Common LinkedIn Carousel Mistakes

A carousel takes real effort to make, which is why a flat one stings. The frustrating part is that the fault is rarely where people look for it. Designers blame the design. Writers blame the caption. In practice the same short list of problems shows up again and again, and most of them are fixable in ten minutes without opening a design tool.

Here is the honest list, with why each one fails and what to do instead.

First, are you even making the right kind of carousel?

This is mistake zero, and it causes a surprising amount of confusion. On LinkedIn today, “carousel” can mean two different organic post types, plus one paid format that is unrelated:

• A document post. You upload a PDF (or a PowerPoint or Word file) and LinkedIn renders it as swipeable pages inside the feed. This is what most people mean when they say “LinkedIn carousel”.
• A multi-image post. You upload several separate images — up to nine — in one post, and the reader swipes between them.
• A carousel ad. A paid LinkedIn Ads format. It lives in the advertising system, costs money, and has its own rules. It is not what you get by posting from your profile.

The mistake is designing for one and publishing as the other. A document post keeps your slides in a fixed order and shows a page counter; a multi-image post behaves more like a photo gallery. Neither is wrong, but they read differently, and the crops differ. Decide which one you are making before you start laying out slides, and check LinkedIn’s own help pages for the current file types and limits rather than trusting a number you read in a blog post two years ago — those specifications change.

Why does nobody swipe past my first slide?

Why it fails. The cover slide is the only one every reader sees. In a scrolling feed it competes with everything else, and it usually loses because it says what the carousel is about instead of giving a reason to keep going. “5 Lessons From Q3” is a label. Labels do not create curiosity, and a reader who does not swipe never sees the other nine slides you spent an afternoon on.

The fix. Make the cover a promise or a tension, not a title. Name the specific problem the reader has, or state the surprising conclusion up front. “We cut our onboarding from 14 days to 3. Here is what we removed.” is a reason to swipe. So is a plain question the reader is actually asking. Keep it short enough to read at a glance, and put the biggest words at the top where the crop will not eat them.

How much text should go on one carousel slide?

Why it fails. The most common design mistake is treating a slide like a paragraph. People paste in a full explanation, shrink the type until it fits, and end up with a wall of small grey text that nobody reads on a phone in a lift. The carousel format rewards pace — each swipe should feel like progress. Dense slides make it feel like homework.

The fix. One idea per slide. If a slide contains two ideas, it is two slides. A practical working limit is a short headline plus one or two supporting lines, and if you cannot say it that briefly, the idea is not sharp enough yet. Cut adjectives before you shrink font size. When a slide genuinely needs more detail, move the detail into the caption below the post, where long text is normal and readable.

Why does my carousel look fine on my laptop and terrible on a phone?

Why it fails. You design it at full size on a large monitor, where small body text looks perfectly readable. Then it appears in the feed on a phone screen, scaled down to something around the width of a playing card, and that same text is unreadable. Most people scrolling LinkedIn are on a phone. Designing for the screen you are designing on is the classic trap.

The fix. Two habits solve almost all of it. First, use type that feels absurdly large on your monitor — if it looks slightly too big on the desktop, it is probably about right on a phone. Second, before publishing, look at the file on your own phone. Not a preview, not a thumbnail: open the actual PDF or image on the device your readers use. Anything you have to squint at gets cut or enlarged. Also keep important content away from the very edges, because crops and rounded corners vary between the LinkedIn app, the mobile browser and the desktop feed.

What is the reader supposed to take away?

Why it fails. Plenty of carousels are pleasant to swipe through and completely forgettable. They list observations without ever landing on a conclusion. The reader reaches the end, feels vaguely informed, and moves on. Nothing gets saved, nothing gets shared, and nothing changes for them.

The fix. Decide the single sentence you want the reader to repeat to a colleague, and write it before you make any slides. If you cannot write that sentence, the carousel is not ready. Then make sure it appears in plain language somewhere in the last third — not implied, not hinted at, actually written down. A carousel that teaches one thing properly beats one that gestures at seven things.

Do I need a call to action at the end?

Why it fails. Two opposite mistakes live here. The first is ending on nothing — the slides simply stop, and an interested reader has no idea what to do next. The second is ending on a wall of asks: follow me, comment below, download the guide, book a call, share this. Five asks are the same as none, because the reader cannot tell which one you actually want.

The fix. One ask, matched to the effort you have earned. A carousel that taught something useful can reasonably ask for a comment or a follow. A carousel that solved a real, specific problem can reasonably point to a longer resource. Make it specific enough to act on immediately: “If you want the checklist version, say checklist in the comments” works better than “let me know your thoughts”. And remember that text inside the slides is generally not clickable — the reliable place for a link is the caption.

Why does my carousel look amateurish even though the content is good?

Why it fails. Inconsistency, almost always. The heading sits near the top on slide two and halfway down on slide three. Slide four is suddenly a different shade of blue. One slide uses sentence case, the next uses title case. Individually these are invisible; stacked across ten slides they read as carelessness, and carelessness quietly undermines the argument you are making.

The fix. Build a tiny system before you build slides: two type sizes (headline and body), two colours plus a background, and one fixed margin. Then never break it. Reuse the same slide as a starting point for every new one rather than creating each from scratch. If you make carousels regularly, save the finished file as your template — consistency across posts is what makes people recognise your work in the feed before they see your name.

Should the main point go on the last slide?

Why it fails. Writers trained on essays save the conclusion for the end. In a feed, that is backwards. Every swipe loses some readers — not because of any ranking mechanic, simply because attention drains. If your best insight is on slide eight, many of the people who opened the carousel never reach it. You are hiding your strongest material behind your weakest.

The fix. Front-load. State the conclusion by slide two or three, then use the rest to show the working, the caveats and the how. This feels wrong the first time and works better almost every time. A reader who gets the answer on slide two and stops has still received value from you. A reader who quits on slide five having received only build-up has received nothing.

The mistakes, side by side

What about the caption and the posting itself?

Two smaller mistakes are worth naming, because they undo good work at the last step.

The first is a caption that just repeats the cover slide. The caption is the only searchable, selectable, clickable text in the whole post — it is where a link belongs, where context belongs, and where anyone who does not swipe still gets something. Repeating the cover wastes it.

The second is publishing at an awkward moment simply because that is when you happened to finish. Making and posting are separate jobs, and treating them as one is why good carousels go out at 11pm on a Saturday. Write it when you have the ideas, publish it when your readers are around.

If you build your slides with AI assistance, you can also publish them straight from the chat. Kalovio connects Claude or ChatGPT to your own LinkedIn account through LinkedIn’s official sign-in, so you can upload a PDF and post it as a document carousel, or send two to nine images as a multi-image post, and approve the exact wording before anything publishes. Two honest caveats: multi-image and document posts go through LinkedIn’s newer versioned interface, which is not switched on for every self-serve app, so it is worth trying on your own account — a refusal is reported clearly rather than failing quietly. And document posts publish immediately; scheduling them is not supported yet, although text and single-image posts can be scheduled.

A short pre-publish check

• Can a stranger tell from slide one what they would gain by swiping?
• Does any slide hold more than one idea?
• Have you opened the file on your own phone, at actual size?
• Is the conclusion visible by slide two or three?
• Is there exactly one ask at the end?
• Do headings sit in the same place on every slide?
• Does the caption add something the slides do not?

None of that requires better design skills. It requires deciding what you are saying before you decide how it looks, which is the mistake underneath all the others.