The Role Agentic AI Can Play For Advertisers

Anyone who works in media planning and buying will be painfully familiar with the issues plaguing the industry. For the uninitiated, I’ll give you my overview.  

While important, we’re too focused on the often-siloed levers of optimisation, i.e. input costs, transparent - or otherwise. Unified measurement is difficult, leaving the door open to risky analysis and costly incorrect insights. Economic pressures mean everyone is having to do more with less, and decreasing investment into true working media is making it feel more acute. An agitator of this growing frustration is that we’re trying to address the above with 30 per cent open agency head count for critical digital roles.  

How did we get here and what can we do to be better? 

The digital and data talent gap has stayed steady over the years. This means that media departments have been trying to do their best with constantly depressed staffing levels while the talent – which is here, capable, and willing – is moving from one company to another in search of a place that can unleash their potential. They’re lured in with the promise of integrated marketing and let down by its soon-to-be-apparent lack. We’re undermining their motivation and excitement to shape what the future of media will become.  

Unlike other sectors of advertising, media has one resource all others are vying for: data. It’s a data-rich sphere but still underutilised at a truly enterprise level. In fact, too many processes are still manual and not integrated across business departments. And let’s be honest – they are also dull, repetitive, and uninspiring.  We’re simply not meeting talent at their potential.  

However, I firmly believe that the time is now to address these woes, and a solution is here that can transform media and allow our brilliant talent to instead focus solving our generation’s largest media disruption yet.  

Yes, I’m talking about AI, and specifically agentic AI.  

While efforts to automate have increased speed to execute, it is still built on an outdated end-to-end process increasingly irrelevant with today’s technological advancements. Highly manual, and even the greatest of case studies in media innovation are still often siloed from the rest of the broader business. Operating in a ‘media only’, or in best cases, ‘marketing only’ vacuum. ‘Better’ but not truly ‘different’.  

Within media buying and planning, the reality is that we’ve been working with what is readily available, scaled or already invested in, and trying to optimise with our limited aperture of access. But automation as a lever just isn’t cutting it anymore. We need autonomous decision-making to remove the tax on human ingenuity, and it’s at the very core of what agentic can do.  

Agentic AI uses sophisticated reasoning and iterative planning to autonomously solve complex, multi-step problems. This is the next frontier for artificial intelligence. Media’s use of Agentic AI can support the broader marketing ecosystem as well as the role that media plays for the entire business by acting, for example, as a barometer of real-time customer insight, behaviours and actions. Just think of the possibilities! 

Gen AI and agentic marketing will no doubt drive exponential business impact and outcomes. However, it is critical that we find a solution and consider these innovations with utmost care.  

Within Accenture, we’ve spent months testing, learning, and applying our findings to understand how agentic can be integrated systematically using an open-source, modular and customised approach at the service of clients’ infrastructure and needs; never monolithic, proprietary tech stacks.  

We’ve also been establishing how to use this frontier technology responsibly, in compliance with local regulations, keeping data secure within national borders, and enabling confident innovation aligned with legal and ethical standards. We’re already seeing the potential and results that this technology unlocks, for ourselves and our clients.  

We have started to embed agentic AI across our service delivery approach, driving efficiency, insights, and accelerating value for our clients through our market leading accelerators. Starting with our own Marketing teams, more than 600 professionals at Accenture are now equipped to use agentic to craft more effective, intelligent and rapid marketing campaigns and enhance their daily activities. And critically, to learn and transform. This journey is leading us to cut the steps in running the average campaign from 135 to 85 and get us to market 25 per cent – 35 per cent faster.  

This is just the beginning, and it’s not without hurdles to overcome. Adoption of new technologies is never linear, so there has to be a space to test, fail, and iterate. Talent will require upskilling and reskilling as the whole media process looks to be transformed. Done well, we can flip the talent model once and for all by providing more human strategy and media SME thought leadership, powered and supported by agentic workforce operations.  

 Embracing agentic AI is then like adding another member to the team, one that can function autonomously, and give everyone else the time and opportunity to apply their expert, strategic, and creative problem-solving skills to the issues that really matter.  

Kristen Kelly is Accenture Song's global head of media