Your impressions fell off a cliff and you want to know what you did wrong.
Possibly nothing. That is the first thing to establish, and almost every guide on this topic skips it. "Here is what you did wrong" is a more satisfying article than "this happened to everyone".
Reach declined across LinkedIn in 2026. So before you change anything about how you post, find out whether your decline is the platform's or yours. The fixes for those two situations are completely different.
This guide separates them. Then it works through the six things that actually suppress an individual account, in the order that explains the most.
Key Takeaways
- Reach fell platform-wide in 2026, with independent analyses reporting anywhere from 25% down on engagement to 66% down on reach.
- Those sources disagree with each other, so treat any single precise figure carefully.
- If your decline sits inside that band, you are seeing a market change, not a penalty.
- If it is steeper, the most common causes in order are outbound links, topic drift, weak hooks, unsaveable content, and pod residue.
- Give any change six to eight weeks before you judge it. One week of LinkedIn data is noise.
Is It You, or Is It Everyone?
Start here, because everything downstream depends on the answer.
Independent analyses published through 2026 report substantial platform-wide declines. The figures vary a great deal depending on who is measuring, and what they are measuring:
- Engagement: down around 25%, by some accounts.
- Reach: down as much as 66%, by others.
- Follower growth and views: somewhere between the two.
We quote the range rather than a single number on purpose. When credible sources disagree this widely, the honest summary is a band, not a point.
What is not in dispute is the direction. Reach went down, broadly, for reasons that have nothing to do with your individual account.
So the diagnostic question is not "why did my reach drop". It is "did my reach drop more than everyone else's".
Here is how to answer that with your own data.
Open your LinkedIn analytics and export the last 12 months of post impressions. Split them in half: the six months before, the six months since. Compare the averages, not the peaks, because one viral post will distort everything.
Then read the result:
- Inside the 25% to 66% band. You are living through a platform change. Your posting is fine.
- Down 80% or more. Something specific to your account is contributing, and the rest of this guide is for you.
One honest limitation. LinkedIn's native export covers a rolling 365 days, so a longer baseline only exists if you have been exporting regularly. Our roundup of LinkedIn analytics tools covers the options. If you were a Shield user, our guide for Shield users explains what to export before you lose access to older history.
The Six Checks
- Platform-wide or just you
- Outbound links
- Headline and topic alignment
- The first three lines
- Saves instead of likes
- Pod residue and posting gaps
Work these in sequence. They are ordered by how much they typically explain, so if you find your answer at step two, you can stop.
Step 1: Separate a platform-wide decline from your own
Covered above. Pull 12 months, compare the halves, and place yourself against the reported band.
This step exists to stop you fixing a problem you do not have. That is the most expensive mistake in the whole process, because it usually means abandoning something that was working.
If you are inside the band, keep going. Adjust for the format and signal changes described in our guide to the 2026 algorithm. Do not panic.
Step 2: Audit your last 10 posts for outbound links
Count how many of your last ten posts carried a link in the body, and how many promised one in the comments.
Forbes reported in July 2026 that sharing an external link reduces visibility by approximately 60%. The same reporting found that putting the link in the first comment now carries that penalty too, instead of avoiding it.
That second part is the one that catches people. Link in first comment was the standard workaround for years. Every LinkedIn growth guide taught it, and it stopped working.
If seven of your last ten posts carried a link, you have found your answer before checking anything else. This is the single largest and most fixable factor on the list. Our full guide to links and LinkedIn reach covers what to do instead, and it is not "never share anything".
Step 3: Check whether your headline matches what you post about
Read your headline. Then read the topics of your last ten posts. Could a stranger predict one from the other?
Multiple 2026 analyses report that misalignment between your stated expertise and your content topics suppresses distribution.
The reasoning is intuitive. To justify showing your post to someone who has never heard of you, a ranking system needs a basis for believing you are worth reading on this subject.
This is the check people resist most, because the fix feels like a constraint on what they are allowed to say. It is not. It is a constraint on how many unrelated things you can say and still be findable. Our guide to building topic authority covers what drift looks like in practice and how to recover, and FeedBoss's Authority Map shows which topics you have actually established.
