Why the rate ranking and the likes ranking disagree

Read down the median-likes column and you will notice it does not march in step with the rate column. That mismatch is the most important thing on this page, so here is the mechanism in plain terms.

Engagement rate versus followers falls mechanically as accounts get bigger. The denominator is the follower count, and a 100k-follower account would need 100 times the likes of a 1k account just to post the same rate, which never happens because reach does not scale with followers. So any topic written about disproportionately by large accounts shows a lower engagement rate, not because the content is worse, but because the arithmetic divides by a bigger number. This is the same size effect spelled out in the pillar on average engagement rate, where the median rate vs followers drops from 1.39% under 1k followers to 0.13% above 100k.

That is exactly why every row above carries median likes and a post count, and the disagreements are easy to spot:

• Career Development sits near the very bottom on rate (0.24%) yet posts a median of 50 likes, more than Graphic Design's 33 at the top of the table. The rate is low because career content is written by larger, more established accounts; the raw reach is fine.

• AI lands second-to-last at 0.24% but earns 42 median likes, again above the topic that leads on rate. A typical AI post collects more likes than a typical graphic-design post; it just collects them against a far bigger follower base.

• Work Culture carries the highest median likes in the table at 71, while ranking only mid-pack on rate (0.51%). High absolute reach, average efficiency.

The opposite case exists too: Graphic Design wins on rate (1.16%) with a modest 33 median likes, which is the signature of a topic posted mostly by smaller accounts whose every post lands hard against a tight follower base. Neither ranking is "the truth." Rate tells you efficiency for your size; likes tell you raw reach. Judge a topic on both, and never on rate alone.

Where do you stand on these numbers? MagicPost's LinkedIn analytics compute your engagement rate per post against your own followers and your own history, so you can see how your topics perform for your account instead of an internet median. And if you want these reference numbers refreshed without redoing the math, the MagicPost Benchmark recomputes them on fresh data and delivers them to your inbox every month.