The term, which means generative engine optimization, refers to optimizing web content to show up in results from AI-driven search platforms, like ChatGPT or Google Gemini. It’s a new way for brands to stand out and reach customers, especially considering that nearly 60% of U.S. consumers have used a generative AI tool for help with online shopping, per an August 2025 survey from Omnisend. If a customer asks an AI engine about buying lip gloss or a T-shirt, the relevant brands want to make sure they show up.

But it’s not an easy journey, executives conceded this week in interviews with Modern Retail and during panels at Shoptalk Fall in Chicago. Brands and retailers have spent decades crafting their SEO playbook, figuring out which keywords to include on product pages, blogs and additional web content to show up in Google or Yahoo. Now, with GEO, platforms like ChatGPT and Perplexity are pulling and ranking information — oftentimes, in response to hyper-specific inquiries, rather than one-word or two-word requests. This could also have financial consequences for brands, especially as these agents start enabling checkout.

Getting ahead of GEO is a new reality for brands, as Jose Nino, vp of global digital and e-commerce for the U.S. Polo Assn., said during a panel on Wednesday. He mentioned searching for running shoes on Google Search a month ago, and then a week ago, and getting completely different results from the “AI mode” at the top. With GEO, he said, “You can bid all you want, but the Gemini responses are taking over, and you don’t even know exactly what’s pushing that response. … Suddenly, someone is controlling that narrative, and you don’t know how they operate it yet. If anyone says they know exactly what’s driving it, they’re not 100% sure.”

At the same time, GEO is also a big opportunity — a white space where brands are eager to experiment. “If you’re not keeping an eye open and really understanding what’s happening in the GEO space, you are going to miss out,” Erica Randerson, chief digital officer and gm of Edible Brands, said on Thursday.

How companies are gaming GEO

Brands and retailers are taking different steps to get ahead when it comes to agentic AI results, although many acknowledged that there are a lot of unknowns. As Brad Larabell, director of performance marketing at Every Man Jack, told Modern Retail, GEO “is really still in its infancy, compared to SEO.”

Ranjeet Bhosale, vp of digital product management at Target, said that “the future of SEO is not just about keyword ranking.”

“The future of SEO is to come into GEO,” he said during a panel on Thursday. “We have to make sure we are training the agents so they can understand and represent our products effectively.”

Bhosale explained that Target now prioritizes five key aspects with search: price, product, promotions, availability and policies. It’s making sure its website is updated accordingly to make this data “machine readable” and ready for AI engines to pull from — whether internally (through Target’s agents) or externally. “That way, when the guest is searching for ‘healthy dinner for four, gluten-free options, less than $20, available for pickup now in Atlanta,’ the guest is getting results which are relevant and are in stock in their particular locality,” Bhosale said.

At Edible Brands, the company is “focused on long-term content that helps answer questions,” Randerson said. “Tapping into the content our teams have already been creating, and having it be much more of a Q&A-type dialogue, we have found, really helps in terms of GEO optimization.” From a technical standpoint, she said, Edible Brands is also making sure metadata is “structured appropriately.”

“It’s really important that these agents and generative LLMs are able to read and understand the content [and] the value that you can bring to consumers,” Randerson said.

At Credo Beauty, the company is “still very early on” when it comes to GEO, Rebecca Armstrong, director of e-commerce at Credo Beauty, told Modern Retail. But the retailer is also making sure it’s staying on top of what people are searching for. “Just this morning, we saw people searching for ‘foundation for olive skin tones,’” she said. “So, how do we incorporate that [into our messaging]?”

However, she acknowledged that brands can’t tweak their websites for every hot search term. “We are small, so we can’t be updating every little thing, only for it to change two days later.” She mentioned the brand wants to take “about a month” in between updates, as the e-commerce team is currently made up of two people.

Every Man Jack is in more of a test-and-learn phase when it comes to GEO, said Larabell. But it’s also taking concrete steps to affect search results. The brand has found that AI LLMs tend to favor “genuine, long-tail conversations” on places like Reddit when making recommendations, Larabell shared. So, he explained, “We’re working more closely with our community and integrated marketing team to make sure we can facilitate authentic conversations on these platforms to drive incremental visibility.”

Every Man Jack is also creating content on its website that’s only visible to LLMs to “understand what the models look for and serve in answers,” he said.

Not throwing away SEO

Even as brands prepare for the new dawn of GEO, they’re not totally abandoning SEO.

At Target, the majority of customers are still using “the traditional one-keyword, two-keyword-based search,” Bhosale said. Randerson said the same, saying, “We’re very still rooted in traditional SEO.” Nino agreed that SEO is “still a big channel.”

“Thirty to forty percent [of traffic] comes from SEO,” he said.

But, retail leaders acknowledged, customers aren’t slowing down their use of tools like ChatGPT and Google Gemini. And many are actually changing their behavior, as a result, when they use more traditional search functions on retail web pages. At Target, “We are seeing guests adapting and using longer queries, being more conversational in that particular fashion,” Bhosale said.

Ultimately, brands are trying to get ahead of GEO, but they still have questions about how it all works. And many want further transparency from the engines themselves.

“What is ‘important’ to LLMs?” said Larabell, pointing to the unanswered questions. “What information do we need to provide to them in order to maximize the opportunity for our business, and how do we provide that information?”