You open Campaign Manager, you get to the targeting step, and you do the obvious thing. You type in your buyer's job title. "VP of Marketing." "CFO." "Head of Operations." It feels precise, like you're walking straight up to exactly the person you want.
It's the most expensive way to target on LinkedIn, and it quietly hides half of the people you're trying to reach.
That's not an opinion about creative or budget. It's how the platform's targeting actually works under the hood, and almost every business owner gets it wrong the same way. So let's fix the part of LinkedIn that decides more about your cost per lead than your bid ever will: who you point the ad at.
On LinkedIn, Targeting Is the Lever. Not Your Bid.
Every ad platform sells you something a little different. On Google, you're bidding on what someone typed into a search box. On Meta, you hand the algorithm your creative and it goes hunting for the people most likely to respond. LinkedIn sells the one thing neither of those can: you get to choose the exact professional in the room by who they are. Their job. Their level. The company they work at.
That precision is the whole reason LinkedIn costs what it costs. A click here can run many times what it would on Meta. You are paying a premium for the ability to be selective, which means the selection itself is where your money is won or lost.
Here's why it matters more on LinkedIn than anywhere else. B2B buying is slow. Someone can see your ad this month and not be ready to do anything about it for another six. The job of the campaign is to be in front of the right people, over and over, until the day they're finally ready. Now do the math on that. If you're going to pay LinkedIn's premium to be seen repeatedly by an audience, the single most expensive mistake you can make is filling that audience with the wrong people. No bid adjustment, no clever creative, no landing page tweak fixes an audience aimed at the wrong crowd. Targeting is upstream of everything.
Every LinkedIn Audience Answers Two Questions
Before you touch a single targeting option, understand that every good audience is really two audiences stacked together. Who is this person? And what kind of company do they work at?
Most owners obsess over the first question and skip the second, and the second is often what saves the budget. Company size is a spending filter. If you sell a tool that costs 1,200 dollars a month, the CFO of a two-person company is a perfect job-title match and a completely useless lead, because that company will never have the budget. Layering in company size, or industry, or a list of named accounts you actually want, is how you stop paying to reach people who fit the title but could never say yes.
A quick example of how much this saves you. Say you sell a scheduling platform priced for teams, and your buyer is an operations lead. Target "operations" by itself and you'll pour budget into solo operators and ten-person shops who will never have a team to schedule. Add a company-size floor of, say, 50 employees and up, and the same targeting suddenly points only at companies built the way your product needs them to be. You didn't change who the person is. You changed which version of that person can actually write the check.
So set the company frame first. Decide what kind of organization can actually buy from you, then move to the harder question: how do you find the person inside it?
The Four Ways to Find the Person, and What Each One Costs You
LinkedIn gives you several ways to target an individual. Four of them do the real work: job title, job function plus seniority, skills, and groups. They are not interchangeable. Each one buys you a different mix of cost, reach, precision, and lead quality, and knowing the difference is most of the game.
Job title is the trap. Two things make it the weakest option, not the strongest. First, it's a free-form field. People type whatever they want into their own headline, so LinkedIn's system can only confidently match a fraction of the job titles that actually exist. AJ Wilcox, who runs one of the only agencies in the world that does nothing but LinkedIn ads, estimates the platform truly understands only about 40 percent of titles. That means when you target by title, you may be reaching less than half of your real audience and never knowing it. Second, job title is where every marketer instinctively starts, and crowded targeting is expensive targeting. So you get the worst of both: a smaller audience and a higher price.
The size of that gap is easy to miss until you see it. In one live example, targeting "marketing VP and above" by job title produced an audience of roughly 120,000 people in North America. The same audience built by job function and seniority came out around 250,000. That's about 70,000 qualifying buyers the job title simply could not see.
Job function plus seniority is the durable default. Job function is the department someone sits in. Seniority is their level, from individual contributor up through manager, director, VP, and the C-suite. Put them together and you describe a person by what they actually do and how senior they are, rather than by a title string they invented. It's broader, it's cheaper because it's less crowded, and crucially it catches all those people whose odd or non-standard titles the platform couldn't parse. This is the option most owners underuse, and it's usually the one that should carry the account.
