How AI is revamping out-of-home advertising
McDonald’s and Burger King used ChatGPT to make billboards as brands rely more on artificial intelligence in out-of-home campaigns
Fast food rivals McDonald’s and Burger King recently duked it out via ChatGPT-written billboards in Brazil, which became one of the most public examples of how major marketers are using artificial intelligence in out-of-home campaigns.
McDonald’s was the first major marketer to use ChatGPT to make an outdoor ad, asking the buzzy chatbot ‘Which is the most iconic burger?’ and plastering the answer (the Big Mac) on a billboard in Brazil. Burger King, not one to shy away from a good troll, responded by placing its own ChatGPT-written billboard next to the McDonald’s poster that read: ‘And which is the biggest?’ (the Whopper).
It was a comical stunt, but for marketers, it was also an indicator of what AI can do in out-of-home advertising and a signal that major marketers are ready to play in this space.
“McDonald's is always on the lookout for trends that impact society's consumption habits and conversations that can connect customers with the brand. In this context, the launch of ChatGPT quickly aroused the interest and adhesion of Brazilians, emerging as an interesting tool to reaffirm, once again, our relationship with the public,” a McDonald's Brazil spoksperson said.
McDonald's tested ChatGPT before launching the campaign by “teasing the platform by asking ChatGPT some questions. In this exercise, we noticed a pattern of answers and built questions that answered and positioned McDonald's in a true and authentic way,” the spokesperson said.
McDonald's is still measuring the results of the campaign, as it was so recent, but the spokesperson said “our main goal was achieved: we generated conversation and became a topic of conversation, in Brazil and abroad.”
“Burger King Brazil saw an opportunity, using the hype and trend of ChatGPT, to participate in a culturally relevant and topical conversation for all consumers,” a Burger Brazil spokesperson told Ad Age, declining to comment on the company’s broader AI strategy when it comes to outdoor advertising. “We are proud of the quality and love for our flame-grilled Whopper and know that our guests prefer its bigger size, which was the main message in this campaign.”
McDonald’s billboard came from Galeria.ag and Burger King’s was from the agency David.
Using AI to create campaigns for out-of-home ads, as McDonald’s and Burger King did, is one obvious way brands can adopt the tech in OOH. But AI is transforming the medium in other ways as well, including by making it more measurable and results-driven.
The global market for outdoor advertising was estimated at $28.2 billion in 2022 and is projected to grow to $39.2 billion by 2030, according to a recent Outdoor Advertising: Global Strategic Business Report, but some in the industry believe with the rapid adoption of AI, even more investment could be made in the space than current estimates call for.
“It's been really interesting in the out-of-home space because part of it is developing more dynamic, personalized creative,” said Jessica Romaniuk, president of the Chicago division of independent agency Two by Four. “But where we've been using [AI] more is in location intelligence data. So, tapping into AI and machine learning from a planning perspective, and then a reporting and analytics perspective. That's what's really changed the game.”
There are, of course, privacy concerns that come with using AI, especially in out-of-home, as the technology can be used to better track people in their daily lives, and brands are having to figure out how best to walk the line between innovative and creepy.
AI helps advertisers create more effective outdoor ads
AI farms a ton of new data that can alter an out-of-home campaign in real-time including weather, location, traffic patterns and date and time triggers. That means brands can much more effectively alter creative using AI, and deliver, through programmatic targeting, much more personalized messaging on a digital billboard based on whether it’s sunny or rainy, night or day, as well as hyper-localize billboards.
Huge Global Head of Data & Insights Frisco Chau said the Interpublic Group of Cos. agency is using AI to collect such data and serve “contextual-based offers” to clients via outdoor ads.
“So, it's a hot day, I'm near a supermarket,” Chau said, noting a supermarket could already tailor its surrounding billboards to promote discounts on ice cream using existing tools within digital out-of-home. He said generative AI can go even further by tailoring creative and offers on billboards to respond in real-time to cultural events and topics trending in the news.
“Programmatic targeting [with AI] is going to be the key to driving a better approach of dynamic creative,” Chau said.
