How the new ranking works
LinkedIn now ranks comments by relevance to each viewer rather than strict chronological order, using signals like professional interests, connections and engagement activity. You see the replies most likely to matter to you first, not the ones posted most recently.
The change follows a reported 18% year-over-year rise in time spent in post comments, which sounds promising until you read what LinkedIn found when it examined the quality of that engagement.
The AI comment problem
Pangram's analysis of public LinkedIn posts between April and June 2026 found 30% of comments were entirely AI-generated. LinkedIn analysed 57,000 public posts and confirmed the scale of the problem internally before shipping the relevance ranking update.
This is not about helpful AI assistance. LinkedIn started ranking comments to fight back against a wave of AI spam, the generic replies that add nothing but game engagement metrics. Chronological sorting made that easy. Post a bland "Great insight!" within seconds, and you sat at the top of the thread. Relevance ranking kills that incentive.
The same Pangram study found 41% of long-form LinkedIn posts were flagged as fully AI-generated over the same period. When both posts and the comments beneath them are increasingly machine-written, the platform has an authenticity problem, not just a ranking one.
What this changes for engagement strategy
Relevance ranking rewards substance over speed. The first reply no longer holds privileged visibility. A comment that adds context, answers a follow-up question, or connects to a reader's professional interests will surface higher than a vague affirmation posted seconds after publication, even if that affirmation came first.
Comment quality now carries significantly more weight, with the algorithm detecting whether replies meaningfully contribute to the topic. If you have been coaching teams to comment early for visibility, that guidance is weaker now. Coach them to comment usefully instead.
This also changes what "good engagement" looks like in your reporting. A post with 40 comments is not automatically performing better than one with 15 if half those 40 are generic and suppressed by relevance ranking. Track comment depth and replies to comments, not just volume.
LinkedIn has been tightening comment enforcement for months. The platform now blocks hundreds of thousands of automated comment attempts daily, part of a broader crackdown on engagement pods and AI-generated activity that accelerated over the summer.
What to do differently
First, stop optimising for speed. Being first in the thread no longer guarantees top placement. Your reply needs to be relevant to the reader seeing it, which means it needs substance. If you cannot add something specific, wait until you can, or skip it entirely.
Second, write comments that other professionals in your niche would actually find useful. The relevance signal is built on professional graph matching. A thoughtful reply from someone in the same industry or with adjacent expertise will rank higher than a cheerleader comment from outside your domain.
Third, if you are running employee advocacy or executive visibility programmes on LinkedIn, audit what your people are actually posting as comments. Template replies and one-word affirmations hurt more than they help now. Building credibility through genuine expertise matters more than hitting a comment quota.
LinkedIn's move mirrors a broader platform trend. Instagram and Reddit both shifted to visual or threaded comment formats this year, prioritising depth over volume. The read-only comment section is over. Platforms want conversations, not performative engagement.
Comments are now ranked by relevance to each viewer rather than shown in strict chronological order, rewarding replies that are genuinely useful to the reader.
FINN Partners, Boom Scroll social media digestThe bigger shift underneath
Relevance ranking is a symptom, not the disease. Time spent in post comments rose 18% year over year while overall content consumption increased 10%, which tells you comments are becoming a bigger share of how people use LinkedIn. The platform is responding by making comment threads more usable and harder to game.
But the AI saturation rate forces a harder question. If 30% of comments and 41% of long-form posts are fully machine-generated, what does "engagement" actually measure? Marketers optimising for comment count may be optimising for bot activity without realising it.
LinkedIn is not unique here. Every platform with a feed is dealing with the same problem. What makes LinkedIn's version harder is that professional credibility lives in those comment threads. A generic AI comment on Instagram is annoying. On LinkedIn, it erodes trust in the entire platform as a place where real professionals share real expertise.
The relevance ranking update is an admission that volume-based engagement metrics broke under AI load. Platforms that survive the next two years will be the ones that figure out how to surface signal through the noise. LinkedIn just made its first real move.