We Ranked All Four LinkedIn Formats. One Wins.

Native documents win. In Socialinsider's study of 1.3 million posts, document posts averaged a 7.00% engagement rate against 6.00% for video, 5.30% for images and 4.50% for text. The gap from first to last is real but modest, which matters more than the ranking itself.

Format is the easiest thing to change about your LinkedIn posting and the least important. Worth knowing anyway, because the difference between the best and worst format is free, and nobody should be leaving free things on the table.

Where do these numbers come from?

One study, and it is worth understanding before leaning on it.

Socialinsider analyzed 1.3 million LinkedIn posts across 16,645 business pages between January 2024 and December 2025. That is a large sample with a stated method, which already puts it ahead of most LinkedIn advice.

Two caveats they state themselves, and I will repeat because they change how you should read this. The figures published as 2026 are 2025 values, because the year was too young at publication to have enough data. And the sample is business pages, not personal profiles, which is a meaningful limitation given that profiles substantially out-reach pages.

So read this as a ranking of formats on company pages during 2025. That is still the best evidence available, and it is not a law of nature.

The ranking

Multi-image posts sat between documents and video at 6.45%, and two formats came in below all four above: polls at 4.20% and links last at 3.25%.

The platform average across everything was 5.20%, up 8% year over year.

Why do documents win?

Because the format forces the thing that earns engagement anyway.

A native document is a PDF that LinkedIn renders as a swipeable carousel. To make one you have to break an idea into discrete slides, which means deciding what the idea actually is. The format punishes waffle in a way a text box does not.

It also holds people in the feed. Swiping through eight slides is time spent on the post without leaving the platform, and platforms reward that consistently.

There is a scarcity effect too. The same study found pages posting an average of two documents a month against seven images. Less competition for attention in that format, at least for now.

Should you just post documents constantly?

No, and the numbers explain why better than the advice usually does.

The spread from first to last is 7.00% against 4.50%. That is roughly a 1.5 to 1 ratio. Real, worth having, and nowhere near large enough to override whether the thing you are saying is any good.

A sharp text post from someone with something to say will beat a mediocre carousel comfortably. Format is a multiplier on the content, not a substitute for it, and a multiplier on nothing is nothing.

The other reason is that all-documents is exhausting to produce and monotonous to follow. Rising engagement rates by format also reflect what is currently uncommon, so a feed of nothing but carousels forfeits the scarcity that helped them.

What about video?

Second place, and the trend underneath it is the interesting part.

Video engagement rose 7% year over year in the same study. But video views declined 36% year over year across all page sizes, and posting frequency doubled from two to four a month.

Put those together. More pages posting more video, each getting fewer views, while the people who do watch engage slightly more. That is a crowding pattern, not a growth pattern.

Which does not mean skip video. It means treat "post more video" as advice that was better two years ago than it is now, and be skeptical of anyone still selling it as an easy win.

Why are links last?

Because you are asking the platform to send people away, and it would rather you did not.

Link posts came in at 3.25%, less than half the top format. This is not unique to LinkedIn. The same pattern shows up on Facebook, where Socialinsider's parallel study put link posts at 0.05% against 0.20% for plain status updates.

The practical workaround is well known and still works: put the substance in the post, and the link in the first comment or added afterwards by editing. You keep the distribution and still get the click from people who actually want it.

If your company page is mostly link posts, that alone explains a good part of your reach. I go through the rest of the picture in the LinkedIn best practices post.

What should the mix actually be?

Something you can sustain, weighted toward the top of the table.

A reasonable shape for most pages is roughly one document a week, one or two image or multi-image posts, and text whenever there is something worth saying that does not need a picture. Video when you have something genuinely visual, not to hit a quota.

Then leave it alone for a month. Format experiments run for a week tell you nothing, because a single post that happened to catch a good hour will swamp the effect you are trying to measure.

The formats are close enough together that only sustained differences are real.

How do you make a document worth swiping?

Structure it like an argument, not like a slide deck.

The failure mode is a carousel that is really a blog post cut into pieces at arbitrary points. Nothing pulls you from slide three to slide four, so most people leave at slide three and the format's main advantage is wasted.

What works is giving every slide one idea and a reason to continue. The first slide carries the promise, stated specifically enough to be worth the swipe: not "Lessons on hiring" but "The hiring mistake that cost us six months". The middle slides deliver one point each, in words large enough to read on a phone held at arm's length. The last slide closes the loop the first one opened, and asks for one thing.

Eight to twelve slides is the range that tends to hold. Below eight there was not enough to justify the format. Above twelve, completion falls away and the last slides are talking to almost nobody.

The constraint that improves these most is size. If a slide needs small type to fit, it has too much on it. Cutting the words until they fit at a large size does the editing you were avoiding.

What I see in B2B content calendars

Effort allocated in almost exactly the wrong order.

Enormous production time goes into video, which sits second and is getting more crowded. Very little goes into documents, which sit first and are still relatively rare. And a large share of posts are links out, which sit last by a wide margin.

The other pattern is format-chasing. Someone reads that documents win, switches everything to carousels, and produces eight slides of thin content because the format was the plan rather than the idea. The engagement rate on a carousel with nothing in it is not 7.00%.

What to do on Monday

Count the last twenty posts on your company page by format. If links out are the largest group, move the links to the comments and put the substance in the post itself. That is the highest-value change on this page and it takes no new production.

Then make one native document this week. Take something you already published, a blog post or an internal explainer, and cut it into eight slides. You are not creating new thinking, you are reformatting existing thinking into the format that currently travels furthest.

FAQ

Which LinkedIn post format gets the most reach? Native documents. Socialinsider's analysis of 1.3 million posts across 16,645 business pages put them at a 7.00% engagement rate, ahead of multi-image at 6.45%, video at 6.00%, images at 5.30% and text at 4.50%.

Are LinkedIn carousels and native documents the same thing? Effectively yes. A native document is a PDF uploaded directly to LinkedIn, which the feed renders as a swipeable carousel. That is the format the engagement figures describe.

Do link posts hurt your LinkedIn reach? They perform worst of any format, at 3.25% against a 5.20% platform average. Putting the substance in the post and the link in the first comment is the standard workaround and still works.

Is video still worth posting on LinkedIn? Yes, but with lower expectations than two years ago. Engagement rate on video rose 7% year over year while video views fell 36% and posting frequency doubled, which points to crowding rather than growth.

How much does post format actually matter? Less than the advice implies. Best to worst is about 1.5 to 1, so format is a multiplier on content quality rather than a replacement for it. A good text post beats a weak carousel.

About the author

Adam Palmer is President of Inertia Digital Marketing, the agency behind Social Fonts. He has spent 15 years running paid and organic campaigns for clients ranging from single-location small businesses to national franchise networks, and has managed over $50 million in paid digital spend.