Why Your LinkedIn Reach Is Collapsing

If your LinkedIn posts feel like they're landing in a void lately, it's not your imagination — the platform has quietly rebuilt how content gets distributed, and most marketers haven't caught up.

The shift is structural. LinkedIn's 360Brew AI system — a 150-billion-parameter model deployed to the live feed — moves distribution away from "broadcast-to-followers" toward precision delivery, pushing posts only to people the algorithm predicts will find them genuinely useful. That means generic, broad-appeal content now gets filtered out before it ever reaches the people who followed you.

The numbers behind this are stark. According to AuthoredUp's dataset of 621,833+ posts, average reach dropped 34% in 2025, with 98% of users seeing declines. Van der Blom's 2025 report, analyzing over 1.8 million posts, shows views down roughly 47%, engagement down 39%, and follower growth down 42%.

Company pages have been hit hardest. Organic company content now captures only about 2% of the feed, while page reach has dropped 60–66% since 2024. Personal profiles, by contrast, still earn 3–5x more organic reach with identical content — reaching 10–15% of connections versus 2–5% of page followers, per algorithm research.

The squeeze happens in three stages:

• An initial spam/quality filter rejects 40–50% of posts within the first 60 minutes
• Survivors go to a test audience of 100–500 close connections
• Only posts that perform early earn extended distribution to thousands

This is why the first hour now decides everything — and why one strong angle matters more than five mediocre ones. It's also why testing creative cheaply, before committing budget, has become the smart play. As one analysis puts it, running creative tests purely through paid ads means "paying full price to learn which message resonates, when a few organic posts could have revealed that for free" (imarcamplify).

That logic shapes how we approach creative at Worqd: test multiple angles on an offer organically first, then amplify only the proven winners with paid budget. Precision delivery rewards precision testing — the algorithm now filters for depth signals like dwell time and saves, not raw volume. Posts holding attention for 61+ seconds achieve 15.6% engagement versus 1.2% for posts skimmed in under three seconds (SocialPilot).

The takeaway is simple: reach isn't dying, it's being rationed — and it flows to content engineered for the algorithm's new rules.

The New Ranking Signals: Dwell Time and Saves

For years, likes were the scoreboard on LinkedIn. That era is over. The algorithm now rewards how long people stay, what they save, and whether the conversation actually goes somewhere — and the gap between posts that win on these signals and posts that lose is enormous.

Start with dwell time. Posts that hold a reader's attention for 61+ seconds achieve 15.6% engagement, compared to just 1.2% for posts skimmed in under three seconds, according to recent algorithm research. That's a 13x difference driven entirely by whether someone keeps reading. Dwell time alone carries an estimated 40%+ of the ranking weight, per one analysis of LinkedIn's ranking system — which means your first two lines and your closing question matter more than your hashtags ever did.

Then there's the quietest signal of all: the save. One save carries 5x the algorithmic weight of a like and roughly 2x the weight of a meaningful comment, according to a dataset of 621,833+ LinkedIn posts. Saved posts also give readers a 130% higher chance of following you. In practice, this means a post with 200 genuine saves and 30 substantive comments from your target audience can outperform a post with 2,000 likes and a wall of emoji reactions.

Comments matter too — but depth beats volume. Replies to comments (what researchers call indirect comments) can boost reach up to 2.4x, and the top 1% of creators reply to comments 741% more often than average creators do. The lesson: the conversation after your post goes live is part of the post itself.

So what does this mean for how you test creative? It changes the pass/fail metrics. When we test ad angles organically at Worqd — running multiple hooks and concepts against the same offer before any paid budget goes out — we don't judge winners by likes. We look at:

• Average dwell time — are people reading past the first three lines?
• Saves — is this useful enough to come back to?
• Substantive comments — is the right audience actually talking?
• Reply depth — are we staying in the conversation during that critical first hour?

This matters because organic testing is the cheapest research you'll ever run. Testing three different angles on the same offer organically costs nothing but time, while learning the same lesson through paid ads burns budget just to find out what a few posts could have revealed for free. Once a message proves itself on these depth signals, it becomes a much stronger candidate for paid amplification.

If you want help building a creative testing rhythm around these signals — or want more tactics like these — the Worqd insights hub goes deeper, and a growth call is the fastest way to talk through your specific funnel.

Personal Profiles Beat Company Pages by 561%

Post the exact same content on your company page and your personal profile, and the personal profile wins — not by a little, but by 561% more reach per post, according to research comparing organic and paid LinkedIn marketing. The algorithm simply trusts people more than logos.

The gap is structural, not cosmetic. Personal posts reach 10–15% of your connections, while company page posts reach only 2–5% of followers — a 3–5x advantage baked into how LinkedIn distributes content. Worse for brands, organic company content now makes up only about 2% of the feed, and page reach dropped 60–66% between 2024 and 2026.

Why does LinkedIn favor people? Because engagement follows people. Employees' shared content consistently outperforms the brand's own posts, and LinkedIn's ranking system rewards the authentic interaction profiles generate. As one analysis puts it, organic LinkedIn marketing today is really personal-profile marketing wearing a company's message.

