LinkedIn Carousel Ads: 2,174 Real Examples, With Specs and Observed Durations

One creative per template, drawn from 309 companies across the industries below. Each caption carries the observed duration: how long AdPeekly watched that ad stay live, not an estimate. These 40 are a sample of 2,174 distinct carousel creatives in the full library.

Carousel is widely sampled and almost never scaled. Of the 430 companies where AdPeekly holds 200 or more published ads, 239 (55.6%) have run at least one carousel. But among the 309 companies running any carousel at all, the median one puts just 2.22% of its ads into the format, the 90th percentile is 10.87%, and only 34 of those 309 go past 10%. Roughly half of well-covered B2B advertisers have tried carousel; almost none of them commit to it.

The 200-ad denominator is deliberate. Adoption rises with how many ads AdPeekly holds for a company, so a figure computed across every tracked company would measure our own coverage as much as advertiser behaviour. Scope note: this describes the companies AdPeekly curates, not B2B advertisers as a whole.

About as long as everything else, and that is the finding. Across 941 carousel creatives with a reported start date, the median is 23 days, against a 19-day dataset median. That gap looks like a format effect and mostly is not.

The control that settles it is comparing each company against itself. Across the 24 companies running both carousel and single image at scale, carousel is the longer-running format at 10 of them, single image at 13, 1 are tied, and the median of the per-company differences is 1 day shorter for carousel. Carousel does not make a campaign run longer or shorter. Advertisers treat it like the rest of their mix; the pooled gap above is mostly which companies happen to run carousel.

The distribution's upper middle is where carousel campaigns run longest: the 75th percentile is 49 observed days, against a 25th percentile of 11 and a 90th of 116.

How to read these durations. They cover only creatives where LinkedIn reported a campaign start date, 941 of 2,174. For the rest, the only measurable span is how long AdPeekly has been watching, which describes our scraping schedule rather than the advertiser's decision, so those are excluded rather than mixed in. Full detail on the methodology page.

LinkedIn publishes the permitted limits and is the authority on them. What nobody else publishes is the gap between what is allowed and what gets used, which is the part that changes how you design a card.

On headlines, LinkedIn allows 45 characters. Across the 2,169 AdPeekly carousel creatives carrying one, the median is 34, the 90th percentile is 43, and only 83 of them, 3.8%, use the full allowance. The useful number is not the limit, it is the working range.

On shape, of the 2,059 carousel creatives whose first card AdPeekly captured at one of the two permitted ratios, 91.5% are 1:1 square and 8.5% are 4:5 vertical. A further 114 captures landed at other ratios, screenshot-fallback artifacts rather than carousels actually shipped at those shapes, so they are excluded rather than reported as if advertisers ship 16:9 carousels. One scope note: this is the first card as rendered in LinkedIn's Ad Library, not the uploaded asset, and a carousel has several cards of which AdPeekly observes one.

What surprised me building this page was how little format choice explains, once you control for which companies use it. Pooled across the whole corpus, carousel campaigns look 4 days longer than single image on the site's own duration methodology. Split by company, that gap nearly disappears: comparing each advertiser's own carousel campaigns against its own single-image campaigns, the median difference is 1 day, in single image's favor, not carousel's. Format choice does not make a campaign run meaningfully longer or shorter, and the pooled 4-day gap was mostly measuring which companies happen to run carousel, not the format itself. The more durable finding is adoption: just over half of well-covered advertisers have tried carousel at least once, and the ones who have put a median of 2.2% of their ads into it. Widely sampled, almost never the backbone of anyone's mix.

Daniele Viero, founder of AdPeekly