What are the best ways to run carousel ads on LinkedIn?

Let me start with a confession. The first LinkedIn carousel campaign I ever ran had beautiful cards, tight copy, and a click-through rate that made me want to close my laptop and go work in a bakery.

The problem wasn't the creative. It wasn't the targeting. It was that I was treating each element like a separate task to complete, not a single system to build. Fix the creative, check. Write the headlines, check. Slap on a landing page, check. Ship it.

That is exactly how you burn a LinkedIn budget and learn nothing from it.

What follows is what I wish someone had told me before that first campaign. Not a checklist. A way of thinking.

Why carousel ads earn their higher cost in B2B pipeline campaigns

Let's talk about the cost problem first, because it's the first objection you'll face internally.

LinkedIn is expensive. CPCs run three to five times higher than Google Ads. Your CFO will notice. Your VP of Demand Gen will ask you to justify it. And honestly? They should.

So here's the justification, and it starts with one question: what format are you comparing it to?

A single-image ad delivers one message. One argument. One angle. It fires once and either hits or it doesn't. Now consider who you're selling to. Most B2B buying decisions involve six to ten stakeholders, each with different priorities, different fears, and different definitions of "good enough to move forward." A single-image ad is trying to speak to all of them at once, which means it's effectively speaking to none of them particularly well.

A carousel delivers a sequence. That's the core argument.

According to Socialinsider's 2025 LinkedIn Benchmarks Report, multi-image carousels average a 6.60% engagement rate. That's the highest of any LinkedIn content format. And for optimized B2B SaaS campaigns, CTR can reach between 0.8% and 1.2%, compared to a platform median somewhere between 0.45% and 0.65%. That ceiling is meaningfully higher. Not guaranteed. But reachable.

Four out of five LinkedIn members drive business decisions. Sixty-one million senior-level influencers are active on the platform. The format's cost reflects who it reaches, and the format's structure reflects how those people actually buy.

But here's where I'll push back on myself a little. Carousel ads are not the right unit for every campaign objective. If you're running pure awareness plays at the top of the funnel, you're probably wasting the format on people who haven't yet registered they have a problem worth solving. Carousels earn their keep in the middle and bottom of the funnel, where multi-argument persuasion is the whole game.

Use them there. Don't use them everywhere.

The card sequence that moves a B2B buyer from scroll to click

Most people design carousel cards. Fewer people design carousel sequences. That distinction is everything.

Think of the sequence like a short conversation. You don't open a conversation by pitching your product. You open it by saying something the other person immediately recognizes as true about their own situation.

Card one is not about you. Card one names the pain. Its only job is to earn the swipe to card two. That's it.

From there, cards two through four should each carry one aspect of the solution, framed as an outcome rather than a feature. "Reduces onboarding time by 60%" lands differently than "automated onboarding workflow." One tells someone what their life looks like after. The other describes a thing that exists. People buy the former.

Cards five through seven are where you bring proof. Client results. Specific metrics. Named outcomes. This is the section where specificity does the heavy lifting, and vagueness kills you. "Customers love us" is not proof. "One customer cut implementation time from 14 weeks to 4" is proof.

The final card is a single, specific call to action matched to what the sequence has earned. If you built toward a demo, ask for a demo. If you built toward a case study, offer the case study. Don't pivot at the last second.

Now, a few practical calibrations on length. Start with three to five cards and test adding more. More cards is not automatically better. Sequence coherence matters more than sequence length. A tight five-card story beats a meandering eight-card one every time.

One structural decision worth making consciously: if you're telling a full story and want to reward the person who swipes all the way through, put the primary CTA on the last card. If you're targeting warm audiences who already know you, an earlier CTA can work because they don't need the full setup.

And here's the thing about card order that most people skip entirely. The sequence is a hypothesis. You think the pain-point-first approach works. But what if your audience finds a specific outcome more compelling as an entry point than the problem framing? You won't know unless you test it. Swapping cards one and three and running both versions is one of the more instructive tests you can run.

Creative and copy decisions that determine whether cards get swiped

Here's something subtle that I've seen separate good carousels from forgettable ones. Visual continuity.

