LinkedIn document post vs carousel: same feature, two names

A LinkedIn document post and a LinkedIn carousel are the same thing. There is no carousel button in the LinkedIn composer: you upload a PDF, PowerPoint, or Word file through the document icon, and LinkedIn renders each page as a swipeable slide with a page counter. LinkedIn calls the feature documents. Everyone else calls the result a carousel, because that is what swiping through panels is called on every other platform. The reason this comparison gets searched at all is that four different products share the vocabulary, and almost nobody separates them clearly.

Is a LinkedIn document post the same as a carousel?

Yes, for organic posting. If someone tells you to “make a LinkedIn carousel,” the action they are describing is: design pages, export a file, click the document icon in the composer, upload, add a document title, write a caption, publish. That is a document post. There is no separate feature to find and no setting to toggle.

The confusion is worth taking seriously, though, because four distinct things go by overlapping names and they behave nothing alike.

If you only take one thing from this article, take the second and third rows. Most bad LinkedIn carousel advice on the internet is carousel-ad advice that wandered into an organic context, or multi-image advice wearing the wrong label.

Why do two names exist for one feature?

Because naming follows behaviour, not buttons. LinkedIn built a feature for sharing files: a deck, a report, a whitepaper. The name it gave the feature describes what you upload. In the feed, though, the result is a panel you swipe through, and swiping through panels has been called a carousel since Instagram made the interaction common.

So the two words come from two different vantage points. “Document post” is the uploader's word. “Carousel” is the reader's word. They describe the same object at different ends of the pipe.

The presentation vocabulary that trails behind both (slides, decks, cover slide) comes from the tools people use to build them. Nobody designs a carousel in a PDF editor. They design it in slide software and export. That is why the format inherited a language that does not appear anywhere in LinkedIn's own interface.

The practical consequence is a search problem, and it is worth knowing which word to use. Searching “document post” gets you LinkedIn's own documentation and hard specs. Searching “carousel” gets you design templates and idea lists. We split our own coverage the same way: the data and the build process in one piece, the concept list in another.

What file formats and limits does a LinkedIn document post have?

These come straight from LinkedIn's help page on uploading and sharing documents, which states: “The file size cannot exceed 100MB and 300 pages,” and “The following file types are supported: PPT, PPTX, DOC, DOCX, and PDF.” Hootsuite's carousel guide reports the same numbers.

Note which rows are documented and which are not. The file types, the size cap, and the page cap are published by LinkedIn. Everything else in that table is either behaviour anyone can observe in the feed or a judgement call we are making. We have labelled them because most spec sheets on this topic present all of it as if it came from LinkedIn, and it does not.

Which file format should you actually export?

PDF. The other four are accepted, and each of them is a worse idea than it sounds.

When you upload a PPTX or a DOCX, it gets converted for display. Conversion is where design goes to die. Fonts that are not embedded get substituted, and substitution changes line lengths, which changes line wraps, which pushes text into places you did not put it. Spacing shifts by a few pixels per element, which is invisible on one page and obvious across twelve. Anything that relied on a text box sitting exactly where you left it is now a gamble.

A PDF is already rendered. What you exported is what people see. The extra step in your workflow is one keystroke, and it removes an entire category of failure that you cannot fix after publishing because the file is not editable.

Two practical notes on the export. Embed your fonts if your tool offers the option, since that also protects you against anything downstream. And export at the page size you designed at rather than letting a template default to A4 in portrait, which produces a tall, narrow frame that wastes horizontal space in the feed.

How does a document post behave once it is in the feed?

This is where the format differs most from everything else you can post, and it is the part that specification lists never cover.

• It plays inline. Readers swipe or click arrows without leaving the feed. Nothing opens in a new tab, nothing loads a new page. Every other long-form option on LinkedIn asks the reader to go somewhere. This one does not, which is the single biggest structural advantage the format has over a long-form article.
• Page one is the thumbnail.Before anyone swipes, the first page is sitting in the feed as a still image. It is doing the same job a video's first frame does. If page one is a title slide with your logo on it, you have spent your only guaranteed impression on a logo.
• There is a page counter. Readers can see how many pages they are committing to. This cuts both ways: a small number invites the swipe, a large number warns people off. It is also why numbering pages inside your design is worth doing.
• The document title renders under the frame.It is a line of copy that readers see, not a filename. “Q3-deck-v4-FINAL.pdf” is a wasted line in a format that gives you very few.
• Do not depend on hyperlinks inside the file. Links embedded in a PDF are not reliably clickable in the in-feed viewer, and the behaviour varies by device. Put any URL you need people to visit in the caption or a comment, never only on page eleven.
• It travels. People screenshot pages, save the post, and send it on. A document that gets saved on Tuesday can produce a reply on Friday, which is a slower and more durable loop than a text post gets.

