AI visibility is table stakes for travel brands now, and if you’re not there yet, here’s where to start.
But let’s say you’re there. Someone finds your brand through one of those engines and goes looking for more. What happens next?
What we’re seeing is that most brands haven’t really built for that moment yet. The media plan, the targeting, the measurement: all of it was designed for a consumer who started with Google. Here’s what it actually takes to build for the one who’s arriving through AI:
The journey has more entry points than your media plan accounts for
Travel search is no longer a single channel. Someone planning a trip doesn’t start with Google, and work down a list of results anymore. They ask an AI chatbot for an itinerary, watch trip content on YouTube, check Reddit for honest takes on specific hotels, and then come back to search when they’re ready to book.
The problem is that most campaigns weren’t built for a journey that spans this many touch points. They run independently, with no visibility into what the other channels are doing, so no single campaign knows enough about a given user to respond to them in the right way.
What the demand capture layer actually needs to look like
So what does it actually look like to build campaigns that can see the full journey? It starts with understanding that not every AI-referred user is in the same place when they arrive.
They’re either still researching, going to other channels to validate what the AI told them, or they’re heading straight to a branded search query to find you directly.
Each path need a different response than your standard acquisition campaign, and the setup for each is distinct.
For users still in research mode, the channels that matter are the ones where validation happens: programmatic media across the open web, paid social on the platforms where travel content actually lives, and Reddit where there’s meaningful conversation volume around your brand. The goal at this stage isn’t conversion. It’s presence and engagement. Showing up consistently enough that the positive first impression from the AI referral doesn’t get eroded by a competitor who happens to be everywhere you’re not. And taking that next step with a user by providing them meaningful information as they look to validate what they’ve already read.
For users heading into search, the opportunity is in how you tag and segment them. When someone clicks through from an AI engine to your site, that referral source is trackable. You can use it to build an audience segment, then feed that signal back into your paid search campaigns so those users get different bids, different landing pages, and different messaging than someone arriving cold. They’ve already been told your brand is worth considering. The campaign should know that.
The piece that ties both together is making sure those users are recognized as the same person across touch points. Someone who came in via an AI referral, saw a YouTube pre-roll ad, and then searched your brand directly is one user at the end of a short journey. Without the right audience architecture in place, your campaigns treat them as three unrelated strangers and bid accordingly.
Why measurement breaks when teams are siloed
Here’s where the structure problem gets expensive: if you’re not measuring this as a connected system, you’ll never see where it’s breaking.
The standard setup for most travel brands has paid search, paid social, and programmatic each reporting back to their own platform numbers. Google says it drove X bookings. Meta says it drove Y. The programmatic DSP has its own figure. Add them up and you’ll get a number that’s two or three times your actual bookings, because all three platforms are claiming credit for the same user.
That’s not a new problem, but it becomes a bigger one when the AI discovery moment is happening upstream of all of them. None of those platforms can see the ChatGPT or AI Overview referral. So the user who was recommended your brand by an AI engine, searched for you, and booked gets attributed entirely to paid search. The role of every other channel in that journey is invisible.
The fix isn’t switching attribution models. It’s building a measurement framework that’s designed to see across the full path: platform attribution for in-flight optimization, marketing mix modeling for the long-term view of what’s actually driving revenue, and incrementality testing to separate what’s genuinely driving bookings from what’s just claiming credit.
The incrementality piece matters more in travel than most categories because the baseline intent is so high. People who were going to book a vacation were going to book a vacation. The real question paid media needs to answer isn’t whether a campaign reached someone who booked. It’s whether the campaign caused a booking with your brand that wouldn’t have happened otherwise. Running geo-based tests, where you activate paid media in some markets and hold it out in comparable ones, tends to produce more sobering results than brands expect. A meaningful share of what gets credited to paid media turns out to be demand that was always going to convert. And critically, the right response to that isn’t pulling back spend. It’s moving it toward the touch points that are genuinely creating new demand rather than harvesting intent that was already there.
The system has to be designed together
Investing in AI visibility is the right move. But routing that demand into a paid media setup that wasn’t built to receive it, is the travel equivalent of running a brilliant brand campaign and sending people to a broken booking engine. The awareness did its job, but everything else downstream didn’t.
Holding and growing share in travel over the next few years means building the whole system together: AI visibility feeding into a paid media architecture designed for multi-entry-point demand, measured by a framework that can actually distinguish what’s actually driving incremental bookings..
That requires the strategy, the channel execution, and the measurement to be owned in one place. Not coordinated across multiple agencies. Not aligned in a quarterly review. Designed together from the start, with a shared view of where users are coming from and what’s moving them to book.
