5 AI-Driven Shifts Reshaping Innovation in Performance Marketing

On a recent episode of the B2B Influence Podcast, I sat down with Simon Poulton, Head of Innovation at Tinuiti, to discuss where B2B performance marketing is headed next. Our conversation covered everything from measurement challenges to emerging AI-driven discovery—but one theme came through clearly: many of the systems marketers rely on to define “performance” are being fundamentally reshaped.

Buyers are researching through AI tools. Attribution models are becoming less reliable. And new technologies—from generative search to agentic AI—are influencing how prospects discover, evaluate, and engage with vendors.

For B2B marketers, the question isn’t just how to optimize existing tactics. It’s whether the way we define and measure performance needs to evolve altogether.

Here are five pearls of wisdom that Simon dropped for our audience:

1. The Traditional Lead Funnel Is Losing Relevance

One of Simon’s first observations was that B2B marketers remain heavily focused on forcing prospects through traditional conversion flows—particularly gated content experiences. For years, the model has looked familiar: attract a visitor, require a form fill to access a whitepaper or report, and count that interaction as a lead.

But buyer behavior has changed dramatically. In B2B, purchasing decisions are highly considered and research-heavy. Increasingly, much of that research happens outside a company’s website—through industry communities, analyst reports, peer discussions, and search-driven discovery.

Many organizations still maintain what Simon called a kind of “gate infatuation.” The problem is that gating content often optimizes for the wrong metric: lead volume. If a company generates 100 leads but only one converts to pipeline, the number of leads alone isn’t a meaningful measure of performance.

The brands seeing stronger results today are often those publishing more information openly and focusing on building awareness and credibility earlier in the buyer journey.

2. The Rise of the LLM-Informed Buyer

Another major shift Simon highlighted is the emergence of what he described as the LLM-informed buyer. More professionals are now using AI-driven tools, such as generative search experiences or platforms like Perplexity, to research unfamiliar categories and complex solutions.

In many cases, these tools aid buyers to translate industry jargon, understand categories, and frame the right questions long before they ever interact with a vendor. That shift changes where and how brands must show up, and today, content visibility across the broader digital ecosystem matters more than ever. And the information marketers publish isn’t just serving human readers—it’s also shaping how AI systems interpret and represent their brands.

Simon noted that many B2B organizations still treat SEO as a narrow search discipline. In reality, modern SEO—or what some now call GEO or AI search optimization—is increasingly about holistic brand presence across digital conversations.

3. How Agentic AI Stands to Transform Lead Gen

One of the most intriguing ideas Simon raised during our conversation was the concept of agentic lead generation. We’re already seeing early examples of agentic commerce, where AI systems help consumers research products and even complete purchases. Simon believes similar models may emerge in B2B environments.

In this scenario, buyers might rely on AI assistants to support vendor evaluations, compare solutions, and narrow down options. Instead of manually visiting multiple vendor websites, the AI agent could coordinate initial conversations or product demos with a shortlist of providers. And, instead of filling out forms across multiple vendor websites, the AI agent could facilitate those introductions directly.

In that scenario, the traditional B2B lead form—the entry point to the funnel for decades—may start to lose relevance as AI agents handle early research, vendor comparison, and even demo scheduling on behalf of buyers.

While this shift is still emerging, it highlights how quickly the mechanics of lead gen could evolve in an AI-mediated research environment.

4. Attribution Is Becoming Less Reliable

We’ve discussed this with many of our podcast guests: measurement remains one of the most challenging areas in performance marketing. Like others, Simon believes many marketers are relying too heavily on deterministic attribution models.

The problem is that these models often credit the most visible touchpoint, not necessarily the most influential one. For example, last-click attribution frequently gives disproportionate credit to search ads because they appear at the moment of conversion. But that doesn’t mean search created the demand in the first place.

Simon is increasingly advocating for a shift toward incrementality testing, which focuses on understanding whether marketing activity actually drives incremental results. Incrementality experiments examine what would happen if a campaign or channel were removed entirely. By comparing performance in test and control groups—often through platform or geographic experiments—marketers can better determine whether a channel truly contributes to growth.

In a world where privacy restrictions and platform modeling make attribution more complex, incrementality testing offers a more reliable way to guide media investment decisions.

5. Agencies Are Evolving From Executors to Innovators

Finally, we discussed how the role of agencies is changing in an environment where many advertising platforms now automate campaign optimization. Tools like Google’s Performance Max (Pmax) and similar AI-driven systems have commoditized many tactical aspects of media buying. But that doesn’t eliminate the value of agency partnerships.

Instead, Simon sees agencies adding the most value in two key areas. The first is advanced experimentation and optimization—finding the incremental performance gains that automated platforms alone may miss. The second is technical marketing innovation. Simon described the rise of what he calls the “hyphenated marketer”—professionals who combine marketing expertise with coding, analytics, or data science skills.

Agencies increasingly accelerate this capability for brands by building tools, testing frameworks, and building data models that internal teams can adopt and scale. In that sense, the agency-client relationship may be entering a new phase—less about executing campaigns and more about enabling innovation.

For B2B marketing leaders, the biggest takeaway from the conversation with Simon isn’t simply to adjust tactics—it’s to rethink how performance itself is defined and measured.

We invite you to connect with Simon here, and please reach out to our team to find out how we use marketing data to drive measurable growth for clients. Whether you want to discuss audience building, B2B2C data, intent data, email campaign management, or any of our other services, we welcome the conversation.