Skills are broad but noisy. Every LinkedIn member can list dozens of skills, and people list skills they'd never buy anything for. Someone who took one class on a piece of software will list it right next to the things they do all day. So skill targeting reaches a lot of people, cheaply, but with more waste mixed in. The fix is to get specific: target a precise two-word skill, not a single broad word. "Marketing" as a skill catches students and salespeople. "Demand generation" or "marketing operations" catches the actual buyer.
Groups are small, precise, and high-intent. When someone joins a LinkedIn group about their profession, they've raised their hand and said this is genuinely my world. A salesperson might list "marketing" as a skill, but they're not going to go join a working group full of marketers. Group audiences are the smallest of the four, but they tend to be the most active people on the platform and the highest quality leads. Layer a seniority level on top and you've got a small, sharp audience worth its size.
The Move That Separates the Pros
Here's the part that changes how you should think about all of this. You don't have to pick the one right facet. The people who run LinkedIn well don't.
Instead of agonizing over whether job function beats skills for your business, run them side by side. Build the same persona four times, once by job title, once by function and seniority, once by skills, once by groups, as separate campaigns. Yes, they'll overlap. That's fine. What you get in exchange is data. Over a few weeks you'll see that maybe skills bring the cheapest clicks but the softest leads, while job title costs more but converts to actual sales conversations at a higher rate. Now you know something real about your market, and you can bid up the winners and quietly shut off the losers.
And notice what that protects you from. If you'd bet the whole account on job title from day one, you'd never have learned that groups were bringing you leads at half the cost, because you'd never have run them. Structuring targeting as a set of parallel bets instead of a single guess is what turns a LinkedIn account from a monthly gamble into a machine that gets cheaper the longer it runs. The account teaches you where your buyers actually live, one campaign at a time.
You can push this further and turn it into free market research. Instead of one audience of "managers and above," split it into two: managers and directors in one campaign, VPs and the C-suite in another. Same content to both. Whichever one responds tells you which level your message actually lands with. That's the kind of insight most companies pay a research firm for, and here it's just a byproduct of structuring your targeting like a test instead of a guess.
The Two Settings That Quietly Wreck Good Targeting
You can do all of this correctly and still hand it all back with two defaults.
The first is audience size. LinkedIn is a small, concentrated network compared to the consumer platforms, so the sizing math is different. A cold audience in the tens of thousands is a healthy target. Too small and it can't generate enough data to learn from; too big and you've lost the precision you paid the premium for in the first place. We wrote about exactly where that line sits in Your LinkedIn Audience Shouldn't Be 300,000. It Also Shouldn't Be 300.
The second is the one that undoes more careful targeting than anything else: audience expansion. LinkedIn turns it on by default, and it quietly widens your carefully chosen audience to include people the system thinks are "similar." You did the work to reach a specific kind of buyer, and a checkbox you never noticed opened the doors to people you deliberately left out. Turn it off. Turn off the LinkedIn Audience Network too, which shows your ads on third-party apps outside LinkedIn. Both are part of the same lesson we covered in LinkedIn Ads Aren't Expensive. Your Setup Is: the platform's defaults are not built for your cost per lead.
What This Means for You
You don't need to build these campaigns yourself. But if someone is running LinkedIn ads for you, you now know the three questions that separate a good setup from a leaky one. Are we targeting by job title alone, or by job function and seniority? Are we running the different targeting types as separate campaigns so we can actually see which one produces qualified leads? And is audience expansion turned off? If the answers are vague, that's where your budget is going.
LinkedIn's entire value is precision. It's an expensive room, and it's only worth the price of admission if you spend it on exactly the right people in it. That's a targeting decision, and it's the one you have the most control over.
If you'd rather have someone build and test that targeting properly instead of guessing at it, that's the kind of thing we do all day. And if you'd rather run it yourself, you're now ahead of most of the people paying to reach half their market by the wrong name.
