LinkedIn trials a feed built only on your connections

LinkedIn is testing a new tab. Social Media Today reports that the feature offers a feed built entirely from a user's direct connections and follows, stripped of the algorithmic recommendations that otherwise pad the main feed with content from strangers three degrees removed. Call it the Network tab. Call it, more accurately, a confession.

For years LinkedIn's main feed has behaved like most platform feeds: it optimises for engagement regardless of relationship, surfacing viral posts from accounts a user has never followed because the algorithm has decided they will scroll longer and click more. That is why a founder's post about a supply-chain outage can outperform a connection's product launch, and why a mid-tier account with one lucky post can outreach the head of communications at a multinational bank. The Network tab is LinkedIn admitting that this design has a cost: it does not necessarily deliver you back to the people you actually chose to follow.

The tell in the design

A platform does not build a "just the people I follow" tab unless the main feed has drifted far enough from that experience to need one. Instagram did this with Following feeds after algorithmic ranking buried friends under Reels from strangers. X did it by keeping "Following" as an option alongside "For You." LinkedIn's version is notable because LinkedIn has spent three years selling itself, correctly, as the place where professional reach compounds through algorithmic distribution rather than follower count alone. A test like this concedes that some users, plausibly the ones LinkedIn most wants to retain, want the platform to behave less like a discovery engine and more like a directory.

That distinction matters enormously for who wins attention on the platform. If the Network tab expands beyond test, it creates a second, smaller, higher-intent surface where reach is gated not by hook rate or dwell time but by whether someone actually chose to connect with you. For an executive at a multilateral institution or an industrial group, that is a very different audience than the algorithmic feed's drive-by scroll. It is procurement officers who follow the World Bank's private-sector arm because they work adjacent to it. It is compliance officers who connected with an ISO standards lead at a conference two years ago and never unfollowed. Low volume, high relevance. The kind of audience that turns into a DM, not a like.

Two feeds, two disciplines

The practical consequence is that brands and executives may soon be optimising for two different feeds with two different logics, and treating them as one will cost reach on both.

The main feed still rewards format novelty, native documents, carousels, and comment velocity in the first 60 to 90 minutes. It is a stranger's feed, and strangers need a hook. The Network tab, by contrast, has no strangers in it. Its currency is not the scroll-stopping headline but consistency and specificity: the connection who posts every week about pricing trends in reinsurance, or the comms director who reliably explains what a UN resolution actually changes on the ground, builds standing there simply by showing up on a topic people already associate with them. Virality is structurally impossible in a feed with no algorithmic amplification; familiarity is the entire mechanism.

For financial-services and policy institutions, this should reframe how executive presence gets measured internally. A CFO's post that gets 40 comments from strangers on the main feed is a different asset than one that gets 12 comments from people who actually sit on the other side of the table in deals or negotiations. The second number has always mattered more; the Network tab is the first LinkedIn feature to structurally separate it from the noise of the first.

The test may not ship widely. LinkedIn has quietly killed engagement experiments before. But the impulse behind it will not go away, because the underlying tension, between a feed optimised for time-on-platform and a network optimised for relationship value, is structural, not cosmetic. Brands that have spent three years chasing algorithmic reach on the main feed should treat this as an early warning: the platform is signalling that reach and relevance may soon be measured, and rewarded, separately. The safer bet is to start building for both now, rather than relearning a channel from scratch when LinkedIn decides which feed actually keeps people coming back.