Writing the First Slide of a LinkedIn Carousel

A LinkedIn carousel is a stack of pages the reader swipes through — either a PDF you upload, which LinkedIn renders as swipeable pages (a document post), or a set of 2 to 9 images in one post (a multi-image post). Either way, only one page is visible before someone decides whether to engage. That is slide one. Everything you wrote on slides two to nine is invisible until it earns the swipe.

That is why the first slide deserves more of your time than the rest of the deck combined. It is not the title page of a document. It is the cover, the headline and the whole pitch, compressed into something roughly the size of a postage stamp on a phone.

Why does the first slide matter more than the others?

Three things stack up on that one page.

It is the thumbnail. In the feed, your carousel appears as a single preview image with the caption above it. LinkedIn does not show slide two. So the first slide is doing the job a headline does on an article — except it is an image, so nothing about it is searchable, and the reader is scrolling with their thumb.

It is seen small. Much LinkedIn browsing happens on a phone, where the preview is a few centimetres across. Text that looks fine on your laptop at full size can turn into grey mush at that scale.

It sets the contract. Whatever the cover promises is what the reader expects to receive. If slide one says “5 pricing mistakes” and the deck delivers three vague ones plus two slides of self-promotion, you did not just lose that reader — you taught them what your posts are worth.

What should the first slide actually say?

Make a specific promise. That is the whole rule, and almost every weak cover breaks it in one of two ways: it states a topic rather than a promise, or it makes a promise so broad it means nothing.

A topic tells the reader what the deck is about. A promise tells them what they will walk away with.

Three ingredients make the right-hand column work, and you can check any cover against them in seconds:

• Who it is for. Either stated outright (“for first-time founders”) or implied strongly enough that the right person recognises themselves.
• What they get. A number, a named outcome, a mistake to avoid, a decision made easier. Something they could not have written themselves.
• Why you. Not a credentials block — just a hint that this comes from doing the thing rather than reading about it. “What we got wrong” carries that. “Best practices for outbound” does not.

Numbers work on covers for an unglamorous reason: they tell the reader how long this will take. “4 mistakes” is a finite commitment. “Everything you need to know about pricing” is open-ended, and open-ended loses to a thumb.

How many words should be on the first slide?

Fewer than you want. A useful working limit is about six to ten words in the main line, plus an optional short second line. Not because there is a rule, but because the cover has to survive being shrunk, and every extra word forces the type smaller.

Most first drafts can lose a third of their words with no loss of meaning. The usual candidates:

• Warm-up phrases. “A few thoughts on…”, “Here is why…”, “In this carousel I will…”. Cut all of them. The reader can see it is a carousel.
• Hedges. “Some”, “a bit”, “maybe”, “possibly”. Hedging on a cover reads as having nothing to say.
• Your own branding. A small logo is fine. A logo plus a tagline plus a handle plus a website plus a headshot is five things competing with the one thing that matters.
• Slide furniture. “Slide 1 of 9”, a date, a decorative subtitle. Nobody swiped because of a page number.

A structure that holds up well: one large line carrying the promise, one small line carrying the qualifier. For example, a large “4 pricing mistakes that cost us 10 deals” above a small “for B2B founders selling their first product”. The big line stops the scroll; the small line confirms it is meant for them.

Will anyone be able to read it at thumbnail size?

This is where good copy dies. Legibility at small size is not a design nicety on a LinkedIn cover — it is the difference between a promise and a grey rectangle.

What reliably survives being shrunk:

• High contrast between text and background. Dark text on a light background, or light text on a genuinely dark one. The accessibility standard most teams follow, WCAG 2.1, asks for a contrast ratio of at least 4.5:1 for normal text and 3:1 for large text. Free contrast checkers give you that number in seconds, and it is a sensible floor for images too.
• Big, plain type. The headline should be the largest thing on the slide by a clear margin. Thin, decorative or very light-weight fonts are the first casualty of shrinking; a medium or bold weight holds its shape.
• Space around the words. Crowding text to the edges makes it harder to read and risks an awkward crop. Leave a generous margin.
• A calm background behind the text. Photos, gradients and patterns behind a headline reduce contrast unpredictably. If you want a photo, put a solid block or a strong overlay behind the words.

One test catches almost everything: export the first slide, shrink it to roughly the width of your thumb on your phone, and look at it from normal reading distance. If you cannot read the promise, the slide has failed, however good the sentence is.

What hook patterns work without being clickbait?

A hook is honest when the deck actually delivers what the cover implies. It becomes clickbait when the cover withholds the answer on purpose to force the swipe. The difference is not the wording — it is whether the payoff arrives.

Patterns that hold up:

• The counted list. “7 questions to ask before signing an agency contract”. Finite, honest, easy to deliver.
• The specific mistake. “We priced per seat for two years. It nearly killed our renewal rate.” Concrete, and it admits something, which is rare enough to stand out.
• The before and after. “Our old onboarding email, and the one that replaced it”. The reader can picture the payoff exactly.
• The correction. “Post daily is bad advice for most B2B teams. Here is what we do instead.” Takes a position the deck can defend.
• The plain how-to. “How to run a 20-minute customer interview”. Unfashionable and consistently effective, because it is a promise you can obviously keep.

Patterns that erode trust after a couple of posts:

• Manufactured shock. “Nobody is talking about this”, “This changes everything”. Someone is, and it does not.
• The withheld answer. “The one thing that fixed our funnel — slide 9”. The reader feels the transaction, and the format is not long enough to justify it.
• Borrowed numbers. A big multiplier on the cover that you cannot source. If a figure is yours, say whose it is and over what period. If it is not, leave it off — a cover built on a number the reader can smell as invented poisons the whole deck.
• Fake vulnerability. “I was rejected 47 times…” with a business lesson stapled to the end. Readers have learnt this shape.

A quiet test before you post: if the first slide were the only thing anyone ever saw, would it still be true and still be useful? Covers that pass that test tend to get swiped anyway.

How do I check the first slide before I publish?

Five checks, in order. Any failure sends you back to the words, not to the design.

• The promise check. Read only slide one, then say out loud what you expect to get. If you cannot, the reader cannot either.
• The thumbnail check. Shrink it and look from arm’s length, as above.
• The delivery check. Read slide one, then the last slide. Was the promise kept?
• The swap check. Could a competitor put their logo on this cover unchanged? If yes, it is too generic to be worth your morning.
• The caption check. Your caption sits above the image in the feed. It should add to the cover, not repeat it word for word.

How do I publish the carousel once the cover works?

The cover is the hard part. Publishing is not. You can upload the PDF or the images through LinkedIn directly, or, if you already draft in an AI assistant, publish from there.

Kalovio is a free hosted connector that lets you publish a document post (your PDF carousel) or a multi-image post by chatting with Claude or ChatGPT. It shows a preview and waits for your confirmation before anything goes live, so the cover you approved is the cover that publishes. Two honest caveats: carousels 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 rather than assuming; and document posts publish immediately — they cannot be scheduled yet.

Whatever you publish with, the sequence is the same. Write the promise first, in plain words. Cut it until it fits. Make it big and readable. Then build the rest of the deck to keep it.