Study reveals Agentic AI divide: Some marketers are scaling - and others are stalling

75% of agentic AI early adopters say they’re using AI to its full potential; half have already incorporated quantum into their innovation roadmaps

Agentic AI is reshaping marketing - and not everyone is keeping up, says a new global study from SAS and Coleman Parkes, Marketers and AI: Navigating New Depths. While just over half (51%) of marketing decision-makers say their organisations plan to invest in agentic AI within the next year, only 21% are actively testing it in live environments. The result? A widening gap between those who are scaling agentic AI and seeing ROI and those still watching from the sidelines.

In EMEA, the picture is even more stark. While 85% of marketers in the region report ROI from GenAI tools - close to US levels - only 5% say they have robust governance frameworks in place for agentic AI, compared to 15% in the US. The lack of governance maturity risks slowing adoption in a region otherwise well positioned to capitalise.

1 in 5 marketers are testing Agentic AI

Agentic AI goes beyond content generation. It enables autonomous and active decision-making, continuous learning, and real-time optimisation. But it doesn’t happen by accident. A strong generative AI (GenAI) foundation - with mature data pipelines, ROI proof points, and a strategic framework - is what makes scaled autonomy possible.

In EMEA, marketers have leaned heavily into text generation (48%) as their leading GenAI use case, ahead of APAC (44%) and the Americas (41%). This content-first maturity is a potential springboard into agentic AI, if governance and strategy can keep pace.

The three populations of agentic AI

Marketers now fall into three distinct maturity groups: Observers, Planners, and Adopters. Adopters are already using agentic AI, Planners intend to within the next year, and Observers intend to within the next two years. While most are still planning or piloting, Adopters are already scaling agentic systems - and the performance gap is growing fast.

  • 63% of Adopters are implementing enterprise-wide AI strategies, compared to just 25% of Observers.
  • 42% of Adopters use AI extensively across campaign channels for real-time decision-making — versus just 1% of Observers.
  • 3 in 4 Adopters say they’re using AI to its full potential. Among Observers? Just 7%.

The difference isn’t just in deployment - it’s in mindset. Only 3% of Observers report a strong understanding of agentic AI, compared to 61% of Adopters. Familiarity breeds confidence, and confidence fuels innovation.

What Adopters are doing differently

Agentic AI Adopters are already deploying advanced use cases like continuous learning agents (47%) and performance reporting (45%).

Trust is conditional

Despite growing adoption, trust in agentic AI remains conditional. While 90% of marketers express some level of trust, fewer than 1 in 20 say they trust it fully. And while 79% of marketers feel somewhat confident in their organisation’s AI governance, very few are strongly confident.

In EMEA, governance is the clearest barrier: just 5% of organisations describe their agentic AI governance as well established. This is the lowest share of any region - and a clear signal that confidence and oversight must improve for the region to scale.

Quantum enters the chat

Agentic AI maturity isn’t just about what marketers are doing today - it’s also about how they’re preparing for what’s next. And for many Adopters, that future includes Quantum Computing.

Half (50%) of agentic AI Adopters globally have already incorporated quantum into their digital or innovation roadmaps - compared to just 25% of Observers. For EMEA organisations, the journey is likely to be slower given governance gaps, but the trajectory is the same: once maturity is established, quantum is firmly in view.

About the research

The report is based on a new survey of 300 organisations worldwide, ranging from SMBs to enterprises with 10,000+ employees. The respondents are marketers who work at organisations across a range of industry sectors.

About SAS

SAS is a global leader in data and AI. With SAS software and industry-specific solutions, organisations transform data into trusted decisions. SAS gives you THE POWER TO KNOW®.

SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of SAS Institute Inc. in the USA and other countries. ® indicates USA registration. Other brand and product names are trademarks of their respective companies. Copyright © 2025 SAS Institute Inc. All rights reserved.

Editorial contacts:

Visit the SAS Newsroom
https://www.sas.com/en_gb/news.html