When each card looks like it belongs to the same story, a consistent color palette, the same typographic system, a coherent layout, the reader feels pulled forward. When cards look loosely connected, the brain registers it as a series of disconnected ads, and swipe momentum dies.

A significant share of LinkedIn's users access it on mobile. That means your text needs to be legible at a small size without squinting, and your visual needs to communicate its point in about two seconds of glancing. If a card requires more than that to decode, you've already lost the swipe.

The discipline that keeps swipe rates up is one idea per card. One visual, one message, one point. It sounds obvious. It is also constantly violated in practice, usually because someone in the approval chain wants to add one more thing.

The 45-character headline limit for URL destinations is a useful constraint, but it creates a challenge. That headline needs to stand alone. Someone who pauses on card four without having seen cards one through three should still understand what's being offered. Write every headline as if it's the first thing someone sees, because sometimes it is.

On the question of voice: founder-led or executive-voice creative, where the card reads as a considered point of view from a named person rather than a brand, tends to outperform generic company creative in B2B contexts. Why exactly does this happen? Because B2B buyers are also people, and people respond to other people. A named perspective carries credibility in a way a logo doesn't.

One production shortcut worth knowing. A blog post can become a card sequence. A webinar's key findings can become the proof cards. You do not need to generate net-new content for every carousel. Repurposing reduces the production lift without sacrificing relevance, and it often surfaces content that already performed well organically.

It's also worth considering Thought Leader Ads as a companion format. These promote organic posts from executives and deliver meaningfully higher CTR than standard company ads. They're worth running alongside a carousel campaign or testing against one to see what your specific audience responds to.

How targeting and audience segmentation shape carousel performance

All of that work on sequence and creative becomes irrelevant if the wrong person sees it.

Carousel ads belong at the middle and bottom of the funnel. The multi-argument format is largely wasted on cold audiences who haven't yet registered a problem worth solving. You're running a persuasion sequence for someone who has no context for why they should be persuaded.

For middle-to-bottom-of-funnel targeting, layer Matched Audiences on top of firmographic filters. Website visitors from pricing pages, demo pages, and other high-intent URLs are telling you something through their behavior. The LinkedIn Insight Tag captures that signal. Use it.

Audience size matters more than people give it credit for. Target somewhere in the range of 50,000 to 300,000 members. Too narrow and you'll burn frequency before the campaign has time to work. Too broad and relevance collapses, and you're back to speaking to no one particularly well.

Multi-dimensional segmentation is where this gets interesting. Layer company size, industry, and revenue filters with known technologies your audience uses, and add off-platform intent signals from providers like Bombora or ZoomInfo. Someone who fits your ICP and is actively researching your category is a fundamentally different audience than someone who just fits your ICP.

But here's the segmentation point that most carousel campaigns miss entirely. A CFO and a VP of Engineering are both on the buying committee. They are not the same audience. Running the same carousel at both of them is a compromise that serves neither particularly well.

A CFO-targeted sequence leads with ROI and cost-efficiency. A VP of Engineering sequence leads with integration depth and implementation speed. These are different stories, structured differently, emphasizing different proof points. The creative architecture from the previous section only works if the right version reaches the right person. That means different campaigns, different sequences, different tracking.

Spray-and-pray targeting across multiple industries or roles at once doesn't just dilute impact. It actively drives up CPCs without improving pipeline quality. The targeting decision is itself a creative decision.

Using per-card URL tracking to read buying intent by stakeholder

This is the part that most teams either skip or underuse, and it's one of the more powerful things the carousel format makes possible.

Each card in a URL-destination carousel can carry a unique destination URL. On the surface, that sounds like a UX convenience. In practice, it's a data collection mechanism.

Think about what a click on a specific card actually means. A CFO clicking the ROI-data card and a VP of Engineering clicking the technical documentation card are sending different signals about where they are in the buying process and what argument they found compelling enough to act on. That's not just useful for reporting. That's sales intelligence.

Per-card click data also reveals which argument in the sequence resonated most across your audience. Over time, this shapes which cards get promoted to card one in future campaigns. The sequence improves because the data is telling you what's actually landing.

