Advertising can be a Sisyphean task. Just when a problem is nearly solved, a new marketing platform, optimization method or three-letter acronym shows up and brings you back to square one.
The latest hill for advertisers to conquer is understanding generative engine optimization (GEO), which is the art and science behind how marketers influence the way their products show up in AI search engine responses.
Evertune, one such GEO tech startup, announced a feature on Wednesday that lets advertisers run programmatic ad campaigns directly on the sites and pages most often cited by AI chatbots.
The capability includes partnerships with The Trade Desk (where, notably, much of the Evertune team used to work) on the buy side and Index Exchange to sell the ads.
Waiting for results
The primary focus of the emerging GEO category is how to influence the way AI models cite a brand’s product, such as by generating earned media or providing better data feeds to the LLM, said Evertune CEO Brian Stempeck.
But Evertune’s new feature is focused on reaching a consumer after they’ve seen the generative search response and clicked through to a site, Stempeck said.
That might mean doubling down on the same message conveyed by the chatbot, like a running shoe company mentioning within the ad how highly ChatGPT ranks its product. Or, vice versa, the ad might attempt to assuage a concern raised by the chatbot, like a shoe company referring to its “new and improved arches” in response to issues with arch support cited by LLMs.
The feature offers a leg up to brands who aren’t satisfied with how they’re showing up in AI models: Now, they can determine the top pages where the AI chatbot pulls its recommendations for a given prompt (for instance, a question about best coffee machines might source to a BuzzFeed roundup) and advertise directly on that page.
Right place, right time
Advertisers have only recently begun to “take GEO really seriously,” said Eraaj Selvadurai, head of organic performance at agency OMD EMEA, an Evertune customer. Now, he said, advertisers are seeing about a 12% click-through rate from the links cited by AI chatbots.
Being able to advertise directly on those pages “at the right moment” (that is, when the user is already considering a purchase and weighing the pros and cons of various products) is crucial, said Selvadurai.
In both programmatic solutions, like Evertune’s and LLMs’ own ads businesses, the challenge for brands is to make their ads compelling enough to influence consumers on the page, especially considering their ad might run counter to the chatbot’s recommendation.
After all, many of these brands choose to run ads because they’re dissatisfied with how they appear in generative AI searches.
The to-do list
But an AI bot’s suggestion isn’t the end of the shopping journey.
For a CPG purchase, for instance, Stempeck said shoppers are more likely to take the AI recommendation at face value and go straight to purchase. But for high-consideration categories, which could be anything from banking to insurance offerings, “that’s where we see the click-out behavior.”
AI chat “condenses [the] research space,” said Selvadurai, which means sharper and newer brands can “start the conquest against some of the slower-moving established brands” by outmaneuvering incumbents in generative search tactics.
For instance, AI chatbots can sometimes identify a fault from customer reviews, social posts or forums like Reddit and thus discourage use of certain products or companies.
If an AI chatbot is consistently citing a negative feature of a product, Stempeck said, the brand should approach that with the urgency it might if influencers or bad press made the same remark, thus dampening sales.
Advertising on the page cited by AI chatbots – and likely accessed by the consumer for further research – is “the perfect time” to emphasize a message to potential customers, Stempeck said. But, he added, the ad has “got to be specific” about reinforcing strengths or remedying shortcomings that were addressed by the LLM.
