LinkedIn Ads Guide 2026: How to Run B2B Campaigns That Actually Convert

LinkedIn Ads are the most expensive CPCs in digital advertising — often £8–£20+ per click for B2B audiences. That cost makes most marketers dismiss them prematurely. But LinkedIn also offers something no other platform has: the ability to target by job title, company size, industry, seniority, and skill — simultaneously. When the targeting is right and the offer is strong, LinkedIn Ads can generate qualified B2B leads at a cost that no other channel can replicate.

When LinkedIn Ads make sense (and when they don't)

LinkedIn Ads work when: your average deal value is high enough to justify £15–£50 per click (typically enterprise SaaS, B2B services, consulting, financial products); you're targeting a specific role in a specific type of company and Google can't isolate that audience effectively; you need to reach decision-makers who are actively researching solutions in a professional context. LinkedIn Ads don't work when: your product serves a broad consumer audience; your deal value is low; you're expecting direct response from cold audiences — LinkedIn requires nurturing.

LinkedIn campaign types

• Sponsored Content (single image, carousel, video): appears in the LinkedIn feed — best for awareness and lead generation at the middle of funnel
• Message Ads (InMail): direct messages to targeted inboxes — high open rates but must offer genuine value to avoid spam perception
• Dynamic Ads: personalised ads using the viewer's name and photo — effective for follower acquisition and brand building
• Text Ads: small sidebar ads — low CPCs but also very low CTRs; useful for remarketing to known audiences
• Lead Gen Forms: native forms that auto-fill LinkedIn profile data — significantly reduce friction vs landing pages, typically 3× higher conversion rate than external landing pages

LinkedIn targeting: how to set it up right

• Job title targeting: LinkedIn's most precise B2B tool — target specific titles (CFO, Head of Marketing, VP of Engineering) but also include seniority + function combinations to capture people with the same role but different titles
• Company size: filter by number of employees to focus spend on the companies you actually want
• Industry targeting: combine industry with job function and seniority to reach the right people at the right type of company
• Matched Audiences: upload your own lists (existing customers, leads from your CRM) to create lookalikes or exclusions
• Lookalike audiences: LinkedIn's algorithm finds profiles similar to a Matched Audience — useful for expanding reach while maintaining quality
• Audience expansion off: turn off LinkedIn's 'Audience Expansion' feature by default — it broadens targeting beyond your specifications

Making the economics work

The only way to justify LinkedIn's cost is to track revenue, not just leads. A £1,000 CPL (cost per lead) is fine if each lead has a 20% close rate and a £10,000 average deal value. You need: conversion tracking on LinkedIn (use the LinkedIn Insight Tag and set conversion events); CRM integration to track lead quality through to close; a content offer worth the high CPL (a gated white paper, a free audit, a demo — not a 'contact us' CTA). Calculate maximum acceptable CPL = deal value × win rate × target margin, then engineer campaigns to stay under that number.

LinkedIn Ads creative that performs

• Lead with the pain, not the product: 'Most [job title]s waste 20% of their ad budget because of this tracking gap' outperforms '[Product name] — the B2B analytics platform'
• Use social proof natively: reference specific company names, job titles, or industries your product has helped — LinkedIn users respond to professional peer validation
• Video performs well for awareness but poorly for direct response — use single image ads with a clear value proposition for lead gen
• Keep copy concise: LinkedIn users are in a professional browsing context, not an entertainment context — get to the point in 2–3 lines
• Test 3–5 creative variations per campaign: LinkedIn's relatively small audience sizes mean statistical significance requires more time than Google Ads — run tests for at least 2 weeks before concluding