The post that pulled three thousand views last year now stalls at three hundred. You did not change anything, and that is exactly the problem, because the quiet conclusion creeping in is that you lost your touch. You did not. LinkedIn rebuilt the machine that decides who sees what, and the new machine rations reach on rules most people on the platform have not read.
By one widely cited annual analysis of millions of posts, average LinkedIn reach fell roughly fifty percent year over year. Engagement and follower growth bent down with it. This is not a few unlucky creators. It is the whole feed.
Here is the part that should change how you respond to it: for most people, the drop was deliberate, not a punishment.
For years the feed ran on a patchwork of narrow ranking systems, each scoring a separate signal. In early 2025 LinkedIn’s research team published a paper describing a single large language model, built to replace that patchwork and rank the entire feed on its own. It is a foundation model, the same broad technology as the chat assistants you already use, pointed at deciding what belongs in your feed.
The practical shift is everything. The old system counted clicks and hashtags. The new one reads language. It looks at your profile, your post, the topic, and the reader’s history, and infers whether the two actually match. If a profile reads as a supply-chain consultant and the post is a generic motivational quote, it sees the mismatch and holds the post back. If a specific, useful take lands in front of the exact people who care, it pushes harder. The feed went from a popularity contest to a relevance match.
So the reach you lost was mostly reach that was never going to do anything for you. LinkedIn would rather show your post to five hundred people who care than five thousand who scroll past. That reframes the whole game. You are no longer chasing volume. You are earning a smaller, tighter slot, and you have to be unmistakably worth it.
Dwell time is the currency now
If relevance decides whether your post enters a feed, attention decides whether it keeps traveling. Dwell time, how long someone’s eyes actually rest on your post, has quietly become the cleanest quality signal the platform has, because it is hard to fake and it tracks the thing LinkedIn wants: people finding the feed worth their time.
The gap is not subtle. By the same analysis, a post that holds attention for around a minute can earn on the order of ten times the engagement of one skimmed in a few seconds. The first line and the format do most of that work. A flat opener and a wall of text get the three-second flick. A sharp first line and a format that invites a swipe buy the seconds that tell the algorithm to keep going.
This is also where generic AI writing quietly dies. A post that reads like a model wrote it gives a reader nothing to stop on, so dwell collapses, so distribution stops. Nobody flags anything. In 2026 LinkedIn went further and announced it would actively limit the reach of low-effort AI content, but it does not need a detector to do it. It reads behavior. Slop earns no dwell, no saves, no comments, and the model reads that fingerprint and stops distributing. Whether a human or a machine typed it is almost beside the point.
For anyone creating with AI, that is the real lesson, and the real limit. Speed only helps if what ships still sounds like a person who did the work. A faster way to publish the average is just a faster way to be ignored.
What still works
Strip away the model names and the panic and the levers are short and durable: show up on a real cadence, sound like a human, pick one lane so the model knows what you are about, and lean on formats that earn dwell, carousels and native Documents, where every swipe is another second of attention. None of it is a hack. The new feed just pays for unglamorous, compounding behavior instead of volume.
The full breakdown of why reach is falling and the four levers that still move it goes deeper than I can here, including the 30-day version of the plan. But the one-line version is simple enough to start today: LinkedIn stopped paying for reach and started paying for attention. Earn the second, and the reach comes back.