You will compare the cost per click with Google’s and conclude that LinkedIn is expensive.
It is. And that comparison still tells you nothing, because the two numbers are prices for different goods.
A search click is somebody who typed the thing. A LinkedIn click is somebody who was reading about a colleague’s promotion and looked up. Paying several times more for the second is not automatically wrong. Paying several times more for the second to do the job the first does is always wrong, and that is what most small B2B accounts are quietly doing.
So the question is not whether LinkedIn is expensive. It is whether you are buying the one thing it sells that nothing else does.
The one thing it sells that nothing else does, and the test that follows
LinkedIn’s unique asset is not the feed. It is the list.
No other advertising platform lets you upload three hundred company names and put a message in front of the people inside them, by job function, reliably. That is the product. Everything else in the account, broad interest targeting, lookalikes, awareness campaigns, is a worse version of something available cheaper elsewhere.
Which gives you a test you can apply this afternoon, before you spend anything.
Can you write the list?
Not a description of it. An actual file: two hundred named companies, or a job title so specific that the people holding it are identifiable and are not yet searching for your category. If you can produce that, LinkedIn is doing something no other channel can, and the premium is a reasonable price for it.
If you cannot, what you have is a demographic guess, and you are about to buy ordinary reach at premium rates. The honest answer there is to go and do the work that produces the list, because it is the same work that makes every other channel cheaper too.
LinkedIn’s own numbers contradict the reason you want to use it
Here is the tension nobody puts in the pitch, and it comes from LinkedIn’s own documentation rather than from me.
The minimum audience needed to run a campaign is 300 members. On the same page, LinkedIn recommends aiming for 50,000 or more for single image and text ads, and more than 15,000 for message ads, to get what it calls optimal scale.
Sit with that gap for a second. The platform will let you run at 300 and tells you to aim for 50,000. Which means the precise, named-account targeting that is LinkedIn’s entire reason to exist is not what its delivery is built around.
It gets more pointed. Location is a required targeting facet and it shrinks your audience, so a matched segment should comfortably exceed 300 before you apply geography. And when a segment comes in too small, LinkedIn’s own troubleshooting advice is to combine several segments or build a lookalike until you clear the floor. Read plainly, that is the platform telling you to make your targeting less precise so that it can deliver.
Now run your own arithmetic. Two hundred target companies with three relevant people in each is six hundred people. You clear the 300 floor. You are also roughly two orders of magnitude below the scale LinkedIn says its main formats want.
That is the real small-company problem with paid LinkedIn, and it is not the click price. The product you want to buy and the product that performs are different sizes.
About the cost figures
Worth one short note, because you will be shown a number.
LinkedIn does not publish an average cost per click or click-through rate, so every figure in circulation is a third-party estimate. The most carefully sourced I found is Metadata’s 2026 B2B paid media benchmark, built from the 2025 spend of 138 advertisers, which puts LinkedIn CPC at $9.39 with a 0.67% click-through rate and $202 per lead. Vendor guides elsewhere go as low as $5.26 and as high as $10.44, and a second platform-based report comes in lower than all of them.
Treat the Metadata figure as the best available and still as a sample of companies that run paid media through a paid media platform, which is not a random draw of small businesses.
And notice that the spread does not change the decision in front of you. Whether a click is six dollars or ten is a budgeting detail. Whether a click is the thing you are buying is the actual question, and no benchmark answers it.
The pair of numbers worth keeping is the 0.67% click-through rate against the $202 per lead, because together they tell you the funnel compresses hard and the per-lead figure is doing a lot of work.
A lead and a conversation are further apart here
Lead gen forms do what they claim. They prefill from the profile, friction drops, and your form-fill rate goes up. That is real and I would use them.
The cost is systematic rather than occasional. The person never visited your site, never read your positioning, and tapped a button that took no effort. So the gap between cost per lead and cost per conversation is wider on LinkedIn than on any channel I have worked in, and the dashboards show you only the first of the two.
Which means the number to hold yourself to is cost per conversation with somebody who knew what they were agreeing to. Count it by hand for a month if you have to. If a $202 lead becomes a conversation one time in four, your real figure is $800, and that is still fine for a $40,000 contract and absurd for a $4,000 one. The arithmetic is category-specific and nobody can hand it to you.
For what to actually run on a small budget: one single-image ad making one specific claim, pointed at a named-account list. Not video, not carousels, not thought leadership formats, because those are brand spending at a budget that cannot do brand, which I went through when writing about awareness as a goal.
Where I come out
For most companies under a few million in revenue, the right paid LinkedIn budget is zero. I want to be precise about why, because that sounds like a dodge and it is not.
It is not that the targeting does not work. It is that at the list size where the targeting is worth paying for, six hundred people, you are in a range where an auction is the wrong instrument. Six hundred people can be reached by one person, deliberately, with research in each message. That person costs less than the campaign and gets replies the ad never would.
The crossover is roughly where the list gets too big to work by hand and still small enough to be real. Several thousand named people, a sales team that cannot cover them, and a deal size that survives an eight hundred dollar cost per conversation. At that point LinkedIn is the correct tool and the premium is worth paying.
Below it, put the money into the list itself, and put the effort into the founder posting and into six hundred messages that somebody actually wrote. I covered the posting half of that for B2B founders and the argument transfers across markets unchanged.
If you’d like a second opinion on whether your list is big enough to advertise to or small enough to work by hand, drop me a line on email, WhatsApp, phone or LinkedIn and we can have a quick chat, and my notes on performance marketing cover the rest. I’m taking on a small number of contracts at the moment, so I’ll tell you straight whether it’s something I could help with.