How to stop wasting money on
LinkedIn ads that target
low-intent audiences.

LinkedIn Ads are expensive. Bad targeting makes them ruinous.

LinkedIn advertising already costs more per click than any other major platform. Average CPCs run $5 to $12. CPMs sit between $30 and $55. That's the baseline cost of doing business on the platform. There's no way around it.

The real damage doesn't come from LinkedIn's pricing, though. It comes from how most advertisers set up their audiences. They go too broad. They target by job title alone, by industry alone, or they let LinkedIn's default settings do the work. The result is an audience that looks big on paper and converts at near-zero in practice.

The 95-5 rule applies directly. At any given moment, only about 3 to 6% of your total addressable market is actively in a buying cycle. The other 94 to 97% aren't looking, aren't comparing, aren't going to act on your ad no matter how sharp the creative is. When you target a broad audience, you're paying LinkedIn prices to reach a group that is overwhelmingly out of market.

"Only 5% of B2B buyers are in-market at any given time. The other 95% won't buy no matter what you put in front of them." (Ehrenberg-Bass Institute, How Brands Grow)

Broad targeting doesn't just waste money.

It corrupts your data too.

Run ads to a 300,000-person audience where only 5% are remotely in market, and every metric you collect is distorted. Your click-through rate reflects how interesting the ad looks to people who have no reason to click. Your conversion rate measures how well your landing page performs against traffic that never intended to convert. Your cost per lead is inflated by thousands of impressions served to people who were never going to become customers.

Worse, it makes optimisation impossible. LinkedIn's algorithm learns from the data your campaigns generate. If most of your clicks come from low-intent people, the algorithm goes hunting for more of them. You end up in a feedback loop where you're paying more and more to reach people who are less and less likely to buy.

This is why so many B2B teams conclude that LinkedIn Ads don't work. The platform works. The targeting was wrong from day one, and every data point collected afterwards reinforced the wrong direction.

"Bad targeting doesn't just waste budget. It trains the algorithm to find more of the wrong people." (Metadata.io, State of Demand Generation 2025)

How LinkedIn's auction makes bad targeting worse.

LinkedIn runs a second-price auction. You don't pay your maximum bid, you pay just above the next-highest bidder. That sounds efficient. For broad audiences, it creates a specific problem.

When your audience is large and undifferentiated, you're bidding against every other advertiser also targeting that same general pool. VP of Marketing. Director of IT. Head of Sales. These are the most competed-for segments on the platform. Everybody bids on them. The competition drives your cost up without improving the quality of who sees your ad.

Narrow audiences face less competition. When you layer targeting criteria, job title plus company size plus specific industries plus seniority, you carve out a segment that fewer advertisers are bidding on. The audience is smaller, but cost per impression often drops because you're not fighting every other B2B advertiser for the same eyeballs.

The math shifts even further once you factor in relevance scores. LinkedIn rewards ads that get high engagement relative to their audience. Tightly targeted ads tend to land harder, which lifts your relevance score, which lowers your CPC. Broad ads get ignored by most of the audience, tank the relevance score, and drag your costs up further.

"LinkedIn's ad relevancy score directly impacts auction outcomes. Higher relevance means lower CPCs, even in competitive segments." (LinkedIn Campaign Manager Documentation, 2026)

Fix 1: shrink the audience and layer the targeting.

Start with your ICP, not with LinkedIn's suggestions. Most advertisers build audiences by typing a job title into Campaign Manager and accepting whatever LinkedIn suggests. That's backwards. Start with your actual customer data. Who has bought from you? What were their titles, company sizes, industries, and seniority levels?

Layer at least three criteria. Job title on its own is not targeting. It's a starting point. Add company size. Add industry. Add seniority. Add geography if the sales team only covers certain regions. Every layer removes people who look right on one dimension and wrong on every other.

Aim for 20,000 to 80,000 in your audience. LinkedIn will tell you the audience is too small. Ignore that. For most B2B campaigns, 20,000 to 80,000 is the sweet spot. Large enough for the algorithm to optimise, small enough that a meaningful share of the audience is actually relevant.

Exclude aggressively. Strip out job titles that sound right but aren't. Strip out company sizes below your minimum deal threshold. Strip out industries where you have no case studies or references. Every exclusion sharpens the audience and cuts waste.