That intent data can be passed to your CRM and used by sales to personalize outreach. "I noticed you engaged with our implementation timeline content" is a warmer opening than a generic follow-up. It's not creepy if it's relevant. It's the opposite of creepy. It's evidence that you understood what they cared about.

Here's the trade-off worth sitting with. Lead Gen Form carousels, where all cards link to a single form, give you the lead. They do not give you the card-level intent signal. You gain contact information and lose the behavioral data. Whether that trade is worth making depends on your campaign objective and how sophisticated your sales team is at using intent data. There's no universal right answer.

The practical starting point: name your destination URLs with card-specific UTM parameters. Make the signal readable in whatever attribution system you're already using. You don't need a new tool. You need consistent naming conventions.

Landing pages that complete the carousel's argument instead of restarting it

The most common failure mode I've seen in carousel campaigns isn't the creative. It's this: a carousel that tells a specific, carefully constructed story drops the clicker onto a generic homepage that ignores everything they just read.

The carousel did its job. The landing page started over.

The landing page should pick up where the card left off. If card five was a specific customer result, the landing page leads with that result and expands it. If card three was about integration speed, the landing page speaks to engineers about integration. The narrative should feel continuous, not like a topic change.

With per-card URLs, multiple landing pages become viable and worth building. The CFO card leads to an ROI-focused page. The technical card leads to an integration-focused page. This is personalization at the destination level, and it's one of the few places in digital advertising where you can meaningfully match the message to the moment across multiple audience types in a single campaign.

On form length: Lead Gen Forms that pre-fill with LinkedIn profile data reduce friction and typically outperform external landing page forms for cold-to-warm audiences. Fewer fields to complete means fewer reasons to abandon.

One continuity check that teams often miss. The CTA on the landing page must match the CTA the carousel sequence implied. A sequence that builds toward a demo request, then lands on a content download, creates friction and loses the conversion. The visitor arrived expecting one thing and got another. That mismatch is a conversion killer.

The carousel and the landing page are one system. Test them together, not independently. A change to card one may require a change to the landing page headline to maintain narrative continuity. They're not separate deliverables. They're two parts of the same argument.

Testing, iteration, and what to measure to improve the next campaign

Let's talk about how you get smarter over time, because a single campaign is almost never the point.

Test one variable at a time. Card order, card one image, headline copy, CTA placement. Mixing variables makes it impossible to know what moved the number. This sounds like basic advice because it is. It's also constantly ignored under schedule pressure.

The priority test, if you're picking one place to start, is CTA placement. Last card versus an earlier card. This interacts directly with how warm the audience is, and the answer will likely surprise you at least once.

A note on benchmarks. The platform average carousel CTR is around 0.4%. That's a floor benchmark, not a target. If a campaign is sitting there, card one creative or audience match is the first place to investigate. The first card and the audience definition are doing most of the work.

Here's where I want to push back on how most teams measure these campaigns, because I think it's where the most value gets left on the table.

CTR and engagement rate are diagnostic. They tell you whether the format is working mechanically. They are not success criteria. Cost per opportunity and influenced pipeline value are the scorecards. Revenue leadership doesn't care about swipe rates. They care about deals.

But there's a structural measurement problem worth naming directly. Seventy-three percent of B2B organizations use 30-day attribution windows, regardless of how long their actual sales cycle is. For deals that span 12 to 20 weeks, carousel touchpoints from weeks two through eight receive zero credit in a 30-day model. The channel looks underperforming when it isn't. The measurement model is producing a false signal.

Only 21% of B2B marketers say they can measure marketing ROI with confidence, according to the Demand Gen Report 2025. That means most carousel campaigns are being judged on incomplete data.

Organizations that implement multi-touch attribution report meaningful improvements in marketing ROI and reductions in customer acquisition cost, per research from Forrester and McKinsey. The measurement model is itself a performance lever. Fixing how you measure is as valuable as optimizing the creative.

And here's the compounding logic that makes all of this worth doing consistently. Per-card click data from campaign one informs card order for campaign two. Landing page conversion rates tell you which arguments to lead with. Audience segment performance tells you where to concentrate budget. Each campaign should be smarter than the last, not because you're working harder, but because you're building on what the previous campaign actually taught you.

That's the system. Not a checklist. A loop.