Welcome back — jump to Tutorials, Support & Docs, or What’s New. Go to PlatformNot you?Log out
Welcome back — jump to Tutorials, Support & Docs, or What’s New. Go to PlatformNot you?Log out
Welcome back — jump to Tutorials, Support & Docs, or What’s New. Go to PlatformNot you?Log out
Platform AI Agents Customers Developer Docs Services Company Careers Blog 2026 Benchmark Report Resources Comparisons Buyer's Guide Integrations Use Cases Tutorials Events Sign In Book a Demo

Channel cost reference · 2025 data

LinkedIn ads cost for B2B: CPL, CPC, CPM and the full distribution

Every number below comes from 138 B2B advertisers running LinkedIn in 2025, inside a dataset of $57.6M of measured spend with closed-won revenue traced back to the ads. No email gate. The method is published, the dataset is available, and every figure carries the number of advertisers behind it.

How much do LinkedIn ads cost for B2B?

A B2B lead from LinkedIn costs $202 on a spend-weighted basis, at $9.39 per click and $63.19 per thousand impressions. Click-through rate is 0.67% and 5.9% of clicks convert. Across 138 advertisers, 2025.

LinkedIn B2B media metrics, 2025, all campaign goals, n=138
MetricValue
Cost per lead$202
Cost per click$9.39
Cost per thousand impressions$63.19
Click-through rate0.67%
Click to lead conversion5.9%

What is a good LinkedIn cost per lead?

The weighted $202 sits at the 26th percentile. The median advertiser pays $376, nearly double. If you are under $196 you are in the top quarter. Above $658 you are in the worst quarter.

This is the number every other benchmark hides. A single weighted average makes most advertisers think they are failing against it, because a handful of very efficient large spenders drag it down. The spread is the useful part:

LinkedIn cost per lead, distribution across n=88 advertisers, 2025
PercentileCost per leadWhat it means
p10$109Top 10% of advertisers
p25$196Top quarter
Median$376The typical advertiser
p75$658Bottom quarter
p90$1,340Bottom 10%
Spend-weighted$20226th percentile, not typical

Which LinkedIn ad format produces the cheapest leads?

Document ads, at $142 per lead against $200 for image and $265 for video. Document ads also convert at 11.9%, roughly double the channel average. Conversation ads are the most expensive at $362.

LinkedIn cost per lead by ad format, 2025 lead-generation campaigns
FormatCost per lead
Document$142
Image$200
Video$265
Conversation$362

Do LinkedIn lead-gen forms beat landing pages?

Yes, and by more than most teams expect. A LinkedIn lead-gen form produces a lead for $193. Sending the same click to a landing page costs $346, 79% more. Measured across 97 form advertisers and 65 landing-page advertisers.

Is LinkedIn retargeting cheaper than prospecting?

No. On LinkedIn, retargeting costs $234 per lead against $194 for prospecting. The warm audience is the dearer one, which is the opposite of the pattern on Meta. Cold LinkedIn audiences run $194 at 6.6% conversion across 94 advertisers.

What does a LinkedIn customer actually cost?

Cost per lead is not cost per customer. Across 114 advertisers with closed-won revenue traced to their LinkedIn ads, 21.4% of leads closed and the acquisition cost was $63,312. Triggered return on ad spend was 0.57x.

For companies under 200 employees the picture is different again: $35,288 per customer at a 19.5% close rate, across 27 advertisers. Company size moves this number more than any creative decision does.

Where does this sit against other channels?

LinkedIn is the most expensive per lead of the social channels and the cheapest per lead against Google Ads. The full cross-channel comparison, with CTR, CPC and CPM for every channel, is on the B2B advertising benchmarks page. The 2026 benchmark report lets you filter all 111 cuts by company size, industry and campaign goal.

How were these numbers calculated?

Every figure is a spend-weighted aggregate across advertisers who ran LinkedIn in 2025. Suppression thresholds mean no cut is published on a sample too small to be worth anything; cuts that failed were killed, not caveated. The full method names the rules and the dataset is CC BY 4.0.