Marketing, AI, Thought Leadership

AI without infrastructure is just hype: Why CMOs are failing to capitalize on Generative AI

June 25, 2025

By Richard Wilson, CEO of Medialake

The AI arms race in marketing is accelerating—but most organisations are running before they can crawl.

According to IBM’s Institute for Business Value, 84% of CMOs report that fragmented systems are undermining their ability to adopt AI at scale, even as 81% acknowledge its transformative potential. In other words: CMOs are sold on the promise of generative AI, but very few have the foundation required to make it real.

This is the problem we see daily at Medialake. Everyone wants the rewards of intelligent automation—content that optimises itself, campaigns that learn and adapt, operations that scale seamlessly. But AI doesn’t operate in a vacuum. It needs structured data, connected systems, operational readiness, and most of all—visibility. Without these, the most advanced model in the world is just another shiny object gathering dust.

The Infrastructure Gap Is the AI Gap

The hype around AI is deafening. GenAI platforms like ChatGPT, Claude, and Gemini have opened the floodgates of what’s possible—but what’s being overlooked is what’s practical.

Only 17% of CMOs in IBM’s survey say they feel prepared to integrate AI into their workflows. A staggering 54% admitted they underestimated how complex AI adoption actually is. Why? Because it’s not just about installing a tool—it’s about rewiring how marketing teams operate.

According to Boston Consulting Group, over 71% of CMOs now plan to invest more than $10M annually in GenAI. But if that investment lands on top of disconnected tools, siloed content libraries, and legacy workflows, it won’t deliver. At best, you’ll get isolated wins. At worst, you’ll create even more complexity.

Marketing Needs a New Operating Model

Marketing teams today are drowning in complexity. Campaigns are increasingly multi-channel, content volumes are exploding, and compliance demands are rising. The truth is: you can’t automate what you can’t see.

The organizations that will win the next decade are those who treat AI not as an add-on—but as a core operating principle. That means building connected infrastructure where data, content, channels, and teams are unified. It means investing not just in AI models, but in the systems and governance that allow those models to thrive.

Take visibility, for example. If your content exists in 10 different systems with no consistent taxonomy or lineage tracking, how can AI tell you what’s working? If your campaigns are managed across siloed tools, how can AI automate optimisation? If your licensing, usage rights, or compliance data are stored in PDFs, how will GenAI know what it can and can’t use?

That’s why we built Medialake—to provide the infrastructure layer that makes AI useful. We unify the content lifecycle, connect the dots between systems, and enable brands to act on insights—not just collect them.

Culture Change Matters—But Structure Matters More

IBM’s research found that just 23% of CMOs feel their teams are culturally ready for AI—and only 22% of organisations have clear guidelines for how AI should influence decision-making. These are crucial elements. But readiness isn’t only about mindset. It’s about structure.

68% of CMOs said simplifying their technology stack would improve operational efficiency. We agree. Complexity is the enemy of scale, especially in AI. Simplicity isn’t a design choice—it’s a competitive advantage.

This applies directly to revenue impact. The IBM study found that aligning marketing, sales, and operations could increase revenue by up to 20%. For a $14B company, that’s $2.8B in growth left on the table—because systems can’t talk to each other.

The Real AI Advantage: Return on Use

At Medialake, we talk a lot about “Return on Use”—the ability to track not just what you’ve created, but what impact it had. That’s where AI can help most—but only once you’ve laid the foundations.

GenAI can dramatically improve content creation, customer experience, and speed to market. But to get there, you need:

  • Connected systems with shared taxonomies and accessible metadata
  • Operational intelligence that tracks content from creation to impact
  • Governance and compliance embedded in workflows—not tacked on later
  • AI-literate teams equipped with real-time tools—not manual dashboards

Final Thought: Before AI Can Scale, Infrastructure Must Lead

The CMOs we talk to are under more pressure than ever. 64% are now accountable for profitability; 58% for revenue growth. The promise of AI is that it can help meet those targets faster. But the risk is that without infrastructure, it becomes just another cost center.

As IBM’s Jonathan Adashek put it: “The companies that will dominate the next decade are the ones with the deepest AI integrations.” I’d add: they’ll be the ones who understand that integration starts with infrastructure.

AI isn’t a feature—it’s a transformation. And like all transformations, it only works when the foundation is strong.

Sources:

  • IBM Institute for Business Value, “Redefining the CMO” (June 2025)
  • BCG, “GenAI in Marketing: From Promise to Productivity” (June 2025)