Step 4: Measure your first three lines against the see more cutoff
LinkedIn truncates a post at roughly 140 characters on mobile and 220 on desktop. Everything after that is behind a click that most people never make.
Take your last five posts and paste them into a post preview tool to see exactly where they cut off. Then ask an uncomfortable question about each one: if this was all I saw, would I click?
Two failure patterns dominate. The first is a wall of text with no line break before the cutoff, which reads as effort before it reads as anything else. The second is a hook that is still clearing its throat: context, background, a windup. By the time the interesting part arrives, the reader is gone.
This matters more in 2026 than it used to, because dwell time carries so much weight. Dwell time is simply how long someone spends on your post before scrolling on. A post nobody opens generates none of it, and that is the signal deciding whether it gets shown further.
Step 5: Look at saves instead of likes
Forbes reported in July 2026 that likes now carry minimal algorithmic weight while a save drives roughly five times the reach of a like.
So re-sort your recent posts by saves rather than reactions. Most people have never looked at this view of their own content, and it is usually the most informative ten minutes of the whole diagnostic.
If nothing is being saved, you have a content problem rather than a distribution problem. Your posts may be well written and pleasant to read and still give nobody a reason to keep them. Saveable posts tend to be things people will run later: a framework, a checklist, a set of numbers, a process. Likeable posts tend to be observations people agree with.
The fix is not to stop writing observations. It is to publish at least one thing a month that someone would want to find again.
Step 6: Check for engagement pod residue and posting gaps
Two separate things, grouped because they are both about patterns over time rather than individual posts.
Pods first. Forbes reported in July 2026 that LinkedIn's current system is built to detect engagement pods and make them entirely ineffective. It also cuts the reach of posts caught in one.
So a tactic that used to be merely inefficient is now actively harmful. If you are in a pod, leave. That is a single action with no downside, available today.
Then check your cadence. Look at your posting frequency month by month over the last year and find the gaps.
Consistency compounds, so gaps cost more than they look like they should. You are not only missing those posts. You are missing the steady signal that you are worth showing. A content calendar exists for exactly this, though a recurring reminder works too. The tool matters less than the habit.
Four Conclusions People Jump To, and Why They Are Wrong
- Assuming you were shadowbanned
- Deleting old posts
- Posting twice as often
- Judging a change after one week
Shadowbanning. LinkedIn does not use the term or confirm the practice. What is documented is that specific behaviors reduce distribution, and all of them are reversible by changing behavior. That is a far more useful frame than assuming a hidden penalty you cannot see or appeal.
Deleting old posts. The link penalty applies to a post's distribution while it is live, not as a permanent mark on your account. Deleting your archive undoes nothing and costs you the record of what worked.
Posting twice as often. This usually makes things worse. Doubling output while halving quality gives the ranking system more evidence that your posts get scrolled past. Frequency is not the lever people hope it is, and our guide on growing a LinkedIn following covers what actually moves it.
Judging after one week. LinkedIn data at current volatility is extremely noisy post to post. One good week means nothing. One bad week means nothing. Six to eight weeks on a rolling average means something.
What Your LinkedIn Analytics Can and Cannot Tell You
Every step above depends on your own numbers, so it is worth knowing what those numbers actually are before you build conclusions on them. LinkedIn's native analytics have four limits that catch people out.
The export window is a rolling 365 days. You can pull post-level impressions and engagements for the last year on demand, and nothing older. If you want a two-year baseline, you needed to be exporting a year ago. Start now regardless, because a year from today you will want it.
Impressions and reach are different numbers. Impressions count views including repeats, reach counts unique people. A post can hold impressions steady while reaching materially fewer humans, which reads as stability when it is not. If your impressions look flat but your comments have thinned, suspect this.
Dwell time is not exposed to you. The signal most responsible for distribution in 2026 is the one you cannot see. You can only infer it from second-order evidence: whether people who commented referenced something from the middle or end of the post, or only the hook.
Demographic data needs volume to mean anything. The job titles and companies breakdown is genuinely useful for checking whether you are reaching the right people, and it is close to noise under a few thousand impressions.
