Short or long posts: the real reach/engagement trade-off

The short-vs-long debate on LinkedIn keeps spinning in circles because people compare the wrong metrics. Across 359,015 posts with at least 180 days of data, the answer is clear: length correlates strongly with median engagement — and the median doesn't lie.

In short

→ Median likes double between posts under 300 characters (median 12) and posts over 1,800 characters (median 25). → The progression is linear and monotonic: every additional length bracket gains in median engagement, without exception. → Short posts sometimes show a high average, but it's inflated by rare outliers. The median tells you what a typical post actually earns. → Long-form at 1,800+ also wins on comments — average 28 versus 14 for posts under 300 characters. → Long isn't riskier: it's more reliable.

Why median, not average?

The average lies. One viral post at 5,000 likes pulls the average up for its entire category — even if the other 99 posts around it each get 8 likes. The median tells you what a "normal" post in that category will actually get.

This is exactly the trap identified in the analysis of 516,000 LinkedIn posts: some creators showcase their best short posts to prove "short works." Statistically, they're exhibiting outliers. The underlying distribution tells a different story.

When you design a content strategy, you want to know what an ordinary post will do — not what a miraculous one does once every six months.

The raw data: length vs engagement (359,015 posts)

Measured across 359,015 posts from corpus §13.3 of the LinkPost study (9.6M snapshots, minimum 180 days of data):

| Length | Median likes | Avg comments | |---|---|---| | Under 300 chars | 12 | 14 | | 300–700 chars | 17 | 20 | | 700–1,200 chars | 19 | 25 | | 1,200–1,800 chars | 21 | 27 | | 1,800+ chars | 25 | 28 |

The progression doesn't stop. There's no plateau effect in the measured range: every additional length bracket gains a few points of median. Long-form at 1,800+ delivers both the highest median engagement and the highest average comment count.

Is long-form actually better, or just better-written?

That's the right objection. You could argue that creators who write long are also the ones who know their subject better — and that content quality is driving the difference, not length itself.

This selection bias is real. This study is observational: correlation does not imply causation, and sample bias is possible. Read these numbers as a regularity measured across a massive corpus, not as a proven causal law.

That said, two mechanical signals point toward long-form.

First, dwell time. LinkedIn measures time spent on a post before deciding whether to amplify it in the feed. A 1,800-character post mechanically forces a longer read — which sends a strong signal to the algorithm. Dwell time is one of the most underrated levers for LinkedIn reach: LinkedIn interprets every second of reading as a positive vote.

Second, convergence across datasets. The playbook across 438,413 posts confirms +49% engagement for posts over 1,500 characters compared to short posts. Two independent corpora — same trend. That's hard to ignore.

Why does everyone still believe short wins?

Because outliers are loud. A short post that explodes is visible, commented on, shared, remembered. A long post that quietly delivers solid median performance never makes waves. You remember the rare rockets — not the baseline distribution.

The other reason is simpler: writing 1,800 characters without padding is work. Many creators tried long-form once or twice, found themselves filling space, and concluded it didn't perform. The problem wasn't the length — it was the density.

What to do about it

→ Target 1,500 to 1,800 characters as your default work zone. That's where median performance becomes meaningfully better. → If you write short, reserve that format for ultra-specific topics with a strong punchline — you're playing the outlier game, not the median. → Don't mistake a one-off win on a short post for a reproducible strategy. → Before publishing, check your post's density with LinkPost's free analyzer — it flags posts falling into the lowest performance brackets across 300+ measured factors.

The guide on ideal LinkedIn post length details the use cases where short format still makes sense — they exist, but they're rarer than most people think.

The reach/engagement trade-off doesn't really exist in the data. Long wins on both dimensions. The real question is learning to write dense.

Observational study by LinkPost (corpus §13.3: 359,015 posts, 9.6M snapshots, minimum 180 days of data). Results are correlations. Selection bias possible. Methodology and limitations in the full playbook.