The marketing industry and its agency ecosystem stand on the brink of their next AI revolution. The past year has seen the integration of Generative AI into existing third-party platforms – from Microsoft Copilot enhancing productivity to Adobe Firefly transforming creative processes. Meanwhile, many organizations have developed their own AI enterprise platforms or integrated AI into existing platforms – like Omnicom’s Omni, which embeds LLMs and diffusion models throughout its workflow. These platforms use engineered prompts for tasks from audience behavioral data to cultural analytics, accessing both structured and unstructured data. Furthermore, agent stores have emerged, democratizing access and enabling domain experts to create and deploy their own AI agents. Yet these advances, significant as they are, will be dwarfed by what’s coming next.
Beyond Analysis: The Evolution of AI Action
The journey from analysis to action is already underway. ChatGPT Actions demonstrated the first step, allowing AI to take varied actions directly on behalf of users. But the real transformation lies ahead: AI agents that can actually control computers and navigate complex systems independently. Anthropic’s Claude, with its Computer Use capability, already demonstrates how AI can take screenshots, click through applications, and fill out forms – effectively performing the functions of a human operator. OpenAI’s upcoming Operator, set to launch in early 2025, aims to expand these capabilities further, suggesting a future where thousands of virtual machines in the cloud running AI agents, each autonomously fulfilling requests from employees working anywhere in the world.
Breaking Down Marketing’s Biggest Challenge
Marketing organizations and their agencies have long struggled with fragmented systems – data siloed across DSPs, social and commerce platforms, CRM, ad servers, and more. Agentic AI could act across these disparate systems. While data cleanliness and accuracy remain crucial, AI agents could connect to each system in turn, making changes or flagging potential issues for human review.
Most of these systems already use AI to execute routine optimizations, but agentic orchestration could operate strategically across the whole ecosystem. Where today reports enable a marketer to make changes in line with campaign strategy, AI agents could do that automatically, leaving the marketer to focus on setting and evolving the strategy itself. By having agents handle routine decisions and optimizations, marketers could focus more time on strategic planning and creative innovation, leading to breakthrough ideas and market opportunities.
Human Control and Strategic Value
While these capabilities are powerful, they must be implemented thoughtfully. Even as AI models become more reliable, human oversight remains critical – AI agents cannot be held responsible for marketing failures. The approach should mirror digital programmatic campaigns, where AI operates within human-defined boundaries, with professionals maintaining control over strategic decisions while overseeing the multitude of smaller, tactical choices.
Understanding these tactical decisions remains crucial for marketing professionals. This knowledge enables adaptation to novel situations that AI cannot handle – such as sudden market shifts or unprecedented events – and helps develop the strategic leaders of tomorrow. Organizations must carefully balance AI automation with maintaining their core marketing expertise and the essential role of human intuition, creativity, and accountability.
Learning and Evolving in Real-Time
Perhaps the most transformative aspect of agentic AI will be this ability to handle routine decisions and optimizations, freeing marketing professionals to focus on strategic innovation. By managing day-to-day tasks and creating automated feedback loops, these systems will enable teams to operate at a higher strategic level, focusing on breakthrough ideas and market opportunities.
Picture marketing organizations where an orchestration layer handles routine analysis and optimization, while professionals focus on strategic planning and creative innovation. As AI agents manage tactical execution – from programmatic campaigns to content optimization – teams will be empowered to push boundaries and explore new possibilities in marketing effectiveness.
The Path Forward
The agentic future is not here – yet. Technical challenges around interoperability and capability need to be solved. There are also ethical considerations including environmental impact, how the models are trained, and the important areas of oversight and alignment to ensure AI is a superpower for workers rather than undermining human dignity and autonomy. Those issues are all being worked on today – and agentic AI represents an opportunity to enhance marketing professionals’ capabilities significantly. As these advanced AI tools emerge and mature, organizations that embrace these technologies early will empower their teams to achieve unprecedented levels of strategic impact and creative innovation.
This isn’t just automation – it’s about augmenting human expertise with intelligent systems that can handle routine tasks while enabling professionals to focus on what they do best: strategic thinking, creative innovation, and building meaningful brand connections. As these systems mature, they’ll unlock new possibilities for marketing excellence, elevating the entire industry to new heights of effectiveness and innovation.
Authored by
Mark Holden
Chief Strategy Officer, PHD Worldwide
And
James Aylett
Global Chief Data Officer, Annalect
