1d:I[58553,["2550","static/chunks/96cb50b6-60e4ebcfc9e986f2.js","8754","static/chunks/eed5e382-ee92683b99edf94b.js","2534","static/chunks/2534-d4488ec44c364215.js","7101","static/chunks/7101-f413d89e9cba3b63.js","1310","static/chunks/1310-e2a289329247ece1.js","6790","static/chunks/6790-8ccbcaf79fba421f.js","2642","static/chunks/2642-abf34e6aa57e0523.js","9230","static/chunks/9230-abec6248ed19d865.js","1566","static/chunks/1566-140144b37dd19b78.js","3244","static/chunks/app/(app)/blog/%5Bslug%5D/page-32be5b62ae91bd42.js"],"Image"] 1e:T243f,

“Best-performing LinkedIn content” is a format question: which post types get more reach and engagement than that same account’s own median. It is not a secret 2025 algorithm, and it is not Influensea’s private dataset.

Two public samples are worth citing. They do not mix:

Personal profiles: AuthoredUp (https://authoredup.com/blog/best-performing-content-on-linkedin) — 3 million+ posts, March 2025–February 2026. Multipliers are vs each profile’s own median, not vs the whole network.
Company pages: Social Insider’s LinkedIn Organic Benchmarks — 1.3 million posts from 16,645 business pages, January 2024–December 2025. Those are engagement rates, not the same multiplier.

If you need why dwell and comments matter in ranking, read LinkedIn organic reach (https://www.influensea.com/blog/linkedin-organic-reach-2025-how-to-beat-the-algorithm). Stay here for what to publish.

Personal profiles (AuthoredUp)

| Format | Reach vs own median | Engagement vs own median | Share of posts | Honest read | | --- | --- | --- | --- | --- | | Document (carousel / PDF) | 1.39× | 1.30× | 4.88% | Best combined; still rare | | Image | 1.20× | 1.33× | 57.16% | Volume default; strong replies | | Poll | 1.78× | 0.37× | 1.16% | Reach trap | | Text-only | 1.07× | 0.78× | 11.59% | Average reach; weaker engagement unless you can write | | Video | 0.86× | 0.93× | 10.61% | Below median after the 2024 video push | | Article (LinkedIn long-form) | 0.69× | 0.44× | 6.00% | Weak in the feed | | Reshare | 0.29× | 0.22× | 7.69% | Worst |

Source: AuthoredUp, personal profiles only, March 2025–February 2026.

A 1.39× reach multiplier means that account’s documents beat that account’s typical post, not that every carousel goes viral.

AuthoredUp also reports documents at 1.72× reach / 1.63× engagement among the top 5% of profiles, and documents taking 12.92% of saves — about 2.6× their share of posts. Saves are a quality proxy, not a published LinkedIn weight.

Year-over-year in the same vendor’s trend cut: video reach −36%, documents still first on relative ranking even as raw reach fell across formats. Relative rank ≠ “reach is up.”

Company pages (Social Insider)

On business pages, Social Insider’s average engagement rates (Jan 2024–Dec 2025):

| Format | Avg engagement rate | | --- | --- | | Native document | 7.00% | | Multi-image | 6.45% | | Video | 6.00% | | Image | 5.30% | | Text-only | 4.50% | | Poll | 4.20% | | Link post | 3.25% |

Do not paste these next to AuthoredUp’s 1.39× and call it one table. Pages and people are different graphs. Link posts sit at the bottom of the page sample; if you need a URL, put it in the first comment and keep the post native.

What the table is trying to tell you

Documents win combined reach + engagement because they take time. Swipes are dwell. Six to eight slides, one idea each, a title slide that states the claim, a last slide people might save. Vertical (1080×1350) is what most phones show. Padding to 15 slides is not a strategy.

Images win conversations at high volume. A real screenshot or a chart you made beats stock. AuthoredUp’s follower split matters here: under ~5k followers, images led both reach and engagement; documents pulled ahead mainly above ~20k. If you are small, do not copy a 50k-carousel playbook and call the format broken.

Polls are a vanity impression. Votes count as distribution fuel; they are not comments, profile visits, or a follow. Use a poll only as step one of research you will actually write up.

Text still works if it is long enough to be a post. In that sample, text posts with 1,000+ characters sat at 1.18× reach vs the text-format average; stubs under 300 characters were weaker. Top 5% writers still get text to travel (AuthoredUp: 1.26× reach for that cohort). Average accounts should not treat a three-line hot take as the plan.

Video is not the 2024 growth cheat. Native uploads only. In a 36,946-video slice (Oct 2025–Mar 2026), 3+ minute videos beat sub-30s clips on both reach and engagement vs the video median. LinkedIn is a professional feed, not a short-form app. Captions still matter because people scroll with sound off — that is product reality, not a multiplier we invented.

Do not reshare as your “content.” Quote in your own post if you must reference someone.

Length, not as folklore

AuthoredUp’s caption cuts, same personal-profile window:

Text: commit (1,000+ characters).
Image: enough caption to add context (~800–900 characters in their peak band).
Document: short caption (0–100 characters at 1.28×). Let the slides work.

