As AI assistants like OpenAI’s ChatGPT, Google Gemini and Microsoft Copilot increasingly mediate the customer experience, brands must rebuild their mobile apps and websites to support their use, experts told CX Dive.
"I think in a year or two — three years, certainly — this whole concept of having an app that you tap and swipe or type stuff into is going to be completely gone,” Tobias Dengel, president of Telus Digital Solutions, said. “You're just going to want to use your voice.”
The technology is quickly reshaping the customer journey, forcing brands to rethink their digital strategies. AI assistants are the fastest-growing distribution channel, with AI search visitors expected to surpass traditional search by 2028, according to a Semrush study. As of February, nearly two-thirds of websites receive AI traffic, according to an Ahrefs study.
Those trends are accelerating, as tech companies add AI assistants to just about everything, including mobile apps, web browsers and operating systems, and brands like Expedia, Spotify and Zillow begin building apps inside of ChatGPT.
As a result, brands must make their mobile apps and websites discoverable and usable by AI systems — a process called agentic engine optimization, answer engine optimization or generative engine optimization. Though similar to search engine optimization and marketing, it requires additional measures and even greater attention to detail.
The change is similar to other technological developments businesses have had to adapt to.
“Companies had to think about AOL's walled garden, then search engines, then Google's dominance,” Dengel said. “We're in the early, fragmented stage now, and brands need to be experimenting across platforms.”
AI workarounds
Brands have built their mobile apps and websites for an internet based on Google’s search algorithm rather than AI, with many focusing on the digital properties they own and control. But that needs to change as consumer behavior shifts.
“The biggest mistake I see is people putting the blinders on,” Jason Maynard, CTO of North America and Asia/Pacific at Zendesk, told CX Dive. “Increasingly, you have to think about the ecosystem that sits around your digital experience, and more and more of that is going to happen on a platform that you don't control."
Today’s mobile apps and websites are primarily designed to be discoverable through traditional search and to be used by humans. However, brands will need to make their digital properties easier for machines to read as more people rely on AI assistants to find information and take action on their behalf. For instance, a consumer could ask ChatGPT to book a flight for them.
AI agents currently use resource-intensive methods, such as taking screenshots and identifying X and Y coordinates, to navigate websites built for humans. It can be slow, incomplete and error-prone, worsening the customer experience.
In addition, brands could suffer if they fail to make their websites easy for AI to discover, just like they become invisible if they’re not on the first page of traditional Google search results.
"If you're not the top three on Gemini or Claude or ChatGPT, you might see a huge fall-off [in traffic]. So the stakes are going to get really high,” Dengel said.
Making your brand discoverable
Optimizing for AI, however, is more complex than SEO because the technology’s outputs aren’t as predictable or keyword-dependent as traditional search.
For many brands, the initial goal is figuring out how to use AI-mediated search to steer customers into their own "proprietary channels,” such as mobile apps and websites, according to Unnati Narang, assistant professor of business administration at the University of Illinois' Gies College of Business.
In addition to making brands more discoverable by AI systems, agentic agent optimization can improve brand consistency across platforms.
“That'll reinforce the value brands already have created, irrespective of where they're delivering,” Narang said.
However, the extent to which brands will want to integrate third-party AI assistants into their digital customer experience will vary based on needs and capabilities.
"The smart brands are going to be playing both sides,” Dengel said. “They're going to say, 'I need to be super engaged with where the searches are happening, but when I can, I'd like my consumers to come straight to me.'"
Businesses should start by investing in “digital hygiene” to make their websites easier for AI agents to crawl, Maynard said. This may include:
- Structuring websites to make it easy for AI agents to understand the hierarchy and content of a website, including headings, paragraphs, images, links and forms.
- Ensuring HTML tags are correct.
- Providing high-quality metadata, including clear labels and descriptive alt text.
- Vectorizing images to ensure they display correctly across platforms.
Brands should also create plain text files to control how AI models and agents interact with their web content. These AI.txt files are similar to the Robots.txt files companies use to tell search engines which pages or sections they can or cannot access.
In addition, semantic HTML can help brands provide AI systems with detailed information about the content on their websites. Websites with open application programming interfaces, or APIs, take things a step further, as APIs are entirely machine-readable and unencumbered by human-machine interfaces like traditional websites.
"You can kind of direct the AI to the conclusions that you want within your website, from your data and your information,” Dengel said.
While AI assistants are changing consumer behavior, brands don’t need to stand idly by. There are things they can do to maximize their chances of success in a post-search world.
“But at the end of the day, consumer preference will drive where this goes,” Maynard said. "We're probably going to be in a messy middle for a long time because we don't have the infrastructure. We're just going to use all these human interfaces.”
