The algorithm that decides who sees your LinkedIn post changed in 2024. Not in the way algorithms usually change — a tweak here, a weight adjusted there. It changed its objective function. That is a different kind of event.
For the decade before, LinkedIn’s feed optimized roughly the same thing every other social feed optimized: engagement. Time spent, clicks, comments, reactions. The content that got those numbers got shown to more people. The content that did not, did not. That created a predictable set of winning patterns. Ask a question. Post a hot take. Write a listicle. Link to something interesting. Post it late in the day when people are doomscrolling. These worked because they generated the signals the algorithm was looking for.
Then LinkedIn’s engineering team published a post in 2024 stating that the platform would begin prioritizing “knowledge and advice” over engagement bait, as stated in a 2024 LinkedIn Engineering blog post. LinkedIn CEO Ryan Roslansky put it more directly in an interview the same year: the algorithm is designed to “maximize professional learning and connection, not just time spent,” as he stated in a 2024 interview. That is a structural break, not a minor update. The algorithm stopped asking “does this keep people on the platform?” and started asking “does this make people better at their jobs?” The two questions produce very different answers.
The three penalized patterns
Three tactics that reliably generated reach before 2024 now actively reduce it. The data is specific enough to treat them as settled.
External links. Richard van der Blom’s annual LinkedIn algorithm report (2024 edition) found that posts containing external links saw a 40% reduction in organic reach compared to native posts. This is not accidental. LinkedIn wants users to stay on the platform. A link sends them elsewhere, which means the post generates no dwell time, no further scrolling, no additional feed interactions. The algorithm reads that as a low-value signal and deprioritizes the post accordingly. The old strategy of “post a link with a compelling caption” now costs you nearly half your potential reach.
Generic engagement bait. The kind of post that asks “Agree?” or “Comment your thoughts below” or “Which option would you pick?” without offering any substantive content first. These worked because they generated comments cheaply. But the 2024 shift to prioritizing knowledge and advice means the algorithm now weighs whether a post contains original insight, not just whether it generates interaction, as stated in a 2024 LinkedIn Engineering blog post. A post that is all prompt and no substance gets penalized because it does not deliver the professional learning the algorithm is now optimized for.
Late-day posting. Lara Acosta’s 2024 LinkedIn growth analysis indicated that posts published between 8 AM and 10 AM local time saw 25% higher engagement than those posted after 5 PM. The mechanism here is probably simple. Early-morning posts catch users during their first feed check of the day, when they are more likely to read substantively. Late-day posts hit during a lower-attention window, generating less dwell time, which the algorithm then reads as a low-quality signal. The old advice to post when your audience is “most active” turns out to be incomplete. You also need to post when they are most attentive.
The two newly-rewarded patterns
The same algorithm change that penalized those three patterns created two new ones that now outperform everything else.
Video-native posts. Justin Welsh reported in 2024 that his LinkedIn posts with video achieved 3x higher reach than text-only posts. Hootsuite’s 2024 social media trends report noted that LinkedIn video posts had a 5x higher share rate than other formats. These numbers are not independent. Video keeps users on the platform longer, which generates dwell time, which the algorithm amplifies. The video does not need to be produced. A founder talking directly to camera for sixty seconds about a specific business problem outperforms a polished text post that took an hour to write. The format itself carries the signal.
Saved-content signals. LinkedIn’s 2024 algorithm update introduced dwell time as a ranking signal, measuring how long users spend viewing a post before scrolling, as stated in a 2024 LinkedIn Engineering blog post. This is the mechanism behind everything else. A post that keeps someone reading for thirty seconds generates a stronger signal than a post that gets a quick like and a scroll past. The practical implication is that substantive, longer posts now outperform quick-hit one-liners. The old wisdom that shorter is better on social media is inverted for LinkedIn in 2026. Posts that take time to read are rewarded because the algorithm values time on post over click-through rate.
The algorithm stopped asking “does this keep people on the platform?” and started asking “does this make people better at their jobs?”
What dwell time means for post length
The dwell time signal is the most consequential change in the 2024 update because it rewrites the fundamental economics of a LinkedIn post, as stated in a 2024 LinkedIn Engineering blog post. Before, the optimal post was short, punchy, and designed to generate a quick reaction. A one-sentence hot take with a question at the end could outperform a thoughtful essay because the hot take generated more comments per word.
Dwell time flips that. A post that takes thirty seconds to read now generates more ranking signal than a post that takes five seconds, all else being equal. The algorithm is literally measuring how long your content holds attention, and it amplifies content that holds it longer. This means the optimal post in 2026 is longer, denser, and more substantive than the optimal post in 2023. It means bullet points and white space are not just readability choices. They are ranking factors, because they affect whether a user reads to the end or scrolls past at the first dense paragraph.
The caveat is that dwell time only works if the content justifies it. A long post that is boring or padded will generate negative signals because users will scroll past it faster than they would scroll past a short post. The algorithm is not rewarding length. It is rewarding attention capture. Length is just the most reliable way to achieve it when the content is good.
A posting framework for 2026
The available evidence supports a concrete set of choices for anyone posting on LinkedIn in 2026.
Lead with video at least once a week. Video posts achieve 3x higher reach, per Justin Welsh’s 2024 analysis, and 5x higher share rates than text-only posts, per Hootsuite’s 2024 social media trends report. The bar for production quality is low. A founder recording a two-minute reflection on a specific business decision, filmed on a phone with decent lighting, will outperform a carefully crafted text post that took an hour to edit. The format itself carries the dwell-time signal.
Post between 8 AM and 10 AM local time. The 25% engagement premium for morning posting, as indicated by Lara Acosta’s 2024 growth analysis, is large enough that it should be treated as a default scheduling rule. If you can only post once a day, post then.
Remove external links from your posts. The 40% organic reach penalty for external links, per Richard van der Blom’s annual LinkedIn algorithm report (2024 edition), means that linking out is now a strategic choice with a measurable cost. If you need to drive traffic to an article or a landing page, consider posting the content natively and adding the link in the comments or in a follow-up post. Or accept the reach reduction and decide that the traffic is worth it.
Write posts that take thirty seconds to read. The dwell time signal rewards substantive content, as stated in a 2024 LinkedIn Engineering blog post. That does not mean long for the sake of long. It means every post should contain enough original insight or specific experience that a reader gains something by reading to the end. If your post can be summarized in the first two sentences, it is too short.
Post less, but better. The old strategy of posting daily engagement bait to stay top of mind is actively counterproductive under the new algorithm. A single high-quality post that generates strong dwell time and shares will outperform a week of thin content. The optimal cadence is probably three to four posts per week, with at least one being video and the rest being long-form text that delivers on the “knowledge and advice” signal the algorithm now prioritizes, as stated in a 2024 LinkedIn Engineering blog post.
The algorithm changed its objective function in 2024. The tactics that worked before that change are not just less effective. They are actively penalized. The founders and operators who adapt to the new signals will see their reach grow. The ones who keep posting links at 6 PM and wondering why nobody sees them will keep wondering.