Ad Age spoke to 10 experts in the out-of-home space who said most brands have just begun to explore what could be opened up in outdoor advertising using AI. They all agreed it will allow outdoor ads to be much more personalized, creative and measurable.
Kevin Bartanian, founder and CEO of out-of-home media company KEVANI, said he’s working with a client now on delivering creative in real-time, based on trending topics on search engines, to the company’s triple-screen, digital outdoor display, called The Trio, in Inglewood, California. Bartanian couldn’t talk about the specifics of the campaign since it’s not slated to launch until later this year.
“Everyone wants to share a personal experience,” said Bartanian, whose firm focuses a lot on creating viral moments out of out-of-home campaigns. “With AI advancement, we’re looking at where you can bring value for the consumer and create interesting conversations.”
Bartanian added AI can also be used to localize billboards down to a specific neighborhood in a town, which wasn’t possible before because a brand typically buys national inventory that includes many billboards across various locations. These billboards are difficult to keep track of and personalize creative to, especially in real-time, because of the immense manpower that would require. He said it can be as simple as using AI to tweak a digital billboard from saying, “Hi, L.A.” to “Hi, downtown L.A.”
“Those little taglines can personalize the message so it doesn’t feel like it’s a generic ad running everywhere,” Bartanian said. “There is also a library of images that AI can generate and pull based on geolocation to use on a billboard."
In another example, automakers could use cameras to detect the type of cars passing by and use generative AI to then pump out content in real-time based on those observations, he said.
“You can change creative every minute if you wanted to,” Bartanian said. “I don’t think the full usage [of AI] has been tested yet.”
AI opens up the possibility to create more engaging, in-person experiences with any brand, said Judah Longgrear, co-founder and CEO of ad tech startup Nickelytics.
“With augmented reality and computer vision, ads can respond to people in real-time,” Longgrear said. “For example, you could virtually try on clothes or accessories shown on a digital display. These interactive experiences grab attention and create memorable encounters with the brand.”
Wunderman Thompson New York recently started working with Choreograph, the holding company’s data products and tech company, to layer several types of “time and place” AI-derived insights spanning customer interests, weather patterns and social conversations, said Will Sandwick, chief data officer at the WPP-owned agency. From there, he said the agency identifies occasions in which campaigns can be delivered against.
While the agency is planning more than just out-of-home campaigns around these insights, Sandwick said the practice is especially helping to improve outdoor ads.
“Digital out-of-home is the perfect use case for this,” he said.
AI can create an occasion as localized and granular as backyard barbeques, zeroing in on, for example, neighborhoods on a nice, summer evening where data has shown a lot of barbeques are happening and tailor creative on surrounding billboards to accompany that type of “occasion,” Sandwick said. He said to think about, for example, a Brooklyn neighborhood where bus station billboards are visible from backyards or places where barbequing is allowed in parks that have outdoor ads.
Wunderman Thompson is in the early stages of using this type of AI to create out-of-home campaigns. Sandwick said its first pilot using the strategy should launch in a week or two.
He could not disclose specifics but said the agency is working with a lot of consumer packaged goods companies including Coca-Cola because CPG companies typically don’t have a ton of first-party data as they’re primarily bought through retailers.
Sandwick said with the rise in digital in the out-of-home space, geographical and time triggers have always been powerful in driving better ads, “but AI is supercharging it because we can do it on a very granular level.”
Alex Simpson, operations director of U.K. outdoor advertising operator 75Media, said the brands that use his platform are using these types of AI insights to create content and reach the right people at the right time with the right message, “maximizing impact and minimizing wastage. This also helps make DOOH campaigns more cost effective as brands can automate various aspects of campaign planning, content creation and optimization, resulting in cost-savings and improved ROI,” Simpson said.
Driving better measurement
Making more effective creative and driving better results for a brand go hand in hand.
Lucy Walker, head of media for marketing agency MG Empower, said her company has seen much stronger results from its digital out-of-home campaigns over the past few years because they can make smarter decisions on where billboards are placed, but AI will make ads even more impactful.