The first three of those are observable by anyone with a LinkedIn account. The last three are patterns we see rather than published specifications, and we would rather label them than dress them up.

What can you not do with a LinkedIn document post?

The limits matter more than the features here, because most of them are permanent.

You cannot swap the file after publishing. Caption edits are allowed. Document edits are not. A typo on page four means deleting the post and re-uploading, and the new post starts at zero engagement. If the original was performing well, that is a genuinely expensive trade. Proofread at phone size before you publish, every time. Our piece on what editing a post does to its reach covers the caption side of this.

You cannot give pages their own destination links. That is the carousel ad, and it is the single clearest functional difference between the two products. An organic document post has one caption and one comment thread. There is no per-page call to action that goes anywhere.

You cannot run two versions. No organic A/B test exists. If you want to know whether a different cover page works better, you are running the test across different posts on different days with different audiences, which is not a test.

You cannot rely on it being read as text. To a screen reader, a document post is a picture. If the substance lives only in the pages, summarise it in the caption so the post is not a blank rectangle for anybody using assistive technology. This is the accessibility point most carousel guides skip entirely.

Do document posts actually outperform other formats?

Probably somewhat, and the honest evidence is thinner than the confident claims suggest. We argue this properly in our piece on LinkedIn carousel posts, so here is the short version with the caveat attached.

In our cohort of 12,988 posts, document posts were 1.2% of the top-performing decile and 0.1% of the bottom half. That is the widest format gap in our data and the one we trust least, because it rests on 60 document posts from 13 accounts. Fifteen of them are in the top decile and nine in the bottom half. Twenty-four posts across both groups cannot support a conclusion, and we are not going to pretend otherwise. Their median engagement rate was 2.42 per 1,000 followers against a cohort median of 0.40, on that same 60-post sample.

A much larger external sample points the same way. Socialinsider's LinkedIn benchmarks, built from 1.3 million posts across 16,645 business pages between January 2024 and December 2025, put native documents at the top of their format table with a 7.00% engagement rate, ahead of multi-image (6.45%), video (6.00%), image (5.30%), text (4.50%), polls (4.20%), and link posts (3.25%). Their population is company Pages rather than personal profiles, and their engagement rate is computed per impression while ours divides by follower count. Different populations, different maths. What is useful is that two independent samples measured differently both put documents near the top.

For a format finding with a sample that can actually carry it, look at native video instead: 26.3% of our top decile against 10.1% of our bottom half, across 1,613 posts from 46 creators. We wrote that up in the piece on LinkedIn video posts. All of our format numbers are correlational rather than causal, and our cohort skews toward established creators with 1,000 or more followers, so treat the direction as more reliable than the magnitude in every case.

Document post or multi-image post: which should you use?

These are the two formats people actually confuse in practice, because both produce something you swipe. They are built for different content.

The test that resolves most cases: does the order carry meaning? A framework where page three only makes sense after page two is a document. Six photos from a conference where any order works is a multi-image post. If you force photographs into a PDF you get worse image quality and more work. If you force a sequential argument into a multi-image post you lose the page counter and the pacing that made it work.

Multi-image is also the format people under-use, which we cover in our guide to LinkedIn multi-image posts, including the LinkedIn specs for photo size and aspect ratio.

Do carousel ads and document posts do the same job?

No, and this is where most bad advice originates. A carousel ad is a paid LinkedIn Ads format. You build it in Campaign Manager, you attach a budget, and each card carries its own image and its own destination URL. It is a direct response unit. Its whole design is about routing clicks to different landing pages.

An organic document post has no budget, no targeting controls, no per-page links, and no landing page. It lives on your profile permanently and gets distributed by the feed according to the same rules as everything else you post.

So when a guide tells you to “use three to five cards” or “put a strong CTA on every card,” check which product it is describing. Card-count advice for ads is about media cost and click routing. It says nothing useful about how many pages an organic document should have, which is a completion-rate question. Six to twelve pages is the working range for organic, and the reasoning is in the carousel build guide.

How should you write the caption and the document title?

The document title is the most wasted field on LinkedIn. It appears under the frame in the feed, it is visible before anybody swipes, and the overwhelming majority of posts fill it with a filename. It is a headline. Treat it like one: state what the reader gets, in fewer than about eight words, in plain language.

The caption above it does a different job. It has to make the swipe feel worth it, and it is competing against the fold. The visible portion runs to roughly 210 characters on desktop and around 140 on mobile, per AuthoredUp's character limit breakdown, and everything past that costs a tap.

