Episode 08: Carousels

First slide: three skincare bottles on a bathroom counter. One says texture. One says acne marks. One says dry patches.

You instantly ask, “Which one is mine?”

A carousel is controlled pacing. Every swipe is a tiny vote. The viewer is saying, “That last slide was worth one more second. I will give you the next one.”

That is the useful version of carousel advice: what works, what got stale, and how to earn the next swipe.

1. Stop starting with “make it saveable.”

That advice is everywhere. Make a checklist. Put one tip per slide. End with “save this for later.”

Some of that works. But “saveable” is an outcome, not a strategy.

Picture the stale version. Slide one says, “7 skin mistakes ruining your glow.” Slide two says, “Drink more water.” Slide three says, “Use sunscreen.” Slide four says, “Sleep better.” Slide five says, “Be consistent.” Final slide: “Save this for later.”

Big promise, generic tip, filler, filler, fake urgency. The carousel keeps asking for another vote without giving the viewer anything sharper.

The better question is: what behavior is this built for? A save needs future utility. A share needs a person on the other side. Trust needs proof.

Yale School of Public Health is a great non-creator example. One STI prevention carousel was basically, “Ask your besties which STI prevention method you are.”

That gives public health information a social use. You can see the group chat: “Which one are you?” “Send this to your bestie.” The share behavior is inside the idea.

That is the level most “save this” carousels never reach.

2. Slide one is not the carousel. Slide two is the receipt.

Gurus say, “make the first slide a strong hook.”

True, but incomplete.

Slide one makes a promise. Slide two proves the promise was not fake.

If slide one says, “5 mistakes ruining your skin barrier,” and slide two is “mistake number one: be consistent,” the viewer knows they are trapped in a recycled template.

If slide one says, “Your cleanser is making your cheeks worse,” and slide two shows two identical-looking cleansers with different active ingredients, there is a reason to keep swiping.

The first swipe is the most expensive tiny vote. Earn it with a receipt: screenshot, contradiction, before-and-after, weird example, number, proof object.

If slide two could appear in one hundred other carousels, slide one was probably bait.

3. Every slide needs one job.

A carousel is not a stack of tips. It is controlled pacing, where every swipe is the viewer voting to keep thinking with you.

So each slide should have one job: hook, proof, reveal, compare, explain, example, memory, or action.

If the final slide is just “follow for more,” you do not have an ending. You have an escape hatch.

This is why old Canva carousels feel so dead. Every slide has the same job: generic advice, generic advice, generic advice, call to action.

No escalation. No proof. No new belief.

The viewer can feel when a slide exists only because the template had a blank spot.

4. Know which kind of carousel you are making.

There are three different formats people keep mixing together.

First, a slide story. The order matters: recap, tutorial, reveal, before-and-after, or “pick one.”

Second, a proof stack. Each slide adds credibility: result, screenshot, customer, quote, process, number, outcome.

Third, a repurposed post. You take a thread, newsletter, podcast, webinar, or report and rebuild the useful idea for the feed.

The mistake is treating all three like the same template.

Yale’s STI carousel works like a slide story. The first slide gives you the social game: ask your besties. The next slides can reveal the options, make the public health information feel low-pressure, and give the viewer something to send.

An airline baggage-fee carousel can work like a proof stack. Slide one names the common pain: your carry-on might cost more at the gate. Slide two shows the size box. Slide three shows the fare rule. Slide four shows the actual choice: check the bag now, repack, or pay later.

A podcast-to-carousel post is repurposing. Different job, different structure.

5. Steal the mechanism, not the surface.

This is the rule that fixes most bad carousel advice.

Do not steal the Canva template. Steal the hierarchy: one dominant idea, one supporting proof point, one reason to swipe.

Do not steal the X screenshot. Steal the compression: a sharp text idea rebuilt into native slides.

Do not steal the corporate PDF look. Steal the professional object: make the post feel like a memo, benchmark, teardown, checklist, or field guide.

Do not steal the TikTok Photo Mode product list. Steal the decision support: make the viewer one swipe closer to choosing something.

The surface tactic ages. The mechanism survives.

