LinkedIn Video Ad Length and Format Benchmarks for B2B

LinkedIn video ad benchmarks look like specs. They're not. They're a record of what B2B buyers actually watch, when they bail, and which format choices turn into something a sales team can use. LinkedIn's video views climbed 36% year-over-year, and its share of total B2B ad budgets grew from 31% in H1 2024 to 39% by year-end, according to Dreamdata's 2025 LinkedIn Ads Benchmarks. Separately, EMARKETER and Demandbase found 65.4% of marketers use LinkedIn most often for B2B marketing. Video is where the money and attention are going.

Here's the part that should stop you before you pour more budget in. A practitioner analysis of 342 video ads across seven B2B SaaS companies (ZenABM) found video takes 31.72% of LinkedIn ad budget but posts the lowest CTR of any format (0.24% median), the highest CPC ($15.61), and the worst efficiency score (1.5 out of 10). So video is both the format everyone's betting on and the format that looks worst on a spreadsheet. Which one is true?

Both. That's the tension worth naming up front: video's strategic role in a B2B buying journey and its performance as a direct-response ad are two different jobs. Confusing them is exactly how teams end up burning budget on a format that was never supposed to win on clicks. Reading the benchmarks correctly means knowing which metric belongs to which job. The rest of this piece is about sorting that out.

What the platform's own length recommendations reveal about buyer behavior

LinkedIn supports a wide range of video lengths, but platform guidance consistently points to 15 to 30 seconds as the sweet spot for most ad placements. That's not a random cutoff. It's LinkedIn telling you, based on its own data, what people will actually sit through.

The broader consensus across 2025 and 2026 sources sorts by objective:

• 15 to 30 seconds for brand awareness and cold audiences
• 30 to 60 seconds for consideration and education
• 30 seconds to 2 minutes for conversion-focused campaigns and retargeting

The logic underneath this is simple. A cold audience didn't ask to learn about your product. They scrolled into it. Asking for 90 seconds of attention from someone with zero context is a mismatch between what you're asking and what they've agreed to give.

But here's where it gets interesting, and where "shorter is always better" starts to fall apart. Huble's data on conversion-focused campaigns shows those videos averaged 37 seconds, with most running under 30 seconds, and completion rates were significantly higher in those campaigns than elsewhere. So length wasn't the variable that mattered. Audience was.

Think about who clicks into a conversion campaign in the first place. That person is already warmer. They've likely seen your brand before, or they're searching for a solution, or they clicked because the offer spoke to a problem they already know they have. Length tolerance isn't set by the ad's objective label. It's set by where the viewer already stands in their own buying process. The campaign objective is just a proxy for that.

One more data point worth sitting with: view rates are up roughly 5% year-over-year, per Huble. That's a sign video is becoming a normal thing people expect to see in their feed, not a novelty. But normalization is not the same as patience. People are used to seeing video. They still won't wait for it to get good.

How format and aspect ratio choices map to where your buyers actually are

Aspect ratio isn't a design detail. It's a distribution decision, and treating it as an afterthought is one of the more expensive mistakes a video ad can make.

Here's the breakdown:

• 9:16 vertical reaches mobile only. LinkedIn is pushing this hard for mobile optimization.
• 1:1 square works on both mobile and desktop. It's the safe, broad-reach default.
• 16:9 landscape still functions, but it's a legacy format. The platform isn't favoring it anymore.

By 2026, vertical video (9:16 or 4:5) is the strongly recommended path, per aimers.io, largely because it feels native to the feed rather than like a commercial that got dropped in. Vertical video doesn't announce itself as an ad. That alone changes how people engage with it.

The trap here is obvious once you see it: teams take a horizontal brand video built for a website or a YouTube pre-roll and drop it into LinkedIn unchanged. That single decision cuts off most of your mobile reach and kills the native feel at the same time. You lose twice for the price of doing nothing.