One pattern in our document posts is worth flagging precisely because it runs against the cohort. Across those 60 document posts from 13 accounts, 96.7% carried at least one hashtag and 68.3% carried four or more. Cohort-wide, heavy hashtag use points the other way: posts with four or more hashtags were 41.9% of the bottom half against 25.9% of the top decile. We are not going to resolve that with 60 posts, and we are not going to pretend the document number means anything on its own. The cohort-level pattern, which rests on all 12,988 posts, is the one to act on, and we break it down in the hashtag analysis. One to three relevant tags at the end is plenty.

Two more caption notes. Document posts in our sample said “you” somewhere in the visible hook 58.3% of the time, which is higher than the cohort's 36.4%, and second-person language is one of the few habits that separates our top decile (47.5%) from our bottom half (32.7%) across the full sample. And formatting matters more than usual here, because the caption is competing with a picture for attention. Short lines, one idea per line, no wall of text. Our guide to formatting a LinkedIn post covers the mechanics.

What are the most common mistakes with LinkedIn document posts?

1. Uploading through the image icon. Attach a PDF the wrong way and it becomes a file attachment rather than swipeable pages. Use the document icon.
2. Naming the document title after the file. Covered above, and it is the single most common one.
3. Designing at desktop scale. Your pages will be read at roughly the size of a playing card, on a phone, by someone doing something else. If you are shrinking the font to fit, cut the copy instead.
4. Putting the only link inside the PDF. It may not be clickable. Put it in the caption or the first comment.
5. Treating 300 pages as a target. It is a technical ceiling. What matters is the share of readers who reach the last page, and that falls with every swipe you ask for.
6. Uploading a deck that was built for a speaker. Client presentations and webinar decks assume somebody is standing next to them explaining. Without that person they are a stack of headings.

What happens to a document post after the first day?

Most format comparisons stop at the feed. Document posts are one of the few LinkedIn formats with a meaningful second life, and it changes how you should judge one.

A text post is consumed and gone. It exists in the feed for a day or two and then it is a line in your activity that nobody scrolls back to. A document post gets saved, screenshot, downloaded, and forwarded. Somebody sends page six to a colleague in a direct message. A team lead saves it and opens it again during a planning session three weeks later. None of that shows up as engagement in the first hour, which is where most people look.

That has three practical consequences. The first is that impressions are the wrong scoreboard for this format. A document that reaches 4,000 people and is saved by 40 of them is doing more for you than one that reaches 12,000 and is saved by three. Watch saves and comments rather than reach, and remember that comments are the scarcer currency across our whole cohort: the top-decile median was 72 comments against 5 in the bottom half, a far wider gap than reactions (235 against 80).

The second is that the document works as a reply asset. Once it exists, you have a link to send whenever somebody asks the question it answers. That is a slower loop than a viral text post and a more durable one, and it is the same argument for building a small library of reusable pieces rather than chasing individual posts.

The third is that the first hour still decides how many people ever get the chance to save it. Nothing about the format exempts you from the ordinary distribution mechanics, and the early comment window matters here as much as anywhere else, which we cover in the piece on LinkedIn's golden hour. A document post that nobody sees is not saved by anybody either.

Is the format losing its edge as more people use it?

Worth considering, and worth being honest that nobody outside LinkedIn can measure it properly.

Our own document sample comes from a period when the format was newer and rarer than it is now. Formats that are rare get an attention bonus simply by breaking the pattern of the feed, and that bonus decays as the format becomes common. If your feed is now full of pastel gradient slide decks with a bold claim on page one, whatever novelty premium those 60 posts in our cohort might have captured is already smaller than it was.

What does not decay is the underlying mechanic. A document post holds attention inside a fixed frame and asks for a series of small deliberate actions, which no other organic format on LinkedIn does. Saturation reduces the bonus for showing up in the format. It does not change what the format is good at, which is sequential explanation that a reader controls the pace of.

The strategic implication is unglamorous: stop asking whether carousels work in general and start asking whether yours work for your audience. Six documents over six weeks, measured against your own median rather than anyone's benchmark, will tell you more than any published study can, including ours.

So which term should you use, document post or carousel?

Both, deliberately. Say document post when you are talking about specs, uploads, and LinkedIn's own documentation, because that is the word LinkedIn uses and it will get you to the right help page. Say carousel when you are talking about design, structure, and ideas, because that is the word everyone else uses and it describes what the reader experiences.

The thing to be precise about is not the name. It is the boundary between the organic document post and the two products that borrow its vocabulary: the paid carousel ad, which has per-card links and a budget, and the multi-image post, which takes up to 20 photos and has no page counter. Get those three straight and most of the contradictory advice you will read about LinkedIn carousels sorts itself out.