PALM SPRINGS, CALIF. — Marketers and agencies are still just scratching the surface when it comes to understanding how generative artificial intelligence tools like large language models are changing the way brands develop their content strategies. For example, brands are now rethinking the role their own websites play in informing AI programs that consumers increasingly rely on.
“The LLMs are going back to the brand’s website as the first source of truth for information about that brand before they’re going to the retailer’s website,” April Carlisle, evp, commerce at Spark Foundry, told Beet.TV at the IAB Annual Leadership Meeting. “Our clients are having to go back and think about how is my brand website continually being updated with all the information and ratings that we use and the places and spaces that the LLMs are pulling from as their source of truth.”
This reverses years of focus where retail brands invested heavily in Amazon and Walmart product pages while treating their own sites as secondary destinations.
Intent replaces identity starting point
Traditional media planning begins with identity and reach, but LLMs start with indicated intent based on purchase history to project future buying behavior and present immediate purchase options.
“The LLMs get very surgical based on your history of what you have been buying and what you may be buying in the future and it’s projecting what they think your next purchase would be and then putting it right in front of you with a platform in place to purchase that item,” Carlisle said. “It’s a very different way of approaching the consumer experience and the shopper journey.”
Spark Foundry is examining what it calls the “agentic shopper journey experience” and how it differs from traditional paths.
Reddit emerges as key source
Reddit has become a significant LLM information source, prompting brands to evaluate their presence on platforms they may have ignored months ago.
“There’s a lot of discussion right now that Reddit is one of the key places that the LLMs are pulling from,” Carlisle said. “Many of our clients are thinking about, well, how is my brand showing up on Reddit? Something that they may have not spent as much focus on just three months ago or six months ago.”
Live sports enables in-game commerce
Streaming sports creates opportunities for real-time commerce including ordering food and snacks with 54-minute delivery while games continue, reaching next-generation audiences through platforms where traditional shopping interruptions don’t apply.
“You’re in game and you can order food and snacks before the game is over and you push on a button, products are delivered in 54 minutes while you’re watching the game,” Carlisle said.
Spark Foundry established a sports commerce focus helping clients think through in-game experiences that capitalize on live viewing moments.
AI autonomy enables strategic focus
Spark Foundry has used artificial intelligence for several years, with increased automation of management tasks freeing time for strategic thinking rather than tactical execution.
“The more we can autonomize the management, the more we have time to actually look at the strategy,” Carlisle said.
The biggest shift involves retailer partnerships with AI platforms as Amazon, Walmart, and Instacart integrate LLMs into shopping experiences.
“When you look at the activities on Amazon, Walmart and Instacart and how they are thinking about the role that the LLMs will have in the shopping experience,” Carlisle said, “that’s where things are really interesting.”
You’re watching Beet.TV coverage from IAB ALM 2026. For more videos from this series, please visit this page.