6. LinkedIn carousels are work artifacts.

Socialinsider’s 2026 LinkedIn benchmark found native document posts leading business-post engagement. That means the format fits the feed; it does not make every PDF magic.

LinkedIn users are already in work mode. A carousel can feel like a memo, a benchmark, an audit, a case study, or a field guide.

That is why the baggage-fee example works. One slide saying “avoid airline fees” would feel thin. A sequence makes the proof feel earned: bag size, rule, exception, decision.

Each swipe changes the reader’s belief by one notch. “This rule is real.” “My bag might fail.” “The fee happens later.” “I know what to do before I leave.”

7. TikTok Photo Mode is not a corporate PDF.

TikTok supports a lot of photos in Photo Mode. That is permission, not a challenge.

The Ordinary can use a TikTok carousel as a visual decision aid for textured-skin products. Canva can use one as a tutorial. Those work because they are visual, quick, self-paced, and clear.

So do not upload the LinkedIn deck to TikTok and act surprised when it feels stiff.

On TikTok Photo Mode, the viewer should understand the point almost at thumb speed. A product list works if each slide helps a choice. A tutorial works if every frame shows a step.

Same controlled pacing, different size of vote.

8. Stop turning X threads into screenshot dumps.

This was one of the easiest repurposing hacks: write a thread, screenshot every tweet, post it as a carousel.

There are whole tools built around this. Orshot, Contentdrips, TwitterShots, all of that.

The tool demand is real. The mistake is thinking the export is the content.

Instagram’s 2026 originality push makes this more important. The Verge and Engadget both covered Instagram’s move against low-effort tweet roundups. They also flagged unoriginal photo and carousel reposts.

The weak version looks like this. Slide one is a screenshot of a viral tweet. Slide two is the next tweet. Slide three is the next tweet. Maybe there is a black background, a fake phone frame, or a “thread” label.

Nothing has been adapted. The viewer is paging through an export.

Take the insight and rebuild it. Turn one tweet into a cover promise. Turn the best reply into a counterpoint. Turn the proof into a screenshot, chart, quote, or example. Give the ending a cleaner belief.

Repurposing should feel like translation, not photocopying.

9. More slides do not mean more value.

Instagram gives you up to 20 photos or videos in a carousel. TikTok Photo Mode supports up to 35 photos.

That is a budget, not a goal.

The viewer still has to fund every swipe with attention.

Ned Potter’s University of York Library teardown is useful here. Carousels became a huge part of their best-performing Instagram posts.

But even that example needs a caveat: carousel image views can inflate totals because multiple image views can count separately.

The lesson is smaller and more useful: an institution with useful messages found a lower-effort format students could move through at their own speed.

Use as many slides as the idea can earn. No more.

10. The final slide should change the first slide.

Most carousel endings are weak because they panic.

“Follow for more.”

“Save this.”

“Comment guide.”

Those can be fine as business mechanics, but they are not a payoff.

The fake CTA version is obvious. Slide one says, “The exact routine that fixed my skin.” The middle slides give generic advice. Final slide: “Comment ROUTINE and I will send it to you.”

That is a lead magnet wearing a costume. The viewer gave you nine tiny votes, and you spent the last one asking them to work for the real answer.

The final slide should make the viewer feel like the first slide has been resolved.

If the first slide made a promise, the final slide should deliver the cleanest version of the answer.

If it was a slide story, the last slide should reframe the beginning.

Generic carousels ask, “How do I fit my advice into slides?”

Good carousels ask, “What does the viewer need to believe before they choose the next swipe?”

That is the practical test.

Open your next carousel draft and write the job of every slide in the margin. Promise. Receipt. Proof. Example. Tension. Payoff. Action.

If you cannot name the job, cut the slide.

If the job repeats, merge it.

If the job is “vibe,” rewrite it.

If the final slide says “save,” “comment,” or “follow,” ask what the viewer gets if they do nothing except finish. If the answer is “not much,” the CTA is covering for a weak payoff.

Because every swipe is a tiny vote. The best carousel is the one where every slide makes the next vote feel earned.