“DOOH is already providing us an opportunity to be dynamic with our messaging but with AI comes the ability to detect objects, track reactions and understand more of who the audience is,” Walker said. “So we’ll be able to have an understanding of traffic patterns, passer-by reactions and insight into who is passing by. Providing personalization has been something we’ve consistently talked about and through AI this is finally becoming available across DOOH.”
For one of Two By Four’s clients, Shedd Aquarium in Chicago, the agency used AI and machine learning to gain more insight across both web and foot traffic that its local city billboards attracted, Romaniuk said.
“In tandem with our programmatic buys, we’ve helped them understand not only foot traffic but also return on ad spend across channels,” Romaniuk said. “That's always been our focus as an agency: How do we tie everything together?”
There are also tools made better by AI such as facial recognition that are giving outdoor advertisers more insight into how people react to their ads.
Longgrear said AI-powered cameras and sensors on digital billboards can detect people’s emotions and adjust creative accordingly, effectively driving better outcomes.
“If the system detects happiness, the ad could display content that amplifies that positive vibe,” he said. “This emotional connection makes the ad experience more personal and impactful.”
Walker said MG Empower has been able to use some digital out-of-home screens with AI-powered cameras that offer facial recognition for client campaigns, but not at scale.
“We’re currently looking into the options available for our clients and the scale that we’re able to do this at,” Walker said. “Traffic patterns will allow us to fully understand the value of a placement beyond impressions and estimated reach but will provide insight into the demographic passing by. Facial recognition will allow us to not only track reactions but also demographic data that through smart DOOH ads will allow us to serve more relevant ads.”
Sandwick said AI is helping close the measurement loop within outdoor advertising because it’s delivering much more granular data—so, the data is informing brands on how their out-of-home marketing is impacting sales. That could, in turn, drive more investment in the channel, which is already seeing growth post-pandemic and because of other digital advancements in the space.
“I don’t know if I have the results yet to say, but I want to say that closing this measurement loop is going to increase the effectiveness of the channel and that would drive more investment,” he said.
Balancing privacy and innovation
Of course, using tools such as facial recognition and farming insights on people in their daily lives risks infringing on privacy rights.
“Privacy and data protection are probably the two key things brands need to keep in mind,” Simpson said. “There are other ethical concerns brands should be cautious about too, such as bias in AI targeting or content creation. AI technologies are evolving at a rapid rate though so brands should assess its capabilities and limitations before implementation.”
Chau said Huge tracks how people travel to and from work or to and from school, and deploys creative based on insights from those observations.
“That's very powerful, right?” Chau said. “It's also very scary and very creepy.”
To mitigate some of the privacy concerns that would come with doing something like that, Chau said Huge takes insights on a “cohort” versus individual level. That means taking insights from passersby observations without identifying individuals.
AI-powered facial recognition, which many advertisers are anxious to use, could inadvertently identify people though, Chau warned.
“On the creepy side, the technology is there with facial recognition and so forth for us to actually then start to go, ‘hey, that's Frisco Chau because there's a very distinctive facial recognition,’” he said. “Is that possible? Absolutely. I'm not sure we're there yet in terms of how we are applying the governance to stop us from that.”
Many of the people who spoke for this story advised advertisers to use AI as it is intended: as a tool for doing the work they are already doing, and not to overuse it.
“AI, as a whole, has shown to be a tremendous tool,” said Micah Freedman, VP of growth for digital agency Ruckus. “Like all tools, it’s only as good as the people who wield it.”
Freedman said Ruckus uses AI mainly to sell ideas to clients; the agency uses generative AI to create assets that it can use in pitches, for out-of-home campaigns and otherwise. But Freedman said agencies have to be cautious in using it.
“AI is still misunderstood and you have to be careful how you talk about it,” he said. “We haven’t gone to market with any assets completely made from AI and I wouldn’t suggest doing it unless you’re making a point, a la Burger King and McDonald’s.”
Read the article on AdAge.