"The ideal LinkedIn Ads audience size for B2B is 20,000 to 80,000. Larger audiences dilute targeting precision and inflate CPL." (Directive Consulting, LinkedIn Ads Benchmarks 2026)

Fix 2: use buying signals to decide when campaigns run.

Targeting the right people isn't enough. You also need to reach them at the right moment. A CFO who just signed a three-year contract with your competitor is the right person at the wrong moment. A VP of Marketing who just posted about needing to scale demand generation is both.

Watch for intent signals outside LinkedIn. Tools like Bombora, G2, or 6sense surface companies actively researching your category. When a company's research activity spikes, that's when you switch on LinkedIn campaigns targeting their decision-makers. Not before.

Build campaigns around trigger events. New funding rounds. Leadership changes. Product launches. Office expansions. Company moves. These events correlate with buying windows. When a company raises a Series B, they're about to spend on infrastructure, tools, and services. That's when your ad should land in front of their team.

Pause when signals go cold. If the intent data shows activity dropping off, pause the campaign. Don't keep spending on an audience that walked out of the buying window. You can always reactivate when fresh signals appear. This single discipline can cut waste 30 to 40%.

"Intent data-driven campaigns produce 2 to 3x higher conversion rates than demographic targeting alone." (Bombora, B2B Intent Data Benchmark Report 2025)

Fix 3: turn off Audience Expansion and Audience Network.

Audience Expansion is on by default. When you create a campaign in LinkedIn Campaign Manager, Audience Expansion is toggled on. The setting tells LinkedIn to show your ads to people outside your defined audience who it thinks are similar. In practice, it undoes your targeting work.

LinkedIn's definition of "similar" is generous. If you target Directors of Marketing at SaaS companies, Audience Expansion might show your ad to Marketing Coordinators at non-tech firms. They're in marketing. They're on LinkedIn. By LinkedIn's logic, similar enough. By your sales team's standards, unqualified.

Audience Network is the same problem in a different wrapper. The setting places your ads on third-party apps and websites outside LinkedIn. The click costs might look cheaper, but the traffic quality drops sharply. You lose the professional context that makes LinkedIn ads worth running in the first place.

Turn both off immediately. Open every active campaign and disable Audience Expansion and Audience Network. Check new campaigns before launch, too. These settings reset to on with every new campaign. Leaving them on is the single fastest way to undo precise targeting.

"Audience Expansion can increase reach by 25% but often dilutes lead quality by including profiles outside your ICP." (B2Linked, LinkedIn Ads Optimization Guide 2026)

Fix 4: run three audience tiers instead of one catch-all.

Tier 1, hot: retargeting and matched audiences. People who already know you. Website visitors, video viewers, Lead Gen Form openers, event attendees, CRM contacts. They've already engaged. Cost per conversion here is the lowest you'll get on LinkedIn, because they don't need convincing. They need a nudge.

Tier 2, warm: intent-signalled accounts. Companies showing active buying behaviour, either through intent platforms or through direct engagement with your content. They haven't converted yet, but the signals say they're in or near a buying window. Target their decision-makers with mid-funnel offers: case studies, benchmarks, product comparisons.

Tier 3, cold but precise: ICP-matched new audiences. People who match your ideal customer profile but haven't interacted with you yet. This is where the layered targeting matters most. Don't run awareness ads to everyone who might one day buy. Run them only to the tightest possible definition of your ideal buyer, and accept that the audience will be small.

Allocate budget accordingly. Most teams put 70 to 80% of budget into Tier 3 because it feels like growth. Flip that. Put 40 to 50% into Tier 1, 30% into Tier 2, and 20 to 30% into Tier 3. The lowest-cost conversions come from people who already engaged. Feed that engine first.

"Retargeting audiences on LinkedIn convert at 3 to 5x the rate of cold audiences, at roughly half the cost per lead." (Metadata.io, Paid Social Benchmark Report 2025)

Where this lands.

LinkedIn Ads aren't inherently wasteful. The default setup, though, encourages waste at every step. Broad audience, Audience Expansion on, Audience Network on, no intent signals informing the timing, and all budget pointed at cold outreach. That combination burns money reliably.

The fix isn't complicated. Shrink the audience. Layer the targeting. Use intent data to time your campaigns. Turn off the settings that dilute your precision. Structure the budget around audience temperature, not audience size.

Do those four things and your cost per qualified lead drops. Not because LinkedIn got cheaper, but because you stopped paying to reach people who were never going to buy.