TL;DR: LinkedIn clicks cost $10.48 in Q1 and $15.72 in Q3, per HockeyStack’s December 2025 benchmark of $28M in B2B SaaS spend. That price only pays off if your targeting is tight. This guide covers all seven attribute categories, Matched Audiences, Predictive Audiences, and the audience sizes LinkedIn actually recommends.
What LinkedIn targeting costs before it earns anything
LinkedIn sells the best-maintained professional dataset in advertising. Members type in their employer, job title, seniority, skills, and school. Then they update it when they get promoted.
That data carries a price. HockeyStack Labs analyzed $28M in LinkedIn ad spend across more than 70 B2B SaaS companies and published the results on December 15, 2025. Average cost per click was $10.48 in Q1 and $15.72 in Q3.
Compare that to Google Search. Across iClick-managed accounts in 2024 and 2025, our SaaS and B2B software clients averaged $4.88 per click at a 3.80% conversion rate. Our portfolio-wide Google Search average was $2.32 per click.
So a LinkedIn click runs roughly two to three times a Google Search click in the same vertical.
The trade is intent for precision. On Google, someone typed a query. On LinkedIn, nobody is shopping. You pick exactly who sees the ad instead. Loose targeting means you pay the premium and get the worse audience anyway. If you want the mechanics of the intent side of that trade, start with How Does Google Ads Work? The 2026 Plain-English Guide.
The three layers of LinkedIn Ads targeting
Every audience you build in Campaign Manager comes from one of three layers.
- Attribute targeting. LinkedIn’s own profile and company data. Job title, company size, industry, seniority, and so on.
- Matched Audiences. Your data, uploaded or captured. Website visitors, CRM lists, target account lists, and people who engaged with your ads.
- AI audiences. Predictive Audiences and Accelerate ad sets, where LinkedIn models who converts and expands from there.
Most accounts that underperform are stuck on layer one. Most accounts that scale profitably run all three at once, with different budgets and different offers.
Layer 1: the seven attribute categories
LinkedIn’s targeting documentation groups audience attributes into seven buckets: Location, Company, Demographics, Education, Job Experience, Interests and Traits, and Devices.
Three rules apply to all of them. Location is required. The minimum audience size to run an ad set is 300 member accounts. And you get a maximum of 200 selections, counting inclusions and exclusions together, on demographic-based targeting.
Company attributes: the strongest firmographic signal
This is where B2B money should go first. LinkedIn offers Company Industry, Company Name, Company Size, Company Revenue, Company Growth Rate, Company Category, Company Followers, and Company Connections.
Company Size plus Company Industry is the workhorse combination. It is stable, it is large enough to optimize, and it maps cleanly to how most B2B teams define an ICP.
Company Connections is the odd one. It targets first-degree connections of your employees at companies with 500 or more employees. Useful for events and recruiting, thin for demand generation.
Skip Company Name selection once your list passes about 50 accounts. Upload a company list instead. Typing names into the interface is slow and it misses subsidiaries.
Job Experience: title versus function and seniority
Job Title targeting is literal. It matches standardized titles, so “VP of Demand Generation” and “Head of Growth Marketing” are different selections. You will miss half your buyers if you only add the titles you already know.
Job Function plus Job Seniority is broader and more durable. Marketing plus Director, VP, and CXO catches the same people without you guessing at title variants. It also survives your prospects changing jobs.
Member Skills looks precise and is not. LinkedIn derives skills from a member’s job description, employer, industry, and education, not just the skills section on their profile. Treat it as a soft signal, never as your only facet.
Years of Experience is the cleanest seniority proxy when job titles are inflated, which they are in startups.
Interests, traits, and groups
Member Interests targets people by the topics they engage with. LinkedIn says it works best for awareness and consideration objectives. That matches what we see: it fills the top of the funnel and rarely holds a cost per lead target on its own.
Member Traits covers behaviors like frequent travelers or job seekers.
Member Groups is genuinely useful for niche communities, with one catch. LinkedIn removed Member Groups targeting for the European Economic Area and Switzerland on May 15, 2024. If you run pan-European campaigns, split your geos or the facet quietly does nothing for half your spend.
Demographics, Education, and Devices
Member Age and Member Gender are inferred, not declared. Use them only when a legal or compliance reason forces it.
Degrees and Fields of Study earn their place in narrow markets. Targeting law degrees, nursing credentials, or engineering disciplines beats title targeting when the credential is the qualifier.
