Find Clients

Why Your LinkedIn Reach Dropped (And What Actually Moves It)

August 25, 2026 · Bradley Jacobs

Someone said this to me on a call and it stuck, because it's the most common version of the complaint:

I'm not getting any kind of interaction. 70 impressions, 39 impressions. My best friend clicks like, and that's it.

The instinct after a run like that is to blame the algorithm. Sometimes that's right. Usually it isn't, and the more useful question is whether you're measuring the right thing at all.

We backtested 895 real LinkedIn posts from 37 creators against actual engagement, with a held-out split so the patterns had to survive on creators the analysis never saw. What follows is that data, plus how I read post performance day to day.

First, you're probably measuring against the wrong number

Raw impressions tell you almost nothing. A post with 400 impressions is a disaster for one person and a great day for another. The number only means something relative to what you normally get.

So set a baseline. Take roughly your last 20 posts, weight comments about double a like, and work out what typical looks like for you at your follower count. Now every post is above baseline, at baseline, or below it. That's a real signal. "70 impressions" on its own isn't.

Two rules I follow with that number:

Wait 48 hours before you judge anything. Looking at performance in the first few hours tells you about timing, not about the post.

Don't make rash judgments off three posts in a week. Variance on LinkedIn is enormous. A run of quiet posts is usually a run of quiet posts.

What the data says actually moves reach

From the 895-post study, ranked by how well the evidence held up.

Specificity. Posts high on specificity, meaning concrete details, real numbers, named things, beat their author's baseline 68% of the time, against about 50% for vague ones.

Emotional pull. The cleanest lift in the whole dataset. Low emotional pull beat baseline 51% of the time, mid 60%, high 70%.

Format, more than most people expect. Beat rates by archetype: Journey or Milestone posts 72%, Story 60%, Confession 59%, Contrarian 57%, Insight 55%, Client story 55%, Promotional 52%, Question 50%. Personal-arc formats beat expertise dumps, consistently.

A number in the first line. Posts with a number or dollar figure in the first 160 characters beat baseline 61% versus 52%.

An image. 61% versus 54% without.

What actually hurts it

A link in the body. This is the biggest self-inflicted one. Posts with an external link in the body beat baseline 41% of the time, versus 56% for posts without. The fix takes five seconds: put the link in the first comment instead.

Very short posts. Under 400 characters beat baseline only 33% of the time, by far the worst bucket in the study. The advice to keep it punchy is wrong at the extreme.

Promotional framing. Promo-flagged posts beat baseline 53% versus 56%, and Promotional was the weakest real archetype at 52%.

And the one that surprised us: long posts don't underperform. Over 2,000 characters beat baseline 59%, one of the better buckets. Length only hurts when a long post has no payoff.

The thing polish won't fix

We also had an AI judge score every post on quality, blind to how it performed. Its single biggest error was over-crediting posts that were clean, well structured, and had nothing at stake.

Among 432 "Insight" posts, the correlation between the AI's quality score and how the post actually did was +0.02. Statistically nothing.

So if your reach dropped and your instinct is to write more carefully, that instinct is wrong. Polish isn't the variable. Specificity and stakes are.

How I actually read a post that underperformed

When a post lands below baseline I look at three things before I look at the algorithm.

Was there a real hook, or a warm-up? The first line has to earn the second.

Was it specific, or was it a category? "We grew fast" is a category. "Uber had 90% market share in six months" is specific.

Did it commit? Posts that commit to a single unhedged claim do better than posts that carefully cover every angle. Hedging reads as having nothing to say.

Most below-baseline posts fail one of those three. Very few fail because LinkedIn changed something.

When it really isn't you

Reach does move for reasons outside your control. Platform changes happen, audiences go quiet in August and December, and a post that lands during a news cycle gets buried.

The tell is the pattern, not the post. If every post over a month is below your baseline, something structural changed. If it's three posts, that's noise, and reacting to noise is how people talk themselves out of posting entirely.

What to do this week

Set your baseline. Last 20 posts, comments counted double. Write the number down.

Move your links to the first comment. Free, immediate, worth 15 percentage points of beat rate in our data.

Stop writing under 400 characters. The shortest bucket was the worst one.

Add one specific number to your next post's first line. 61% versus 52%.

Then wait two weeks before judging any of it.

If tracking a baseline by hand sounds like work you won't do, that's fair, and it's roughly why Mylance scores every post against your own rolling baseline and tells you which of the three things above it missed. But the method matters more than the tool: measure against yourself, move the links, and give it a month.

Keep reading: What Actually Works on LinkedIn: Lessons From 895 Posts