What 1,000 answers taught me

As travelers increasingly turn to AI instead of opening dozens of webpages, discovery depends on whether a brand can be found, understood, and recommended. In this post, I share our perspective on GEO and what it means for travel and destination leaders.

How AI expands travel intent.

A few months before Google began rolling out AI Mode in the U.S. in May 2025, we conducted a quick analysis of recommendations from ChatGPT, Gemini, and Perplexity for travel planning.

We analyzed more than 1,000 responses across a random mix of small, medium, and large destinations in the Southeastern U.S. Asking questions like "Where should I go?" and "What should I do there?"

The answers mentioned destination websites, well-known attractions, and places to visit. But destination sites appeared less often than I expected. What surprised me was how often travel aggregators, community and creator content, and media outlets appeared. Generative engines were pulling context from many digital spaces where travelers show up: Reddit, YouTube, creators, social, and earned media coverage. A destination’s webpage was one among many.

Consumer behavior has also shifted. Travelers can now interact with AI instead of opening 20–30 pages when planning a trip. It is just more practical. In the first four months of 2026, 68% of U.S. Google searches ended without a click, up from 45% a decade ago.

What travelers told us.

Last fall, I interviewed travelers who were already comfortable using AI. The only caveat is that an AI agent conducted the interviews, so the participants were more likely to be early adopters. They described planning as easier and less stressful. AI helped them find cheaper options and confirm, reshape, and validate their decisions. “Exciting,” “fun,” and “cool” were among the words they used to describe their experience.

They also expected AI to evolve from suggesting trips toward booking them, even as they remained aware of errors in its advice and the need to verify.

Recent research points in the same direction. AI is becoming another entry point into discovery and travel planning. Skift’s 2025 U.S. Travel Tracker found that more than half had used an AI tool for trip planning. Further, McKinsey found that fewer than one-third of travelers had used generative AI for travel-related tasks. Of those, 84% said it improved their experience. IMG's 2026 Travel Outlook Survey showed that one-third of respondents were likely to use AI for travel planning. Among that group, 75% turned to it for recommendations, 70% for itineraries, and 69% for discovering new ideas, while only 13% used it for booking. Expedia raised the AI trust issue, with 66% not trusting it to buy or book on their behalf.

AI discovery introduces a new alphabet.

If you feel buried under new acronyms, it helps to understand how marketers’ roles are expanding beyond SEO and SEM to include GEO and GEM. Some practitioners call it AEO, answer engine optimization. I stick with GEO and really like how Michael Pettiette and Kimberly Whitler describe them:

GEOGenerative engine optimization: The process of optimizing content to boost its visibility in responses from AI-driven search platforms.
GEMGenerative engine marketing: A paid promotional process aimed at getting a brand to appear in responses from AI-driven search platforms.

In January 2026, OpenAI announced plans to test advertising inside ChatGPT. By mid-year, a beta self-serve ads manager was live. Expedia, Marriott, Airbnb, Booking.com, and Royal Caribbean were the earliest travel advertisers to be reported. Google is pushing its own paid with AI Max, an optimization layer within existing Search campaigns. It brings AI-powered reach, targeting, creative tailoring, and reporting.

GEM is still developing. The trap is treating these as either-or options. Marketers have more tools in their toolbox, which turns into a resource allocation problem in a world of static budgets.

Three issues deserve attention as we gain a better understanding of these systems.

Breadth is how context gets built.

LLMs are trained on publicly available, licensed, synthetic, and human-provided information. AI search systems may also retrieve, parse, extract, and represent web content when responding to a question. Consistent information across credible sources can strengthen the context available about a brand.

A destination represented across many communities has a broader presence than one that relies mainly on its website. Think travel aggregators, forums, creators, magazines, YouTube, social media and earned media coverage, and so on. These systems don’t read a webpage as a traveler does. They interpret repetition, relevance, authority, and relationships across many sources. Their answers may reflect information and connections found across many channels and communities.

Structured and well-distributed content also matters for entering AI answers. Breadth across trusted sources can influence whether a brand appears in the narrative.

A recent Notified study illustrates how quickly distributed content can enter AI answers. Their analysis of a sample of 8,000 GlobeNewswire press releases revealed that either ChatGPT or Claude cited 99.3% of the press releases, with an average time to first citation of eight hours. While it is a vendor study, the speed at which clear, well-distributed, new information can become part of the AI narrative is eye-opening.

Traffic is an input, not an outcome.

I have always viewed impressions, volume, and click data with caution. When presented, people get fixated on them. These measures tell us something about exposure and on-site behavior. But they capture only a narrow slice of influence.

As website sessions and clicks start to shrink in a zero-click world, instead of treating those data points as KPIs, ask better questions.

What kind of content holds their attention? Are the right travelers reaching the site? What does their behavior reveal about where they are in the decision process?

Adobe’s findings add another dimension. They found that AI-referred visitors were more engaged and stayed on travel sites longer than visitors from non-AI sources. These visitors may have specific questions or may be further along in their research.

Traffic still matters. Its role is to inform the analysis, rather than serve as the final measure of success.

Early learning has value.

GEM is still novel. Understanding how it works will take time. But the early period is when competition is thin, pricing is low, and learning can be less expensive.

WPP Media estimates that generative search could jump from $5.1 billion in 2026 to $100 billion in annual ad revenue by 2030, becoming the fastest-growing channel ever. Travel and tourism marketers may need to revisit their marketing mix allocation as they plan for 2027 and 2028.

So what can you do today?

Understand your audience.

What are travelers asking in their own words before they reach your brand? How do they find value in your brand? What problems are they trying to solve? Then decide how your brand should appear throughout the travel planning journey.

Make your brand clear wherever it appears.

Build a differentiated and trusted story, then carry it consistently across the wider ecosystem shaping traveler decisions and AI answers.

Can machines reach you?

Check which priority pages search engines and AI crawlers can access and retrieve. Remove duplication, keep information current, and make each page’s purpose clear.

Structure matters.

LLMs rely on data, and your website content is part of that data. Clearly identify authors, organizations, places, and FAQs. Use appropriate schema to clarify entities and how they relate to one another.

Make it cross-functional.

Bring people together from different departments from day one, not just digital and IT.

Measure for insight.

Track AI visibility across major AI platforms. Share of answers, competitive presence, citation coverage, and AI sentiment are starting points. Smaller organizations can use AI to identify thin, outdated, or repetitive pages on their site. The most practical question is how AI systems understand your narrative, and what they get right and wrong about your brand.

Breadth is becoming a strategic asset. Manage AI visibility as one of your channels.

*********

How AI is shaping travel discovery.