Insights Driven By Data: What We Can Learn From AI Travel Assistants

People aren’t just using ChatGPT to cheat on their schoolwork or complete arduous tasks at work more efficiently. Travelers — especially international travelers — are increasingly turning to AI chatbots for trip inspiration, selecting destinations, creating itineraries, and booking accommodations. 20.1% of international travelers use AI when choosing a destination according to a Brand USA Market Intelligence Survey. A whopping 98% of Indian travelers want AI tools for planning trips according to a Booking.com Global AI Sentiment Study.

This creates several questions for DMOs and other organizations: Is AI recommending my destination to travelers? How can I identify and respond to consumer preferences when they consult AI instead of humans?

Fortunately, there is a way to use this emerging trend to your company’s advantage: Build your own AI travel agent, integrate it into your website, and let the data guide your marketing strategy.

That’s what Brand USA did. In collaboration with Mindtrip, they revamped the Visit the USA website to allow users to receive AI assistance and recommendations while searching through the website and looking at potential locations to visit. After just a few clicks, someone can build an entire vacation plan with easy steps to follow. 

Leaning into increased AI usage and making it easier for travelers to plan their visit to your destination is certainly beneficial, but you can take it one step further by compiling data from consumer-AI conversations to see what trends are worth pursuing.

Brand USA recently hosted a webinar sharing information they obtained from nearly 20,000 conversations with the Mindtrip AI from international travelers who opted into data collection. They found the following broad-level data:

• The average length for a planned trip was 9.7 days.
• The average party size was 3.0 people.
• The average chat lasted 6.2 back-and-forth messages.
• When trip planning with AI, users were most interested in attractions (64%), hotels (21%), and restaurants (12%).

This is useful for DMOs looking to build their understanding of the international traveler segment, but it’s also rather self-intuitive. It makes sense that people who are travelling farther distances would stay longer, travel with a few people instead of alone, and mostly take interest in attractions because those are the primary draw.

What I found most appealing from Brand USA’s report was the deeper insights that show particular country-level demographic information. Too often, tourism marketing attempts to appeal to the international traveler in general. In reality, understanding the motivations of consumers from different nations (often driven by different cultural values and popular sentiments) is crucial as it allows you to match the features of your destination with the wants and needs of the traveler.

Here are some findings for countries with the highest chat volume:

1. India: 23% of chats related to wine and 28% of chats related to road trips
1. 59% of trip-planning chats related to national parks
2. Brazil: 38% of chats related to film tourism (my favorite form of tourism)
1. 1.6X over-index for theme parks and 1.7X over-index for luxury resorts
3. Mexico: 27% of chats related to winter sports
1. Emphasis on shorter planned trip length at 6.5 days on average
4. Japan: 35% of chats related to the Northern Lights
1. Other interests include cherry blossoms and urban cities
5. Canada: The top country for cross-border road trips
6. Argentina: 16% of budget-conscious chats
7. UK: 3.1X over-index on Route 66 and 2.0X over-index on Nashville and music
8. Australia: Longest planned trip length at 15.8 days on average

These national preferences are crucial for DMO strategizing. But here’s my single most compelling fact: Whereas traditional web analytics would place Brazil outside of the top 10 most in-demand travel markets, Brand USA’s chatbot analysis puts it at number two. This means that correctly incorporating AI into your consumer-facing content can help you catch shifts in traveler attitudes that would have otherwise flown under the radar.

I would also advise that your brand use AI with the utmost transparency as consumer opinions towards it are still mixed. According to an article from Forbes, 82% of survey respondents “view AI data loss-of-control as a serious personal threat.” Always make it clear when users are communicating with AI, and present a clear option of whether they would like their data to be collected and used for reports like the one from Brand USA.

So how should you go about using AI-driven data to inform your marketing campaign? Here are some key takeaways:

• Analyze AI conversations alongside your traditional analytics. Website traffic tells you where visitors came from. AI conversations tell you what they're trying to accomplish. Looking at both together provides a much more complete picture of traveler intent.
• Create market-specific campaigns instead of one-size-fits-all messaging. If Brazilian travelers are asking about filming locations while Indian travelers are searching for national parks and road trips, they shouldn't receive the same advertisements or website content. Use AI insights to tailor creative assets, landing pages, and promotional messaging for each market.
• Look for emerging demand before it appears in your visitor data. AI conversations can reveal interest from markets that may not yet generate significant website traffic. Treat chatbot data as an early-warning system for new opportunities, allowing you to adjust marketing investments before competitors recognize the trend.
• Fill the content gaps travelers are revealing. Pay close attention to the questions your AI cannot answer well or the information visitors repeatedly request. Those gaps should directly influence future website pages, blog posts, FAQs, itineraries, and social media content.
• Be transparent about your use of AI and customer data. Clearly disclose when visitors are interacting with AI, provide an easy way to speak with a person when needed, and only analyze conversations from users who have explicitly opted into data collection. Trust is just as valuable as the insights themselves.