None of this makes the native analytics useless. It makes them a floor. Work with what they do give you, export regularly so you own your history, and treat any single post as an anecdote.
A Worked Example
Numbers make step one concrete, so here is the comparison as you would actually run it. Say you averaged 4,000 impressions per post over the six months to January:
| Now averaging | Decline | What it means |
|---|---|---|
| 1,800 | 55% | Inside the band. Platform-wide, not you. |
| 1,200 | 70% | The awkward middle. Run the checks. |
| 400 | 90% | Well outside. Account-specific. |
At 1,800 the decline is 55%, which lands inside the reported platform band. Your posting is not the problem. The right response is the format and signal adjustments in the 2026 algorithm guide, not a rewrite of your strategy.
At 400 the decline is 90%, well outside the band. Something specific to your account is contributing. Go to step two and work down.
At 1,200 you are at 70%, outside the reported band but not dramatically. Treat it as account-specific and run the checks. They cost an hour, and the alternative is guessing. If they all come back clean, you are probably in the tail of a platform-wide distribution rather than being penalized, and the answer is patience.
One more thing this example makes visible: use per-post averages, not totals. If you posted 40 times in the first period and 20 in the second, your total impressions halved for reasons that have nothing to do with the algorithm.
What to Track While You Fix It
- Impressions
- Saves
- Comments that received a reply
- Whether it carried an outbound link
- Number of posts published
- Rolling four-week average impressions
- Follower change
Keep this somewhere boring: a spreadsheet is fine. The point is not sophistication, it is having a baseline that predates your changes so you can tell whether they worked.
The one column people forget is whether a post carried an outbound link. Without it you cannot separate the link penalty from everything else you changed at the same time. That is the most common reason people never learn what actually fixed their reach.
Tick that box yourself, per post. For the rest of the picture, FeedBoss's analytics dashboard tracks engagement and format performance over time and compares any period against the one before it, from $29 per month.
Conclusion
Most reach drops in 2026 have one of two explanations, and they call for opposite responses.
If your decline matches the platform, the answer is patience and adjustment: same discipline, updated for what the current system rewards. Changing everything in response to a market-wide shift is how people abandon work that was actually fine.
If your decline is much steeper, it is almost always links, drift, or pods, in that order of frequency. All three are fixable this week, and none of them require posting more.
Whichever applies, give it eight weeks before you decide whether it worked. That is the least satisfying advice in this guide and the one most likely to matter.
Diagnosing a drop is easier when last month and this month sit side by side.
Start growing freeNo credit card required.
Frequently Asked Questions
Is the LinkedIn reach drop happening to everyone?
Largely yes. Independent analyses published through 2026 report platform-wide declines, though the figures vary widely, from around 25% on engagement to as much as 66% on reach. The disagreement between sources is itself useful information: treat any single precise number with caution.
How do I know if my reach drop is my fault?
Compare your own year-over-year decline against the reported platform band. Inside it, you are seeing a market change. Well outside it, something specific to your account is likely at play, and the six checks in this guide cover the most common causes.
Will deleting old posts with links restore my reach?
No. The reach penalty applies to distribution at the time a post is live, not as a lasting mark on your account. Deleting old posts does not undo anything and costs you the archive. Change what you do next instead.
How long does it take to recover LinkedIn reach?
Expect six to eight weeks of consistent posting before you can read a trend, because a single week tells you almost nothing at current volatility. Judge the change on a rolling four-week average rather than post to post.
Does posting more often fix low reach?
Usually not, and it can make things worse if the extra posts are thinner. Consistency at a cadence you can hold beats a burst you cannot. Two solid posts a week that people finish reading outperform five that get scrolled past.
Can a LinkedIn account be shadowbanned?
LinkedIn does not use that term or confirm the practice. What is documented is that specific behaviors reduce distribution, including coordinated pod engagement, comment bait, and outbound links. Those are reversible by changing behavior, which is a more useful frame than assuming a hidden penalty.
Does the LinkedIn reach drop affect company pages too?
More severely. Coverage in 2026 puts company page organic reach down 60% to 66% against 2024. Posts reshared by employees travel dramatically further than the same content from the page itself. If your company page is your main channel, that gap is the thing to act on.