Naïlé Titah6 min read
The essentials
Until late July 2026, “Comment KEYWORD” posts did a little better than their author's average post: 9% more impressions. And 83% more comments.
The edge disappeared in the week of July 27. Not one week since has climbed back to the average post.
Same creators, before and after: 60 of 79 saw their keyword posts drop. Their other posts slipped just 7%.
These posts now reach 62% fewer people than an average post. Comments? Back to average.
A closing question stayed exactly where it was: 4% below an average post, before and after July 27.
Read the numbersEvery + or − percentage compares a post with the same creator's average post from 30 days before to 30 days after it. “+9%” means 9% more impressions than that creator usually gets.
Chapter 1The format
It's a post that ends by asking readers to comment one word to get something: a guide, a template, a PDF. 79 creators used it both before and after July 27. 60 of them saw their keyword posts drop. Three examples:
Before July 27
After July 27
Comment "DECK" and I'll send you the full rubric.
Chapter 2The break
Yes, and it wasn't a slow slide. For 14 months, these posts stayed around their author's average post, from 4% below to 20% above. Then August 2026 hit: 56% fewer impressions. And no, it's not a summer thing. In August 2025, the same format was still 11% above average.
Table 1 · “Comment KEYWORD” posts vs the creator's average post, month by month16 months
| Month | “Comment KEYWORD” | Closing question | Keyword posts |
|---|---|---|---|
| Jun 2025 | +15% | +1% | 629 |
| Jul 2025 | +19% | +1% | 879 |
| Aug 2025 | +11% | 0% | 822 |
| Sep 2025 | +11% | +2% | 864 |
| Oct 2025 | +20% | +4% | 966 |
| Nov 2025 | +15% | +1% | 1,044 |
| Dec 2025 | +13% | +1% | 890 |
| Jan 2026 | +14% | +1% | 635 |
| Feb 2026 | 0% | −2% | 509 |
| Mar 2026 | +20% | −2% | 549 |
| Apr 2026 | +20% | −3% | 455 |
| May 2026 | +7% | −7% | 379 |
| Jun 2026 | −4% | −6% | 440 |
| Jul 2026 | +5% | −2% | 372 |
| Aug 2026 | −56% | −3% | 185 |
| Sep 2026 | −56% | −5% | 58 |
“Keyword CTAs killed my reach.”
Tiffany Spahl-Nally (21K followers), first line of her post on September 14, 2026.See the post
Chapter 3Week by week
The week of July 27, 2026. The week before, keyword posts were still 12% above an average post. That week, they fell to 39% below. And every week since has landed between 47% and 62% below.
Table 2 · The weeks around the break10 weeks
| Week of | “Comment KEYWORD” | Keyword posts |
|---|---|---|
| Jul 6 | +11% | 81 |
| Jul 13 | +17% | 83 |
| Jul 20 | +12% | 96 |
| Jul 27 | −39% | 61 |
| Aug 3 | −58% | 30 |
| Aug 10 | −56% | 43 |
| Aug 17 | −53% | 54 |
| Aug 24 | −52% | 43 |
| Aug 31 | −62% | 28 |
| Sep 7 | −47% | 18 |
Chapter 4Same creators
The format. Put together, the keyword posts of those 79 creators went from 15% above an average post to 48% below. Everything else they posted? Down just 7%. Each line below is one of the 79:
60 went down19 went upAll 79 creators: +15% → −48%
Each line is one creator who used “Comment KEYWORD” on both sides of the break: their keyword posts compared with their own average post. Each step on the scale doubles. 6 creators beyond 16 times or a sixteenth of their average are drawn on the edge.
You can see it in the public comment counts too. Chris Tottman again, every keyword post from June to August:
Chris Tottman150K followers · comments on each keyword post
Median before July 27: 487After: 24
Same story across the keyword posts of every creator we track:
59.5 → 9.5
median comments per “Comment KEYWORD” post
44% → 21%
of those posts had more comments than likes
Raw counts, not compared with each creator's average: 300 posts from 85 tracked creators (Jun 12 to Jul 26) vs 240 posts from 77 (Jul 27 to Sep 19).
Chapter 5What it lost
Mostly reach. Since July 27, a keyword post reaches 62% fewer people than its author's average post. Likes are down 37%. Comments stayed close to average, 2% below. Readers still comment to get the file. Far fewer people get to see it.
