Insight · Social strategy
Nobody's seeing your posts
Read article →
Every major tracker measuring LinkedIn performance in 2026 lands on the same conclusion, using entirely different data sets and different maths to get there. Carousels, technically PDF "document" posts, are the single best-performing format on the platform. Not by a small margin. By multiples.
Only 4.88% of profiles post them regularly. That's not a niche tactic sitting untapped, that's most of LinkedIn ignoring its own best-performing format.
The numbers, and why they don't all match
Depending on which tracker you check, carousel engagement sits anywhere from 6.6% to over 24%. That's a genuinely wide spread, and it's worth being upfront about why: each firm measures engagement differently, some as a percentage of followers, some as a percentage of impressions, some across different account sizes and industries. None of them are wrong, they're just not measuring the same thing.
What matters is that every single one of them, regardless of methodology, puts carousels at the top of the format hierarchy, consistently ahead of video, images, and text. A sample of what's been published in 2026:
- Socialinsider (1.3 million posts): documents at 7.00% average engagement, up 14% year-on-year, the highest of any format tracked.
- Buffer (52 million posts): carousels at a 21.77% median, generating 278% more engagement than video and 596% more than text-only posts.
- AuthoredUp: document posts generating 39% more reach and 30% more engagement than the average post, rising to 1.72x reach among the top 5% of profiles.
Different numbers. Same podium finish, every time.
Why the format wins
The mechanic behind it is consistent across every source too: dwell time. LinkedIn's algorithm, known as LiRank, explicitly rewards how long someone spends actively engaging with a post, not just whether they liked it. A carousel forces a physical action, swiping, that a scroll-past text post or a passively watched video doesn't. Each swipe is a small, repeated signal to the algorithm that someone is genuinely engaged, not just scrolling through.
Each swipe is a small, repeated signal to the algorithm that someone is genuinely engaged, not just scrolling through.
As Semrush's Julia Holmqvist put it in Social Insider's 2026 benchmarks report, document posts work because they behave like free value on a platform where people actively want to learn, and the format itself is simply easy to scan.
LinkedIn format performance, side by side
| Format | Relative engagement | Why |
|---|---|---|
| Carousel / document | Highest across every tracker | Swipe-driven dwell time, seen as a "free value" format |
| Video | Second, but declining | Reach down as much as 36% year-on-year on some trackers |
| Image | Mid-table | Still outperforms text, but no swipe mechanic |
| Text-only | Lowest | Fastest to scroll past, shortest dwell time |
| Polls | High reach, weak engagement | A "reach trap", visibility without genuine interaction |
What to actually do with this
- Build carousels as PDFs, not native slide tools. The most consistent format across every guide: a multi-page PDF, ideally sized around 1080x1350px, uploaded as a document post rather than using LinkedIn's native carousel ad format.
- Open with a cover slide, then a numbered structure. A clear framework, a checklist, a step-by-step breakdown, gives someone a reason to keep swiping rather than stopping after slide one.
- Design for the swipe, not the scroll. Each slide should work as a small, complete unit, since completion tracking (how many slides someone actually views) is part of what LinkedIn's algorithm reads as engagement.
- Don't abandon every other format. One tracker's practical guidance is a 70% carousel, 30% text split, since text posts still build the personal voice and relationship-building that a purely instructional carousel doesn't.
- Save this for content that's genuinely worth swiping through. A weak carousel with nothing to say performs no better than a weak post in any other format, the structure creates the opportunity, the content still has to earn the swipe.
This is the same underlying pattern we've flagged before with organic reach decline across social platforms - the format that rewards genuine engagement over passive scrolling is the one that survives an algorithm update, not the one that's easiest to produce.
Want carousels that are actually worth the swipe?
We build and design LinkedIn carousel content for clients as part of full-funnel social strategy. Book a 30-minute call and we'll talk through what that would look like for yours.
Book a 30-minute callFrequently asked questions
Is this the same as LinkedIn's old native carousel feature?
No. LinkedIn retired its original native carousel format in 2023. What's referred to as a "carousel" now is a PDF document uploaded and displayed as swipeable slides.
Why do different reports show such different engagement numbers for the same format?
Because they're not all measuring engagement the same way, some use followers as the base, others use impressions, and sample sizes and account types vary. The consistent part across every study is the ranking, not the exact percentage.
Does this work for every account size?
Mostly, with one caveat. Under roughly 5,000 followers, single images can outperform documents. Carousels tend to take the clear lead once an account passes around 20,000 followers.
Is LinkedIn's algorithm biased toward the carousel format specifically?
Not by design, according to LinkedIn's own signals. The algorithm rewards dwell time and comment quality, and carousels simply generate those signals more reliably because of the swipe mechanic, not because the format itself is favoured.