What this means in practice:

• Route every campaign through founder, sales, and team profiles — not just the company page.
• Treat employee advocacy as a distribution channel with a plan, not a hopeful afterthought.
• Keep the company page for credibility, hiring, and paid amplification — not primary reach.

This shift also changes how you should test creative. Posting three different angles on the same offer through personal profiles "costs nothing but time," while running the same test purely through paid ads burns budget to learn what organic could have shown for free, as the same organic-vs-paid analysis notes. Winners proven organically then perform better when amplified with paid spend, combining authenticity with guaranteed distribution.

That's exactly how Worqd approaches creative testing: generate a batch of concepts and hook variations, let personal-profile posts reveal which angles earn genuine engagement, then put paid budget only behind what's already working. You pay to amplify proof, not to discover it.

The economics back this up. Organic LinkedIn costs roughly $164 per lead versus about $310 paid, with organic ROI compounding over 3–6 months, per the same research. None of that happens if your best content sits on a page the algorithm barely shows to anyone.

If you take one thing from this section: your people are your reach. Build the content engine around them.

Creative Testing That Costs Nothing but Time

Most advertisers pay full price to learn which message resonates — when a handful of organic posts could have revealed the same answer for free. That's the core insight behind organic-first creative testing: post multiple angles on the same offer, watch what earns genuine engagement, and only then put budget behind the winners.

Here's how the loop works in practice. Draft 3–10 angle variations of your core message and publish them as document carousels or polls — the two formats LinkedIn rewards most. Data from 621,833+ posts shows carousels earn a 1.45x reach multiplier on personal profiles, while polls lead all formats at 1.64x. Text-only posts sit at the bottom at 0.88x, so format choice alone can double your test's signal quality.

Then measure what the algorithm actually values. Research on LinkedIn's ranking signals shows posts that hold attention for 61+ seconds achieve 15.6% engagement versus just 1.2% for posts skimmed in under three seconds. And a single save carries five times the reach weight of a like — saves and dwell time, not likes, are your pass/fail metrics.

Your organic test checklist:

• Run 3–10 angle variations per offer as carousels (8–12 slides) or polls
• Judge winners by saves, dwell time, and substantive comments — never vanity likes
• Route tests through personal profiles, which get 3–5x the reach of company pages
• Boost only proven winners as Sponsored Content

The final step is where the economics flip in your favor. Once a post proves itself organically, boosting it as Sponsored Content combines organic's authenticity signal with paid's guaranteed distribution — you're amplifying evidence, not gambling on guesses.

This is exactly how Worqd structures creative testing: the Creative Sprint model generates 10 ad concepts with 3 hook variations each from a single brief, so clients test up to 30 angles before spending on distribution. Skipping the cheap organic test means paying full price to learn what a few free posts could have shown you — and with organic leads costing roughly $164 versus $310 paid, the savings compound down the entire funnel.

Want more tested angles before you spend a dollar on ads? Explore the Worqd insights hub or book a free growth call at worqd.com/book.

Format, Timing, and Penalty Traps to Avoid

Most LinkedIn reach isn't lost to bad ideas — it's lost to format choices and penalty traps that quietly cut your distribution before anyone even sees your post. Getting these two things right can matter more than the content itself.

Start with format. Reach benchmarks show document carousels earn 12–18% reach — the highest of any format — while link posts sit at just 3–6%. Engagement data puts carousel performance at 6.60% average engagement, with 8–12 slides as the sweet spot. If you're testing creative variations, carousels should anchor every batch.

Timing matters too. The first 60 minutes decide whether LinkedIn's spam filter — which rejects 40–50% of posts before distribution — passes your post into a test audience of 100–500 close connections. Post Tuesday through Thursday, 8–10 AM or 12–1 PM, and respond to every comment in that first hour.

Then there are the penalty traps, and they're expensive:

• External links in the post body — a 50–70% reach penalty. Put links in the first comment, or add them after early comments arrive.
• Engagement pods — LinkedIn detects them with ~97% accuracy, and one documented case saw reach collapse from 8,500 to 340 impressions overnight.
• Generic AI content — posts lacking original insight get ~30% less reach and 55% less engagement, even when the writing itself is fine.

That last point deserves clarity. As analysis of the AI penalty puts it, using AI writing tools is not the trigger — publishing content that lacks authentic insight is what earns the downrank. So the answer isn't avoiding AI; it's layering your own data, opinions, and client-specific stories on top of AI-generated drafts. That's exactly how we build creative at Worqd: AI produces the volume, human insight makes it worth reading.

This is also why testing format and angle together beats guessing. Testing multiple organic angles on the same offer costs nothing but time — while running the same experiment purely through paid ads burns budget to learn what organic would have shown for free. Our Creative Sprint works this way: 10 concepts × 3 hook variations from one brief, posted organically, with dwell time and saves — not likes — as the pass/fail signals.

Pick the carousel format, respect the first hour, keep links out of the body, and never ship AI content without a real point of view. Do that consistently, and the algorithm stops being your opponent.