Match the format. Do not write a newsletter in the document caption.

Cadence vs format

In the same vendor’s frequency table, 4–5 posts/week was the peak median engagement rate (2.60%) and higher impressions per post than weekly-once. 8+ posts/week dropped both rate and impressions per post. 80% of profiles in that cut posted once a week. That is a sample of AuthoredUp users, not LinkedIn’s census. Three to five posts you can keep still beats a heroic week and silence.

How to pick this week (without a 10-format kit)

If you are under 5k followers: one image with a real artifact + one text post you would actually finish. Add a document only if you already have a framework worth saving.
If you have a framework: one 6–8 slide document, caption as a hook, not a recap.
Skip the poll unless you will publish the results.
Skip the reshare. Skip the YouTube link-in-body.
Measure within format. Your best document vs last document, not vs a poll’s inflated impressions.

Preview the thing as a LinkedIn post with a body (and as slides, if it is a document) before it ships. Line breaks and “see more” lie in Google Docs.

Where Influensea fits

Influensea drafts a text post you can paste into a LinkedIn-shaped preview. It is not a carousel designer and not an analytics suite. Use it for the hook-and-body pass; build the PDF where you already design. Five LinkedIn drafts a month on the free plan (https://www.influensea.com/sign-up); no card.

The weekly operating loop (pillars, inbox, replies) is the LinkedIn content system (https://www.influensea.com/resources/linkedin-content-system), not this page.

Questions people type next

What is the best LinkedIn post format?

On personal profiles in AuthoredUp’s 2025–2026 sample: native documents for combined reach and engagement; images if you are still small or you want comments. Polls look big and do little. That is one vendor’s median, not a law.

Do carousels still work?

Yes, as document posts. They are 4.88% of posts in that sample and 1.39× reach vs the poster’s median. They work when each slide teaches. They fail as 20 screens of padding.

Is LinkedIn video dead?

No. It is below the median in that sample after a year of extra uploads. Native, longer, captioned, actually about work — or skip it this week.

Should I post from the company page?

The page sample still ranks documents above link posts. People still usually outperform pages for the same idea. See the reach article for that split.

Ship the format you can finish

Best-performing content is the format that earns time and a reply on your graph, attributed to studies you can name. Pick one, write it in a real preview, publish when you can stay for comments.

Draft the text post (https://www.influensea.com/sign-up), look at it as a body on a card, then add the document or image in LinkedIn itself.

1b:["$","article",null,{"children":[["$","$L7",null,{"className":"mb-8 flex w-max items-center gap-1 text-muted-foreground leading-none hover:text-accent","href":"/blog","children":[["$","svg",null,{"ref":"$undefined","xmlns":"http://www.w3.org/2000/svg","width":24,"height":24,"viewBox":"0 0 24 24","fill":"none","color":"currentColor","className":"mb-0.5 h-5 w-5","children":[["$","path","0",{"d":"M15 6C15 6 9.00001 10.4189 9 12C8.99999 13.5812 15 18 15 18","stroke":"currentColor","strokeLinecap":"round","strokeLinejoin":"round","strokeWidth":"1.5","opacity":"$undefined"}]]}]," All Posts"]}],["$","h1",null,{"className":"my-2 font-bold text-5xl sm:text-6xl md:text-7xl lg:mb-6","children":"Best-performing LinkedIn content, by format |"}],["$","p",null,{"className":"text-lg text-secondary-foreground sm:text-xl","children":"On personal profiles, AuthoredUp’s 3M-post sample puts native documents first for combined reach and engagement. Polls inflate reach and starve replies. Video is below the median. Company-page numbers are a different study. |"}],["$","header",null,{"className":"my-4 lg:my-6","children":["$","address",null,{"className":"mb-6 flex flex-wrap items-center justify-between gap-3 not-italic","children":[["$","p",null,{"className":"font-medium text-primary md:text-lg","children":"lekterable"}],["$","div",null,{"className":"flex items-center","children":["$undefined",["$","time",null,{"className":"mx-1 text-primary text-sm md:text-base","dateTime":"2025-11-24T07:50:10.364Z","title":"2025-11-24","children":"November 24, 2025"}],["$","div",null,{"className":"mx-1","children":" ・ "}],["$","p",null,{"className":"text-muted-foreground text-sm md:text-base","children":[7," min read"]}]]}]]}]}],["$","figure",null,{"children":["$","div",null,{"className":"relative w-full","style":{"aspectRatio":"16/9"},"children":["$","$L1d",null,{"src":"http://zip.kornel.me/u/A40k9e","alt":"Best-Performing Content on LinkedIn in 2025/2026 Big Data Analysis","className":"h-full w-full rounded-lg object-cover","onLoad":"$undefined","onError":"$undefined","fill":true,"sizes":"(max-width: 640px) 100vw, (max-width: 1280px) 50vw, 1280px"}]}]}],["$","div",null,{"suppressHydrationWarning":true,"className":"prose prose-lg prose-li:my-0 mt-8 prose-headings:mt-10","dangerouslySetInnerHTML":{"__html":"$1e"}}],"$L1f","$L20"]}]