Device OS and Device Type matter for one reason: LinkedIn traffic skews heavily mobile, and a form that needs six fields on a phone will not convert. Either fix the form or exclude mobile.
Layer 2: Matched Audiences beat LinkedIn’s data
Attribute targeting tells you who someone is. Matched Audiences tell you what they already did. The second one converts better, every time.
You have five practical options:
- Website retargeting. Requires the LinkedIn Insight Tag. Build separate audiences for pricing pages, demo pages, and blog readers. They are not the same person.
- Contact list upload. Your CRM, matched to member accounts by email.
- Company list upload. Your target account list, matched to LinkedIn company pages. This is the backbone of ABM on LinkedIn.
- Engagement retargeting. People who viewed a video, opened a Lead Gen Form, or engaged with a single-image ad.
- Company page visitors. Small, but high intent.
An audience starts delivering once it reaches 300 member accounts, per LinkedIn’s Matched Audiences documentation. Below that, your ad set sits idle and you will not get an alert about it.
Upload limits on contact and company lists are published by LinkedIn but were not verifiable at the source during this writing. LinkedIn’s own Campaign Manager documentation sets the bar: an uploaded list must have at least 300 rows to upload successfully, and the maximum list size is 20 MB or 300,000 records for a contact list (300,000 companies for a company list), with LinkedIn separately recommending at least 10,000 contacts, or 1,000 companies, so the list actually matches the 300 member accounts required to run an ad set.
Layer 3: Predictive Audiences and Accelerate
Predictive Audiences takes a seed, a Lead Gen Form audience, a contact list, a conversion audience, or a retargeting audience, and models members who look like converters. Accelerate goes further: it builds an ad set from your landing page, your audience signals, and your account’s historical data.
Two honest notes on the AI layer.
First, seed quality decides everything. A Predictive Audience trained on 4,000 form fills from a free ebook will find you more ebook downloaders, not more pipeline. Feed it closed-won contacts or qualified demo requests.
Second, you lose reporting granularity. You cannot see which title or company size drove a conversion. That is a real cost if your next budget conversation depends on showing where the pipeline came from.
Our rule: run AI audiences on 20% to 30% of budget once you have a clean conversion signal, and never before. It is the same logic behind how we sequence campaign types on Google, laid out in Google Ads Campaign Types in 2026: A Decision Tree.
How big should the audience actually be?
This is the setting most advertisers get wrong, in both directions.
LinkedIn’s own audience size guidance is specific. The minimum to run an ad set is 300 member accounts. LinkedIn suggests a minimum of 50,000 to drive results generally, a minimum of 300,000 for Sponsored Content and Sponsored Messaging, and between 60,000 and 400,000 for Text Ads.
Those numbers are guidance for efficiency, not law. Tight ABM audiences of 2,000 to 5,000 accounts work fine. They just cost more per click and take longer to gather signal, so budget and patience have to match.
The failure pattern is the middle ground: an audience of 8,000 built from four stacked facets, running $50 a day, reviewed weekly. It never gathers enough data to optimize and it never gets enough budget to matter. Pick a lane. Go narrow with high budget per person, or go broad with strong creative filtering.
Exclusions are the cheapest lever you have
Every dollar you stop wasting is worth more than a dollar you add. Four exclusions belong in nearly every ad set:
- Your own company, by Company Name.
- Current customers, via a contact list upload.
- Students and unpaid roles, via Job Seniority, unless you sell to them.
- Competitors and agencies, by Company Name or a company list.
Also turn off the LinkedIn Audience Network for your first 60 days. It extends reach to third-party apps at a lower CPM. It also makes it much harder to tell whether your targeting is working. Add it back once your on-platform numbers are stable.
Timing: your audience is not priced the same all year
The HockeyStack data shows a clear seasonal pattern. Click-through rate ran 0.82% in Q1, 0.96% in Q3, and peaked at 1.05% in September. Cost per click climbed from $10.48 in Q1 to $15.72 in Q3, a jump the report attributes to teams ramping budgets after summer.
Read that together: Q3 buys more engaged clicks at a much worse price. If your budget is fixed, front-load Q1 and Q2 for prospecting and hold Q3 spend for retargeting warm audiences, where the higher CPC is easier to justify.
Where LinkedIn should sit in your channel mix
LinkedIn is a targeting channel, not an intent channel. It works best when it feeds cheaper channels rather than carrying the whole number alone.
