How Many Slides Should a LinkedIn Carousel Have?

Slide count is usually the first thing people worry about, and it is the wrong thing to worry about first. The number that matters is however many slides your one idea takes to land. Everything else — format caps, platform limits, what someone told you about the algorithm — sits underneath that.

Still, the caps are real and they differ by format, so it is worth knowing exactly what you are working inside before you start designing.

Why does the answer depend on which format I use?

On LinkedIn, “carousel” means two different things, and they have very different slide limits.

• A multi-image post — several separate image files in one post, which the reader swipes through. Kalovio publishes these with two to nine images, and that range matches what the LinkedIn post composer offers most members. If you need the current maximum, check LinkedIn’s own help pages, because composer limits do get adjusted.
• A document post — you upload one file, usually a PDF, and LinkedIn renders each page as a swipeable slide. This is what most people mean when they say “LinkedIn carousel”. The page ceiling here is far higher than nine.

There is a third thing called a carousel ad, which is a paid LinkedIn Ads format with its own rules. It is not what this page is about, and mixing the two up leads to advice that does not apply.

So how many slides is right?

Here is a sensible starting point by purpose. Treat these as defaults to argue with, not rules.

If you took one number away from this page, take somewhere between five and ten. It fits inside the multi-image cap, it is long enough to develop an idea properly, and it is short enough that a reader on a train platform will actually reach the end.

Do longer carousels get more reach?

Nobody outside LinkedIn can answer this with real evidence, and you should be suspicious of anyone who says otherwise. LinkedIn does not publish how its feed ranking works, so every confident claim about slide counts and reach is either someone’s private testing on their own account or, more often, a number invented and then repeated until it sounds official.

What is widely observed among people who post regularly is softer than that, and it is about attention rather than mechanics. People swipe while distracted. Every extra slide is another moment where they can stop. A carousel that people finish is doing better work than one they abandon on slide four, whatever the raw impression count says.

So the useful framing is not “how many slides does the algorithm like”. It is: how many slides can I make good enough that a reader keeps going. That number is different for everyone and different for every topic.

What is the trade-off between depth and completion?

This is the whole problem in one sentence: more slides let you say more, and fewer slides get read to the end. You cannot have both, so you have to choose which one this particular post needs.

Depth wins when the subject genuinely has parts. If you are walking through a seven-step process, cutting it to four slides does not make it tighter — it makes it wrong. Squeezing three steps onto one slide is worse than having three slides.

Completion wins when you are making an argument. An argument has one point. Stretching it across fifteen slides does not make it more persuasive; it makes the reader look for the exit. Most carousels that feel long are arguments pretending to be guides.

A rough test: go through your draft slide by slide and ask what would be lost if that slide vanished. If the answer is “nothing much” on three or four slides, you are padding. Padding is the most common reason a carousel is too long, and it is almost always the slides in the middle.

Signs you have too many slides

• Two consecutive slides make the same point in different words.
• A slide exists only to say “now let us look at the next part”.
• You added slides to hit a round number.
• You cannot summarise the whole thing in one sentence any more.
• The last third is noticeably thinner than the first third.

Signs you have too few

• A slide has so much text you shrank the font to fit.
• You are covering two separate ideas on one slide.
• The reader has to already know something you never explained.
• It ends abruptly, with no closing thought.

Cramming is the mirror-image failure of padding, and on a phone screen it is the more damaging of the two. A twelve-slide carousel that breathes beats a six-slide one you have to pinch and zoom.

How should I structure the slides I do use?

Whatever your total, the shape is roughly the same, and it is worth budgeting your slides against it.

Slide one carries the whole post. It is the one slide every reader sees, because it is what shows in the feed. It should say what the post is about in plain words, not tease it. Vague covers cost you readers who would have stayed for the actual content. Budget real effort here, not the last five minutes.

The middle slides do one job each. One step, one point, one example per slide. The moment a slide has two jobs, split it. This is also where slide count is decided in practice — count your points, and that is roughly your middle.

The last slide gives the reader something to leave with. A short summary of the idea, or the single thing to do next. A closing slide that only says “follow me for more” wastes the one slide where you have already earned attention. Say the takeaway first, then ask for something if you want to.

So a nine-slide carousel is realistically one cover, seven content slides, one close. Working backwards from that is often easier than counting forwards: how many real points do I have, plus two.

When is a long carousel actually the right call?

Long works when the length is the point. A genuine reference — a full checklist, a set of examples, a walkthrough with a screenshot per step — earns its pages, because the reader came for completeness. People do save and return to those.

Long fails when it is one thin idea inflated with quotes, filler and section dividers. The reader can feel the difference within three swipes.

If you are going long, front-load ruthlessly. Assume most readers will see the first few slides and stop. The value has to start immediately, not build to a reveal on slide eighteen.

How do I tell whether my slide count was right?

Honestly, imperfectly. LinkedIn’s own post analytics show you impressions and engagement on the post, which tells you whether people arrived and reacted. It does not hand you a clean per-slide drop-off figure, so anyone quoting you a precise completion rate is estimating.

What you can do is compare your own posts against each other over time. Publish a few at different lengths on similar topics, and look at which ones people actually engaged with rather than just saw. Your own account is the only benchmark that is genuinely about your audience.

Does the tool I publish with change any of this?

Only in what it will accept. The editorial judgement is yours either way.

If you publish through an AI assistant, the format split still applies. Kalovio is a free hosted connector for Claude and ChatGPT that publishes both kinds: a multi-image post of two to nine images, or a PDF document post for anything longer. Worth knowing before you rely on it: both go through LinkedIn’s newer versioned interface, which is not switched on for every self-serve app, so a refusal is reported plainly rather than hidden — try it on your own account. Document posts also publish immediately; scheduling them is not supported yet, though multi-image posts can be scheduled.

None of that changes the answer to the question. Count your real points, add a cover and a close, and cut anything that repeats. If that lands you under nine, a multi-image post will do. If it does not, build a PDF — and be sure the extra pages are carrying weight.