LinkedIn's 121% ROAS Isn't the Story. The B2B Attribution Shift Is.

If you're still evaluating LinkedIn by CPC or CPL, you're using the wrong map for the territory.

That's the core message of Dreamdata's 2026 LinkedIn Ads Benchmarks Report, which aggregated 66 million sessions and 3.5 million B2B customer journeys. The report generated headlines with its headline number — 121% ROAS, the only positive B2B return among major ad platforms. But focusing on that number alone misses the deeper shift.

The real story isn't that LinkedIn finally delivers positive ROAS. The real story is how it does so — and what that reveals about a measurement system that's been broken for years.

The Contact-Level Trap

Here's the problem in one number: LinkedIn CPC sits at roughly $5.74, while Meta costs $1.60 and Google Search costs $5.19. A marketer looking at those numbers would naturally ask: why would I pay 3.5x more per click for LinkedIn when Meta gives me cheaper traffic?

The answer: because B2B doesn't sell to individuals. It sells to companies.

Dreamdata's data shows the average B2B deal now involves 10 stakeholders, across 88 touchpoints, over 272 days. No single stakeholder makes the decision alone. And 81% of that journey happens outside the formal sales pipeline — meaning most of the influence occurs before your CRM even sees the prospect.

When you shift from contact-level to company-level metrics, the cost comparison inverts:

LinkedIn appears expensive at the individual level because it reaches multiple stakeholders within the same account. A single ad impression on LinkedIn might touch three decision-makers at the same company. On Meta, you're usually reaching one individual with no organizational context.

As the agency managing B2B campaigns for clients across LATAM and the US, we've seen this pattern play out consistently: clients who optimize for CPL see LinkedIn as expensive. Clients who measure cost per account influenced see LinkedIn as their most efficient channel. The difference isn't the platform — it's the metric.

The Three Numbers That Actually Matter

Beyond the headline ROAS figure, three data points from Dreamdata's report tell a more complete story:

1. LinkedIn commands 41% of B2B ad budgets (up from 39%).

This is happening in parallel with something significant: non-branded Google Search budgets dropped from 37% to 33% of total B2B spend. CPCs on those non-branded terms jumped 29%, while CTRs fell 26%. The structural driver is AI Overviews answering informational B2B queries directly — reducing click-through rates for the kinds of research queries that historically fed B2B search campaigns.

Money is moving from search to social in B2B, and LinkedIn is the primary destination.

2. LinkedIn's cost per company influenced dropped from €154 to €70.11.

That's a 54% efficiency improvement in two years. LinkedIn achieved this through better targeting (Predictive Audiences replacing the retired Lookalike model), tighter integration with CRM data via the Conversions API, and the Company Intelligence API that connects organic page activity to revenue.

3. LinkedIn influence grows as deals get closer to close.

This is the most counterintuitive finding. LinkedIn's influence share:

• 24.2% at MQL stage
• 30.2% at SQL stage
• 28.3% at New Business stage (up from 15% the prior year)

Unlike previous years, LinkedIn's influence doesn't decline when sales engages. It strengthens. Buying committees are using LinkedIn to research vendors throughout the evaluation process — not just at the awareness stage.

What Drove the Shift

Three structural forces converged to create this change:

AI Overviews are compressing Google search budgets. As we covered in our Google Search Console AI reports analysis, generative AI answers on Google are reducing click-through rates for informational queries. For B2B marketers who historically relied on Google Search for demand capture, this erodes the channel's cost-effectiveness. Those budgets flow to LinkedIn, which captures demand earlier in the journey.

LinkedIn rebuilt its targeting infrastructure. The retirement of Lookalike Audiences in favor of Predictive Audiences — now powering 41% of Sponsored Content spend — shifted LinkedIn from demographic targeting to intent-based targeting. Combined with the Company Intelligence API, LinkedIn can now connect organic page visits to revenue in a way that was previously impossible. Our guide to LinkedIn campaigns covers how these changes affect campaign structure.

The buying committee is growing. The jump from 6.8 to 10 stakeholders per deal reflects a broader trend in B2B purchasing: more sign-offs, more security reviews, more procurement involvement. When the buying group expands, channels that reach multiple stakeholders per account (LinkedIn) gain relative efficiency compared to channels that reach individuals (Google, Meta).

Mintec's Framework: The Four B2B Metrics That Change Budget Decisions

After running B2B campaigns across all major platforms, we've developed a simple framework for moving beyond the contact-level trap. These four metrics replace the standard CPC/CPL/CPA dashboard:

1. Cost per Company Influenced (CCI) Total LinkedIn spend ÷ number of unique companies that engaged with any ad. This is the single most important metric for B2B budgeting. If your attribution system can't measure this, it can't properly evaluate LinkedIn.

2. Funnel-Stage Influence Share What percentage of touchpoints did LinkedIn contribute at MQL → SQL → Closed Won stages? If LinkedIn influence drops at any stage, that's a creative or targeting problem — not a platform limitation. The Dreamdata data proves LinkedIn can work across the entire funnel.

3. Buying Committee Reach Of the 10 stakeholders involved in a typical deal, how many did your LinkedIn campaign reach? A campaign that reaches 7 of 10 decision-makers is more valuable than one that reaches 2 of 10, even if the CPL is higher.

4. Attributed Revenue Time-to-Close Measure the average time from first LinkedIn impression to closed revenue. Dreamdata reports 281 days. If you're measuring LinkedIn performance on a 30-day attribution window, you're capturing roughly 10% of its contribution.

What This Means for Q3 2026 Budgets

The practical implications for B2B marketers are straightforward:

Extend your attribution window for LinkedIn to 12 months. Anything shorter will systematically undercount the channel's contribution. B2B deals take 272 days on average — your measurement model needs to match the reality of the buying cycle.

Integrate LinkedIn's Conversions API if you haven't. Dreamdata's data shows CAPI users see 20% lower CPA and 31% more attributed conversions. This isn't optional anymore — it's the baseline for accurate measurement.

Layer Account Targeting with Predictive Audiences. The combination performs 2.7x better on conversion rate than industry-plus-seniority targeting alone. Predictive Audiences narrows on conversion-likely individuals; Account Targeting narrows on the right companies. Both are needed.

Monitor Google search ad performance alongside LinkedIn performance. If your Search CTR is declining while CPCs rise — as Dreamdata's data shows across the B2B industry — the window for reallocating that budget to LinkedIn is now.

LinkedIn's 121% ROAS isn't the story. That number is the symptom. The real shift is that B2B marketing measurement has been optimized for a world that no longer exists — where buyers researched alone, decided quickly, and tracked neatly through a CRM pipeline.

The winners in Q3 2026 will be the teams that stop optimizing for cheap clicks and start optimizing for company influence.