Our B2B SaaS results run largely on Google, where the intent already exists: 30% lower CAC, 60% more trial signups, and double the demo requests across the accounts we manage. Trademark Bob cut CAC 30% in under 90 days. Talent Room doubled demo requests in Q1 without raising spend.
LinkedIn’s job is to create the demand that later shows up as a branded search. That is also why video belongs in the plan, and the creative principles carry over from YouTube Ads for eCommerce: Creative Frameworks That Actually Convert.
For how each platform earns its slot, see our paid media channel hub and the LinkedIn Ads channel page.
Your first 30 days
Run this order. It works with a $5,000 test budget or a $50,000 one.
- Week 1. Install the Insight Tag. Upload your customer list as an exclusion and your target account list as an audience. Build one broad ad set on Job Function plus Job Seniority plus Company Size.
- Week 2. Launch two ad sets: broad attribute prospecting and target account list. Same creative in both so you are testing the audience, not the ad.
- Week 3. Add website retargeting once the tag has collected 300 member accounts. Split pricing page visitors from blog readers.
- Week 4. Compare cost per qualified lead, not cost per lead. Kill the losing facet. Do not add a Predictive Audience until you have at least 100 real conversions to seed it.
Most accounts we audit fail at week 4. They compare raw lead volume, keep the cheap audience, and wonder six months later why sales stopped following up.
Get a second opinion on your targeting
If you are spending $20,000 a month or more and your LinkedIn cost per qualified lead is moving the wrong direction, we will tell you why in writing. Our free 47-point written PPC audit is a PDF delivered in five business days. No sales call, no pitch deck. iClick has managed more than $61M in ad spend for 250+ clients since 2017 as a Google Premier Partner, top 3% in North America.
Request your free written audit at /services/ppc-audit.
Sources
- LinkedIn Marketing Solutions Help: Targeting options for LinkedIn Ads
- LinkedIn Marketing Solutions Help: Target audience size best practices
- LinkedIn Marketing Solutions Help: Matched Audiences best practices
- LinkedIn Marketing Solutions Help: Best practices for LinkedIn Ads audience targeting
- HockeyStack Labs: 2025 LinkedIn Ads Benchmark Report for B2B Marketers (December 15, 2025)
- LinkedIn Newsroom: About Us (member count)
- iClick Advertising: 2026 PPC Benchmarks (first-party account data, 2024 to 2025)
- iClick Advertising: Client case studies
- LinkedIn Marketing Solutions Help: Requirements for contact targeting lists in Campaign Manager (company list page states the same 300-row / 20 MB / 300,000 limits)
Frequently asked questions
What is the minimum audience size for LinkedIn Ads?
LinkedIn requires 300 member accounts to run an ad set, and Matched Audiences start delivering at that same threshold. For efficiency, LinkedIn suggests a minimum of 50,000 generally, 300,000 for Sponsored Content and Sponsored Messaging, and 60,000 to 400,000 for Text Ads. Smaller ABM audiences work, but expect higher costs per click.
Should I target job titles or job functions on LinkedIn?
Use Job Function plus Job Seniority as your default. Job Title targeting is literal, so it misses the variants you did not think to add, and it breaks when people change roles. Function plus seniority builds a larger, more stable audience. Layer job titles on top later as a separate test ad set, not a replacement.
Why is LinkedIn Ads so expensive compared to Google Ads?
You pay for targeting precision instead of buyer intent. HockeyStack’s December 2025 benchmark put LinkedIn CPC at $10.48 in Q1 and $15.72 in Q3. Across iClick-managed accounts, our SaaS and B2B software clients averaged $4.88 per click on Google Search. Nobody on LinkedIn is searching for you, so the platform charges for reaching the right person.
Are Member Groups still available for LinkedIn targeting?
Yes, except in the European Economic Area and Switzerland. LinkedIn removed Member Groups targeting in those regions on May 15, 2024. If one ad set covers both EEA and non-EEA geographies, the facet silently stops applying to part of your spend. Split your campaigns by region so the reporting stays honest.
When should I use Predictive Audiences instead of manual targeting?
Only after you have a clean conversion signal, roughly 100 real conversions of the type you actually want. Predictive Audiences model your seed list, so a seed built on ebook downloads produces more ebook downloaders. Start with closed-won contacts or qualified demo requests, and cap AI audiences at 20% to 30% of budget while you validate them.
