How LinkedIn Posts Actually Get Engagement (And Why Yours Might Not)

LinkedIn reach dropped 34% on average in 2025, according to AuthoredUp’s analysis of 621,833 posts. Ninety-eight percent of users saw impressions fall year-over-year. And yet some people are growing fast. The gap between those two groups isn’t talent or luck. It’s mostly the first 140 characters of a post.

I’ve watched dozens of engineers and PMs treat LinkedIn like a broadcast channel. They write a polished paragraph, hit post, and wonder why it gets 12 reactions from teammates. The reason is mechanical, not personal, and it’s fixable.

Why the first line controls everything

On mobile, LinkedIn truncates your post at roughly 140 characters before showing a “see more” button. Over 72% of LinkedIn sessions happen on mobile, and 91% of engagement occurs there (AuthoredUp, 2025 dataset). That means your first 140 characters are doing the job your whole post thinks it’s doing.

Most people write an opener like: “Excited to share some thoughts on my career path and what I’ve learned about professional growth along the way.” That’s 109 characters, zero tension, and no reason to tap “see more.” Every potential reader just keeps scrolling.

The opener that works creates a gap the reader wants to close. “I got rejected from Google three times. Here’s what changed on attempt four.” Or: “Your post got 47 impressions. Here’s the one sentence that caused it.” Specific, slightly uncomfortable, and unresolved.

Notice that neither of those requires the post to be long or clever. They just need to make the reader mildly uneasy about not knowing what comes next.

What the algorithm actually rewards

The platform weights comments at roughly twice the algorithmic value of a like, based on AuthoredUp’s 2025 analysis of 994,894 posts across 63,407 profiles. Comments longer than 15 words carry even more weight than brief replies. Indirect comments (replies-to-comments in the thread) generate up to 2.4x the reach boost of a standard comment.

This matters practically: a post with 3 likes and 1 real comment showing in someone’s feed outperforms a post with 30 likes and no comments. The algorithm treats the comment as evidence that a human thought it was worth engaging with rather than just double-tapping while half-asleep.

Dwell time is the other signal people underestimate. If someone opens your post and reads for 45 seconds before scrolling away, that registers differently than a quick scroll-past. Long-form text posts (1,300 to 2,500 characters) generated 27% higher median engagement than posts under 400 characters in the same dataset. The mechanism is probably dwell time, not length itself.

External links, though, do the opposite. Posts with a link out to an article or website consistently receive 40-60% fewer impressions because LinkedIn’s distribution system deprioritizes anything that takes users off the platform. If you want to share a link, put it in the first comment.

Format: the part nobody wants to admit matters

Formatting feels cheap. “I don’t want to write LinkedIn content that looks like LinkedIn content” is something I hear a lot. I understand the reluctance. But wall-of-text posts perform consistently worse than formatted ones, and the reason isn’t aesthetic. Mobile readers scan before they commit. If the first three lines are a dense paragraph with no visual break, most readers bail before giving the content a chance.

What actually works, structurally:

• One short sentence as the first line (5-12 words, no comma)
• A single blank line
• Your actual point in 2-3 short paragraphs, each separated by a blank line
• A question or an incomplete thought at the end that invites a comment

That’s it. You don’t need bullets, emojis (those are a style choice, not an algorithm factor), or numbered lists. The blank lines just give mobile readers a visual anchor so they don’t feel like they’re committing to a wall of text.

Content topics that actually generate discussion

A few patterns show up consistently among posts that generate comment threads worth reading.

The honest failure. “I applied to 67 companies over 4 months and got 3 offers. Here’s what I learned.” This works because it’s specific (67, not “many”), it has a resolution, and most people who read it have failed at something similar. They comment because they recognize themselves.

The unpopular opinion. “I think daily LeetCode practice is mostly harmful for engineers who’ve been working more than 3 years.” It might be wrong (it probably is wrong in some cases). But it creates a real discussion because people have a genuine stake in whether it’s true.

The process reveal. “Here’s exactly how I structured my system design prep for my Meta interview last month.” Specificity again. Last month, not “recently.” Meta, not “a FAANG company.” If someone is prepping for Meta right now, they’ll read every word and probably comment.

The least reliable category is the “hot take.” Vague contrarianism (“Most resumes are terrible and here’s why”) generates engagement but not usually the kind that helps your professional network. I’m genuinely unsure whether that tradeoff is worth it for most people. If you’re building a personal brand in content creation, maybe. If you’re a software engineer trying to be visible to hiring managers, probably not.

Frequency: how often is enough

LinkedIn’s own research suggests posting at least once a week produces roughly 2x the engagement of less frequent posting, across company pages. For individuals the pattern is similar but less pronounced.

Three posts a week is the number most practitioners suggest, and I think that’s right for someone actively building visibility, maybe while in job search mode. But one genuinely good post a week beats three mediocre ones. The algorithm responds to engagement rate, not volume. A post that gets 4 comments from 200 impressions is more valuable to your next post’s distribution than a post that gets 1 like from 500 impressions.

Best times are Tuesday through Thursday, 8-9 AM in your audience’s timezone. This isn’t magic; it’s just when professionals are checking their feed before the first meeting. The first 60-90 minutes after posting set the trajectory. If early engagement is flat, the algorithm deprioritizes distribution quickly.

The profile behind the post

A post from a profile with no photo, no headline beyond “Software Engineer at [Company],” and 43 connections gets half the distribution of the same post from a complete profile. LinkedIn’s system treats profile completeness as a trust signal. This is one of those facts that’s slightly uncomfortable because it means the platform rewards people who’ve already invested time in it. But it’s real and it’s worth fixing before optimizing your content.

Your headline (which shows under your name in every post) does more work than most people realize. “Senior Engineer | ex-Google | Helping engineers pass FAANG interviews” gives a reason to follow before anyone reads your content. “Software Engineer at Acme Corp” doesn’t.

The same thinking that goes into a strong resume headline applies here: lead with the value, not the title. If you want more attention from hiring managers, your headline should say something a hiring manager would care about. The ATS and LinkedIn headline follow the same logic: keyword specificity over generic role descriptions.

What to do this week

Pick one thing you failed at recently in your job search, your work, or your interview prep. Write three sentences: what you tried, what happened, what you’d do differently. Put the outcome in the first sentence, not the last. Post it Tuesday morning. Don’t put a link in it. End with a question.

That’s a more useful experiment than reading another article about the LinkedIn algorithm. Including this one.