Media Buying Briefing: How agencies balance media and creative as AI projects ramp up

This Media Buying Briefing covers the latest in agency news and media buying for Digiday+ members and is distributed over email every Monday at 10 a.m. ET. More from the series →

As media agencies develop more artificial intelligence-driven work, they are relying on data and media tools to inform the creative side.

Whether it’s developing proprietary media and analytics applications or establishing a wider practice to guide the AI and ethical areas of their campaigns, various agencies specializing in commerce, innovation and media are incorporating these tools into their strategies.

At IPG’s design and innovation arm Huge, the amount of AI client projects has steadily increased as the company added new leadership to support these areas. This June, the agency said it expanded its client projects that involve using an AI component to 25%, including in its work with clients Google, Darling Ingredients and Hublot.

“It was a conscious effort [to increase that number],” said Lisa De Bonis, CEO of Huge, though she did not say how many brands that represented.

Yet one potential challenges with more generative AI content is sometimes the wasted creative output. CreativeX research recently pointed to 90% of global campaign toolkits never getting activated by local markets — and 45% of those core assets related to the brand image are not used in any way.

De Bonis said striking this input and output balance is important to Huge as it incorporates AI into its broader practice: “We’re not just using it to make stuff and pump it out. We’re using it to inform our thinking and to be smarter about what we put in front of our clients and out in the world.”

Building a broader AI practice

With Google, Huge is working on incorporating AI using text-to-speech into the Keyword blog and news site that make product news and content more accessible to readers. With food sustainability company Darling Ingredients, the team is helping launch a new flagship digital experience using imagery generated by AI, and it is similarly developing AI prototypes and personalized experiences for the watch brand Hublot.

The Huge team calls these “intelligent experiences,” or IX, as part of its developing practice, which range from creating personalized experiences to other forms of digital experiences that focus on making everything shoppable and tech-driven.

De Bonis sees it as “the next wave of experiences, [in particular] catapulted by generative AI.”

It’s been a part of Huge’s larger data and AI investments these past three years to be able to accelerate their data-driven insights and client innovations, she explained. Since the explosion of generative AI, Huge said clients have also asked to leverage the tools to boost efficiency and drive brand growth.

Huge’s new executives include a new chief technology officer, chief creative officer and executive creative director — with the team reporting to Huge’s recently-installed chief product officer, Chidi Achara, who will oversee both the creative and technology directions as a way to build on this practice.

“I wanted to structurally create that connectivity and partnership,” De Bonis told Digiday.

Custom tools

At other agencies, the evolving media and creative strategies related to AI have resulted in creating custom platforms or building on existing applications to incorporate AI features.

Josy Amann, co-founder and managing director of Media Matters Worldwide, said AI and machine learning is used throughout all their operations as a media agency — from analytics and audience research to platform optimizations and vendor RFP processing. They include a mix of custom-built and existing tools used throughout billing, data, client metrics and project management.

One of its proprietary products includes Agile Mix Modeling, which uses machine learning to provide clients with weekly ROI readouts by channel, Amann explained. “This is a game-changer because traditional attribution is now obsolete. However, this type of output is essential for accurately assessing each channel’s value, ensuring that we allocate our clients’ budgets responsibly and effectively,” Amann said.

Right now the agency does not charge any differently for its AI-focused work, but Amann mentioned Agile Mix Modeling could become a “standalone product” in the future.

Commerce-focused Mars United Commerce also employs its in-house platform for planning purposes, Marilyn Commerce Media, with an estimated 50% of its client work currently using AI, such as automation across its search and programmatic campaigns, said Ethan Goodman, EVP of global digital commerce at Mars, who didn’t provide other exact figures.

“All of our media work is already involved in AI in some way,” Goodman said.

Goodman sees AI increasingly used for automated bid management and audience or placement optimization in their space. Clients and creative shops will continue using it for their content development, but he said these parts could eventually combine to become part of an “integrated creative audience and placement optimization process.”

In the long run, agencies also see a continued need for people to be a part of the media planning process — even as AI plays a growing role.

“AI is part of the process, rather than the full deliverable,” said Anne Buehner, head of creative at Code3, which uses AI throughout media and creative for project management, drafting content and research.

“While AI will play a huge role in automation and delivering the first 80-90% of the media plan, our people will use intuition to find the unique opportunities that allow media to connect with consumer needs and creative ideas,” Goodman added.

Color by numbers

We’ve been writing about attention metrics for a while, but advertisers are still torn about these measurements’ ability to address current challenges. At Cannes, research from Infillion found 47% of brands and agencies believe attention measurement doesn’t address the new-age problems of ad avoidance and device multitasking. More stats:

  • 42% of advertisers said attention measurement is a proxy metric.
  • 37% said attention measurement fails to distinguish between earned and interruptive attention.
  • 90% of consumers have strategies to evade digital ads — and 25% avoid ads because there are too many. Methods include skipping them, switching devices or closing pop ups.

Takeoff & landing

  • GroupM named Alycia Mason CEO of Wavemaker U.S., effective the end of July when she starts, and reporting to GroupM North American CEO Sharb Farjami. Mason most recently was a vp at McDonald’s but has experience at Publicis, OMD and Mindshare. She also cofounded a confectionary company called Candycopia.
  • Interpublic Group‘s Jarrod Martin, current Global CEO of Kinesso, added oversight of Acxiom, IPG’s data, identity resolution and cloud marketing services unit, to his remit.
  • Omnicom Media Group’s Hearts & Science won media AOR duties for underwear and apparel maker HanesBrands, and the work includes integrated media strategy, planning, buying, and measurement for DTC and retail. Publicis’ Spark Foundry was the incumbent agency.
  • According to reports, Publicis Media landed Yum Brands’ media business in China, representing some $400 million in media spend. The incumbent was GroupM.

Direct quote

“[The holding companies are] looking for smaller clients. Their models are predicated on things like data automation, technology automation. So holding companies’ ability to compete for budgets that probably three or four years ago they weren’t competing for are much more prevalent and paramount, and we have to up our game, so to speak, to make sure we are competitively fit for purpose, to punch at that weight and compete with holding companies.”

— Doug Grumet, evp and general manager, Amp.

Speed reading

  • Michael Bürgi wrote about a new full-service operation called Amp going after mid-market clients.
  • On the AI beat, Marty Swant explained why Toys“R”Us and OpenAI Sora made an AI video.
  • Krystal Scanlon covered NBC Universal’s ad dollars in the Olympics.

https://digiday.com/?p=549081

More in Media Buying

How influencer shops and agencies are adding content studios to boost production speed, revenue

Influencer shops expand content studios to boost production speed and drive additional revenue streams. Depending on who you talk to, they get there in different ways.

AI Briefing: Index Exchange and Cognitiv integrate to use generative AI for programmatic curation

The companies are the latest to bring deep learning and large language models to media buying tools.

Media agency performance analysis

Media Buying Briefing: A tale of two reports (on media agencies)

Where Forrester’s latest Wave report generally gave media agencies high marks for integration of multiple skills, MediaSense/WFA cited marketers’ desire to change remuneration toward business outcomes.