Quick note. Impressions count every time LinkedIn shows the post. People reached counts each person once. Before July 27, a keyword post pulled 83% more comments than usual and reached 14% more people. Since then, the comment count is back to normal and reach sits well below it. More comments no longer go with more reach.
Chapter 6What held
Two of them. A closing question didn't budge: 4% below an average post, before and after July 27. A link in the first comment stayed put at 11% below. Neither adds reach. But neither costs more than it used to. The plain asks did drop, just not as far as the keyword. A simple request for comments or a “♻️ Repost” went from about average to around 20% under.
Closing question−4%before and after Jul 27
Jasmin AlićP.S. What's ONE LinkedIn habit you swear by?
Who took a chance on you before your numbers earned it?
Marina PanovaWhat does your version look like?
Link in the first comment−11%since Jul 27, as before
Laszlo BockI’m excited to join Seedglass as cofounder.
Early access link in the comments.
Comment or repost request, no keywordabout −20%since Jul 27
Ankur Warikoo"I cannot focus. My mind works for 30-35 minutes, but then it starts wandering.
Let me know in the comments.
Eric PartakerAt McKinsey, I advised Fortune 50 CEOs on how to run their companies.
♻️ Save this for later and repost to help others
Table 3 · Each call to action vs an average post5 calls to action
| Call to action | Before | After | Posts before / after |
|---|---|---|---|
| “Comment KEYWORD” + I'll send it | +13% | −49% | 1,660 / 207 |
| “Comment KEYWORD”, no promise | +3% | −58% | 992 / 88 |
| Closing question | −4% | −4% | 18,681 / 4,349 |
| Link in the comments | −13% | −11% | 2,381 / 516 |
| “DM me” | −16% | −26% | 4,321 / 975 |
Three moves, each backed by a number from this study.
Keyword posts now get about half their author's usual impressions. Keep the resource. Just change how people get it.
It didn't move: 4% below an average post, before and after July 27. You still ask for replies, just without the keyword's drop.
A link in the comments sits 11% below an average post. Same as before July 27. A keyword post now gets about half. Schedule the comment with the post, so it goes live on its own.
Schedule LinkedIn comments with your post →One example: the same lead-magnet post, before and after the break. Only the last line changes. The link moves to the first comment.
I rewrote my pitch deck three times before investors actually read it.
Comment “DECK” and I'll send you the full rubric.
I rewrote my pitch deck three times before investors actually read it.
Which slide do investors skip first?
↳ First comment
684,000 posts
From 16,000 creators, published June 2025 to September 2026. We left out company pages, accounts under 100 followers and posts under 3 days old.
We compare each post with the same creator's average post: the mean, on a log scale, of their plain posts from 30 days before to 30 days after it. A random sample of plain posts lands at 0% against this baseline. Confidence intervals come from resampling creators. We checked by hand the posts that ask for a keyword and promise to send something: 119 of 120 were correct.
Comment bait on LinkedIn is a post that asks readers to comment in exchange for something, most often a keyword: “Comment GUIDE and I'll send it.” Between October 2025 and July 2026, 1.5% to 1.9% of the posts MagicPost tracks used the keyword version each month. In September 2026, 0.5% did.
“Comment KEYWORD” posts have gotten about half the impressions of their author's average post since the week of July 27, 2026 (52% fewer), according to MagicPost data on 684,000 LinkedIn posts. The data shows when the drop happened and how large it is. MagicPost doesn't see how LinkedIn ranks posts, so it can't say why.
No: the comment bait drop is not seasonal. Each post is compared with the same creator's posts from 30 days before to 30 days after it, so a summer dip hits both sides. In August 2025, “Comment KEYWORD” posts were still 11% above an average post.
Publishing natively or through a scheduling tool makes no measurable difference to the comment bait drop. Posts published directly on LinkedIn went from 9% above an average post to 52% below. Posts published through MagicPost went from 14% above to 54% below.
“Comment KEYWORD” posts now get about as many comments as an average post (2% fewer), down from 83% more before July 27, 2026. They also reach far fewer people: 62% below an average post.
A closing question is the safest choice: it kept its level after July 27, 2026, at 4% below the author's average post before and after, in MagicPost data on 684,000 LinkedIn posts. A link in the first comment sits 11% below. A “Comment KEYWORD” line now costs about half.
Charts and numbers are free to reuse with a link to this page.
MagicPost (2026). “LinkedIn Comment Bait Now Gets Half the Reach.” Study of 684,000 LinkedIn posts. magicpost.in/studies/linkedin-comment-bait
