Creative Formats & Assets

LinkedIn Carousel Ads

By the AdFlint research team · Last reviewed November 2026

A horizontally swipeable set of cards in the LinkedIn feed, each holding its own image, headline, and landing destination.

You get multiple cards under one sponsored post, which suits sequencing a multi-point argument, listing services, or walking through a framework, and each card can point at a different page. It only works when the first card earns the swipe. The mistake is spreading one idea thinly across cards nobody reaches; put the strongest proof on card one and treat the rest as depth.

Key takeaways

  • LinkedIn Carousel ads hold 2 to 10 swipeable cards, and each card can have its own image and link.
  • Card one does most of the work, so put the strongest promise or proof there.
  • More cards is not better: two or three strong cards can beat ten weak ones.
  • Match each card's destination to its content, or clicks will be wasted.

In practice.

A LinkedIn Carousel ad is a Sponsored Content unit that holds a row of swipeable cards in the feed. LinkedIn's ad guide lists 2 to 10 cards, a recommended size of 1080 by 1080 pixels, JPG or PNG files, and a maximum of 10 MB per file. Each card has its own image and can link to its own destination, so a single ad can point to several pages. You pick the Carousel format in Campaign Manager, upload the cards in order, write the intro text, and add a headline for each card. The ad is published through a company page and appears between organic posts. Think of the card sequence as a mini story with a beginning, a middle, and a next step.

The mechanism is simple: a member sees the first card and decides whether to swipe. Everything after card one depends on that decision, so the first card carries most of the weight. From there the reader moves card by card, and you can use the sequence to build an argument, show steps in a process, list services, or present several customers. Because each card can link to a different page, the format lets you test which topic draws clicks. The intro text above the cards appears with every card, so it should work as a standalone headline for the whole ad rather than a lead-in to card one only. Each card should make sense on its own, since some readers will stop after the first.

Carousel works with the same audiences, objectives, and bidding options as other Sponsored Content, so it competes with Single Image and Video ads for the same people. Compared with a Single Image ad, it carries more information, and compared with a Document ad, it is lighter and the cards link out individually. It does not gate content on its own, so lead capture happens at the destination or with a Lead Gen Form attached to the ad. The number of cards is a creative choice, not a setting that improves delivery. Two or three strong cards can outperform ten weak ones. Keep the visual style consistent across cards, with the same fonts and colors, so the unit reads as one ad.

The usual mistake is spreading one idea across cards that nobody reaches. Most members see card one and perhaps card two, so a punchline on card eight is wasted. Put the strongest proof or the clearest promise on card one and treat the later cards as depth. A second mistake is using a card per feature with no story, which reads like a catalog. A third is inconsistent design, where cards look like they belong to different ads. Also check that each destination matches its card, since a click on a pricing card that opens a homepage loses the reader. Avoid using the carousel to hide weak content behind a swipe, because readers decide quickly whether to continue.

Build a three-card version first. Card one states the result you offer, card two shows the proof, such as a number or a customer, and card three gives the next step. Use consistent colors and large text that reads on a phone. Run it against a Single Image ad with the same audience, budget, and offer for two weeks. Compare click-through rate and cost per lead. If the carousel wins, expand to five cards and test whether the added cards help or just add scrolling. If it does not win, keep the best card as a single image. Look at card-level clicks where available, and rebuild the order if the later cards never get clicked.

Worked example

A three-card test against a single image

Suppose an HR consultancy runs a three-card carousel and a Single Image ad to the same audience with $1,000 each. The single image gets 40,000 impressions and 280 clicks, a click-through rate of 0.7%. The carousel gets 40,000 impressions and 360 clicks, or 0.9%.

Looking at the cards, 70% of clicks come from card one and 20% from card two. Card three, the call to action, gets only 10%. The consultancy moves the offer onto card two and tests again, this time with a two-card version. The data showed that readers rarely reached the last card, and the change came from the click distribution rather than a guess.

LinkedIn Carousel Ads compared with

The settings this gets confused with, and how to tell them apart.

Common questions.

How many cards can a LinkedIn Carousel ad have?

LinkedIn's ad guide lists 2 to 10 cards, with a recommended size of 1080 by 1080 pixels in JPG or PNG, up to 10 MB per file.

Can each carousel card link to a different page?

Yes. Each card can have its own image and link, so one ad can point to several pages.

Is a carousel better than a single image ad?

Not automatically. A carousel carries more information, but most members see only the first cards. Test both with the same audience and budget and compare cost per lead.

What should go on the first card of a carousel?

The strongest promise or proof you have. The first card decides whether anyone swipes, so treat later cards as supporting detail.

Sources

You should not need to know this to advertise.

AdFlint handles the settings for you, inside the Google and Meta accounts you already own.

Try AdFlint free