What Actually Happened
LinkedIn has quietly finished a shift it started years ago: moving from a relationship graph to an interest graph. In plain terms, the feed used to prioritize content from people already in your network. Now it maps each post against topics a user has shown affinity for, through their engagement, their profile, and the subjects they follow, then distributes accordingly.
The 2026 State of LinkedIn tracking put hard numbers on it. Follower-first reach, the share of a post's views that come from your own followers, has fallen from roughly 40 to 60 percent in the pre-2025 era to about 10 to 20 percent today (State of LinkedIn, Spring 2026). The majority of a strong post's reach now comes from people who do not follow the account at all.
Native video sits at the center of this. LinkedIn continues to give video uploaded directly to the platform favorable treatment, and native video posts run an average engagement rate near 5.6 percent, ahead of most other formats (LinkedIn video statistics, 2026). Put the two facts together and the mechanism is clear: video is the format LinkedIn most wants to distribute, and it will push it to strangers who care about your topic, regardless of your follower count.
Why did LinkedIn do this? The same reason TikTok and YouTube did years ago. An interest graph keeps people watching longer, because it serves relevance instead of relationships. LinkedIn has leaned into a dwell-time and completion-rate model to match, which rewards content that holds attention rather than content that simply collects reactions from a familiar network. For B2B, that means the platform is now optimizing for exactly the thing a good video does: keep a relevant stranger watching.
Why Your Follower Count Stopped Mattering
For a decade, the LinkedIn playbook was audience accumulation. Grow the follower base, and every post inherits guaranteed reach. That model is now broken at the mechanic level. If only 10 to 20 percent of reach comes from followers, then doubling your follower count barely moves your ceiling. What moves it is whether this specific post earns distribution on this specific topic.
That reframes the whole game for B2B. The account with 40,000 followers and an inconsistent posting habit is sitting on a depreciating asset. The account with 2,000 followers that ships four sharp, on-topic videos a month is feeding the exact signal the interest graph rewards: topical relevance plus watch time. On an interest graph, consistency is distribution. Audience size is a lagging indicator of it, not a substitute for it.
This is why the follower-count obsession is a trap for marketing leaders. It optimizes a number the algorithm has largely stopped pricing. Worse, chasing followers usually means broad, safe, reach-bait content, which is exactly the content an interest graph struggles to route, because it is not clearly about anything. The interest graph rewards narrow. It rewards a point of view on a specific problem, delivered often enough that the system learns who to show it to.
The Data
Two numbers matter here, one from the market and one from our own book.
The market number: follower-first reach has collapsed from the 40 to 60 percent range to roughly 10 to 20 percent, per the 2026 State of LinkedIn tracking (source). That is not a tweak. That is the platform telling you where reach comes from now, and it is not your follower list.
The first-party number comes from our retainer book. We run structured LinkedIn video for a set of B2B SaaS accounts, and we track per-post reach against follower count every month. Over the last two quarters, the correlation between an account's follower count and its median per-post reach fell to near zero inside our book. A client under 3,000 followers routinely out-reaches a client above 30,000, when the smaller account posts more consistently on a tighter topic.
The cadence effect is even sharper. In our production tracker, accounts posting four or more on-topic videos a month held or grew median reach even in months when follower growth was flat. Accounts that went dark for three weeks or more lost roughly 60 percent of their median reach, and it took about five to six weeks of consistent posting to climb back. The asset that compounds is not the follower list. It is the posting system that keeps feeding the graph.
One more pattern from our book is worth naming: the accounts that grew reach fastest were not the ones that posted the most. They were the ones that posted consistently on the narrowest topic. Volume without focus scattered the signal. Four videos a month on one problem out-performed eight videos a month spread across five unrelated themes, because the interest graph could actually learn who to route them to.
The Counter-Argument, Steelmanned
The strongest case against this: followers are not worthless. A large, relevant following still seeds early engagement, and early engagement is the signal the interest graph uses to decide whether to widen distribution. No followers, no initial velocity, no wider reach. There is truth here. Followers are the ignition, not the engine.
There is also a lead-gen argument. For account-based plays, you may want to reach a named list of buyers, and interest-graph distribution to strangers is not the same as landing in a specific champion's feed. Fair.
Here is the response. Both objections argue for a following as a supporting asset, not as the primary metric. Ignition matters, but you still need fuel, and the fuel is consistent, on-topic video. A 50,000-follower account that posts twice a quarter gets weak ignition on a cold graph. As for named-account reach, the interest graph actually helps: when you consistently publish on the problem your buyers care about, their engagement behavior pulls your content toward them over time, without a follow. The point is not that followers are bad. It is that follower count is the wrong thing to optimize, because the system now prices relevance and repetition, and those come from a content system, not an audience-growth campaign.
What to Do Monday
First, stop reporting follower count as a primary KPI. Replace it with median per-post reach and watch time on native video. Those are the numbers the interest graph actually responds to, and they tell you whether your topic is landing.
Second, pick a lane and narrow it. The interest graph cannot route content that is about everything. Choose one or two problems your buyers own, and commit your video output to them for a quarter. Relevance is now a distribution input, not just a brand nicety.
Third, set a cadence you can hold without heroics. In our book, four on-topic native videos a month is the floor where reach compounds rather than resets. Two is maintenance. Zero for three weeks is a measurable setback. Pick the number you can actually sustain, and build the production system to hit it.
Fourth, shoot for native video first. It is the format LinkedIn is most eager to distribute to non-followers, and it carries the watch-time signal the graph weights. Repurposing a blog into a carousel is fine, but the video is the wedge.
Fifth, treat your best-performing topics as a map. When a video over-indexes on reach, the graph is telling you which audience it found. Make more on that exact topic before you chase a new one.