Then there's sound. Roughly 80% of LinkedIn videos are watched with the sound off. Read that again, because it changes what "creative" means for this format. Captions and on-screen text aren't a nice touch, they're the difference between an ad that communicates and one that's just moving pixels. If your video only works with sound on, it doesn't work on LinkedIn.

One more practical constraint: LinkedIn caps video ads at 200 MB, well under the limit for organic video. That means compression has to be a pre-production decision, not something you scramble to fix the night before launch.

Put it together and the point is this: format decisions made during production determine who can even see your ad and whether the message lands once they do. That's not separate from targeting strategy. It's part of it.

What the performance benchmarks mean when you read them against each other

Numbers in isolation don't tell you much. Numbers read against each other start to tell a story.

Start with in-feed video view-through rate: 20 to 35%, according to yansmedia's April 2026 data. That means roughly one in three impressions results in at least 2 seconds of viewing. This is a volume metric. It tells you the hook worked well enough to stop the scroll. It says nothing about quality.

Now CTR. Bull & Wolf puts it around 0.44%. ZenABM and yansmedia put it lower, around 0.24 to 0.25%. That spread is worth pausing on. A gap that size between sources probably isn't about creative quality. It's more likely a signal of how tightly targeted the audiences were. Loose targeting drags CTR down regardless of how good the video is.

Cost figures: average CPC globally sits around $5.58, average CPM around $33.80 (Bull & Wolf). Useful as a floor. But B2B SaaS campaigns targeting small audiences and senior titles routinely run well above these numbers, because the audience pool is small and everyone's competing for the same inboxes.

Engagement rate benchmark lands at 1 to 2% (Bull & Wolf).

So how do you use any of this? Here's a diagnostic worth keeping in your back pocket. If a campaign has a strong view-through rate but a weak CTR, the video is holding attention but not creating intent. That's a message or offer problem, not a targeting problem. If view-through rate itself is weak, the issue is upstream, in the hook or the audience relevance, before the CTA ever gets a chance.

Seasonality adds another wrinkle. HockeyStack's dataset, drawn from over 70 B2B SaaS companies ranging from $5M to $1B in revenue, found September stood out sharply, with up to 117,095 clicks and over 10.9 million impressions. If you're comparing your Q3 numbers to a benchmark built on a different quarter, you're not comparing like to like.

But here's what none of these benchmarks do: connect to pipeline. A team optimizing for CTR on a video ad is optimizing for the wrong output. CTR tells you about the ad. It doesn't tell you about the business. The benchmark worth building toward lives further downstream, in qualified opportunity rate, and none of the numbers above get you there on their own.

The first three seconds and why most B2B video ads lose before the message lands

Viewers decide whether to keep watching in about 1 to 2 seconds. Not ten. Not five. One or two. That means the hook isn't a stylistic flourish you add for polish. It's the primary engineering problem of the entire ad.

Most B2B video ads fail here, and they fail in a predictable way. They open with a company logo. Or a branded animation. Or some version of a mission statement. All three defer the one question every viewer is silently asking: what's in this for me? By the time the ad gets around to answering that, the viewer has already scrolled past.

What actually works as an opener:

• A bold, specific claim that names the viewer's exact problem
• A scenario the target buyer recognizes instantly, because they've lived it
• A direct answer to a question the buyer is already asking themselves

LinkedIn Creative Labs ran a study across thousands of B2B video ads, analyzing hundreds of thousands of video frames with LLMs and machine learning. The finding: creative approaches built around "Expert Takes" and "Human Touch" drove large lifts in performance. Not production polish. Framing and message structure.

That should reframe how a lot of organizations think about video budget. A 60-second explainer that speaks directly to a specific business problem will outperform a beautifully produced brand video that says nothing in particular. That's a hard pill for any organization that gates video behind a big-budget production pipeline, because it means the bottleneck was never the camera. It was the script.

There's a corollary worth knowing if you're running message ad formats. Video there requires a click to even start playing, which means the thumbnail is doing the exact same job as the hook in a feed ad. A generic branded still gets scrolled past without a second thought. A face, a text overlay, or a clear visual problem statement gets the click that starts the clock.

How funnel stage should govern both length and offer — and where Lead Gen Forms actually fit

Video isn't one format playing one role across a campaign. It plays a different role at every stage, and length and offer both need to shift with it.

• Top of funnel: short, under 30 seconds, problem-aware, no product pitch. The job here is recognition and reach, nothing more.
• Middle of funnel: 30 to 60 seconds, built around education and trust. Case studies, category framing, "here's why this actually matters" content.
• Bottom of funnel and retargeting: up to 2 minutes, built around proof and specificity. Testimonials, demos, ROI evidence, for an audience that already knows who you are.

A common starting allocation across the funnel runs roughly 40% top, 35% middle, 25% bottom, according to ppcblogpro. As your retargeting pool grows over time, shifting more budget toward middle and bottom of funnel tends to pay off, since that's where ROI is typically higher.

Now, Lead Gen Forms. These consistently convert 2 to 4 times higher than sending the same traffic to a landing page, mostly because LinkedIn pre-fills contact information and removes the friction of a form. Sounds like a clean win.

Except that frictionlessness cuts both ways. Easy submission means low-effort submission, and low effort often means low intent. Someone who fills out a pre-filled form in three seconds hasn't necessarily thought about whether they're a fit. Adding one qualifying question, something like company size, current challenge, or role, filters for actual ICP match instead of just volume.

A real example makes this concrete. A B2B cybersecurity company spending $60,000 a month, with 85% of that going to LinkedIn Lead Gen Forms, was generating high lead volume but a very low share of qualified opportunities. When they restructured into a three-layer funnel approach, lead volume dropped sharply, but lead quality improved substantially, per Stackmatix.

That's not a story about video length or aspect ratio. It's a story about what happens after someone watches. Video ad decisions can't be judged in isolation from the offer, the form, and the qualification logic sitting behind them. A great video feeding a bad form still produces bad pipeline.

What good measurement looks like when video's job is pipeline, not clicks

Most LinkedIn video dashboards show you view-through rate, CTR, and CPM. Those describe how the ad performed on the platform. They don't describe what it did for the business. That gap is where a lot of budget decisions go wrong.

What actually connects video to pipeline:

• View-to-opportunity rate. What share of viewers, broken down by audience segment, actually entered the pipeline within a defined window.
• Influenced pipeline. Deals where a LinkedIn video touchpoint shows up somewhere in the multi-touch path. Research on multi-touch attribution consistently finds that LinkedIn-influenced pipeline from multi-stage funnels outperforms single-stage campaigns on close rates.
• Cost per qualified opportunity, not cost per lead. The cybersecurity example above is the clearest possible illustration of why that distinction matters. Cost per lead looked fine right up until someone checked the opportunity rate.

Attribution infrastructure is the quiet variable underneath all of this. Dreamdata's 2025 benchmarks cite LinkedIn's own finding that Conversions API users see roughly 20% lower CPA compared to those without the integration. Without that plumbing in place, video's contribution to pipeline is effectively invisible. You can't credit what you can't see.

Here's the practical diagnostic to run when a video campaign is producing views and even clicks but nothing shows up downstream. The problem is more likely sitting in the offer, the landing page, or the lead qualification logic than in the video itself. Benchmark data alone won't tell you which one it is. That answer only comes from looking across the whole system, not just the ad.

Benchmark literacy gets you far. It tells you where your ad stands relative to the rest of the market. But it stops short of telling you what to actually do with that information. Turning a view-through rate or a CTR into a decision about pipeline requires looking at the full path a viewer takes, from the first frame of video to the moment a deal gets marked qualified. That's a different kind of work than reading a chart. It's the work that makes the chart worth reading in the first place.