What Hoteliers Need to Know About the New AI-Driven Customer Journey

Are you paying attention to Reddit and YouTube? Is your hotel or brand showing up there in a meaningful way? According to Jonny Brentwood, Global President of Data & Analytics at Golin, those are among the sources shaping what generative AI tells consumers.

These were just a few of the surprising reveals at a talk at IMM UK TravMedia Summit in London earlier this week, where Brentwood delivered a presentation on generative AI that felt like a wake-up call. As Global President of Data & Analytics at Golin, a global communications agency within Interpublic Group, Brentwood was drawing on broad, ongoing analysis of how Gen AI pulls, ranks and cites information across platforms for a range of travel clients.

His core message: the way people search, compare and make decisions online is changing in real time, and hospitality businesses need to adapt now.

This session struck me as valuable not just for PR teams and media, but for anyone looking to better understand the customer journey in the age of AI. If customers are increasingly turning to AI platforms for recommendations, comparisons and answers, then the people responsible for operations and revenue need to understand how those answers are being shaped.

In other words, this is no longer just a marketing story. It is a commercial shift. And increasingly, it means brands are no longer just marketing to people. They are marketing to the machines shaping what people see first.

AI is becoming a new front door to hotel discovery

For years, Google was the front door to the internet. According to Brentwood, those days are long gone.

Instead of searching across multiple websites, users are now asking generative AI tools to do the work for them. They want quick answers, curated recommendations and side-by-side comparisons, then usually act on those results without going much further.

For hospitality, that changes the rules. Potential guests no longer begin with a brand website, an OTA, a media article or even a trusted travel advisor. They begin with an AI tool that summarizes the market for them in seconds.

Discovery is no longer just about ranking in search. It is about being present in the sources AI systems rely on.

The customer journey is getting shorter

Traditionally, booking a hotel involved multiple steps. Travelers searched online, read reviews, compared brands, opened several tabs, checked prices and often asked friends before narrowing down their options.

Brentwood’s point was that generative AI is collapsing that process.

People are now using AI not just for inspiration, but for evaluation. They are asking for the best hotels for families, the smartest options for a bleisure trip, the most romantic adults-only resorts or the best luxury stays for food lovers. Then they receive a filtered answer that already feels authoritative.

If customers are outsourcing more of their decision-making to AI, the question is simple: what is shaping the answer they get?

Authority now comes from different places

One of Brentwood’s strongest points was that the sources influencing AI are not always the obvious prestige outlets.

Traditional media still matters. But AI also pulls heavily from practical, accessible, question-led content: comparisons, reviews, niche articles, expert commentary, FAQs and list-based pieces that directly answer what users are asking.

That challenges how many brands still think about visibility. In an AI environment, relevance matters more than prestige.

It is not just about where your brand appears. It is about whether the content around your brand clearly answers the questions potential guests are asking machines.

Owned content matters in a new way

Brentwood was blunt about websites. A hotel website can look polished, cinematic and visually impressive to a person and still be extremely hard for AI to read. JavaScript-heavy websites become almost invisible to generative AI. And if key information sits mainly in images or moving visuals, AI cannot read that properly either.

For an industry that loves beautiful design, that is a serious problem. Pretty websites become nearly impossible for machines to find, interpret or cite.

If your property website is hard to read, if important information is buried in image-led content, or if your structure and metadata are weak, you are making it harder for AI to surface your brand.

Unlike media coverage, your own website is something you actually control.

Content needs to answer real questions

Brentwood was equally direct on content. Generic brand copy is not enough.

He said generative AI favors formats that help people make decisions. That includes listicles, comparison pieces and opinion-led content that answers questions like: should I choose A, B or C? What are the pros and cons of each? What is good, bad or different about these options?

He also pointed to FAQs, how-to guides, itineraries and expert-led content as strong formats for AI visibility.

He also made a surprisingly strong case for press releases. Not because journalists are suddenly hanging on every word, but because releases syndicated across hundreds or thousands of sites create machine-readable distribution at scale. In Brentwood’s view, that broad pickup makes them valuable for Gen AI, especially when they are structured clearly and written to answer real questions.

For hotels, that means producing content that matches the language and intent of what consumers are actually asking. Which hotel in London is best for a business trip with extra leisure time? What is the best family-friendly luxury resort in the Maldives? Where should I stay in Paris for a first visit focused on culture and walkability?

The more directly a hotel answers those questions in clear, useful language, the more likely it is to be picked up. This is not just a content exercise. It is a visibility strategy.

YouTube, Reddit and other AI-friendly sources deserve more attention

Brentwood also highlighted YouTube and Reddit as two platforms that matter far more in an AI environment than many hotel teams realize.

His point on YouTube was specific: AI does not understand the video itself, so brands need to paste the full transcript into the description. Not auto captions. The actual transcript.

That is a practical fix for any hotel brand working with creators, publishing destination content or investing in video as part of its marketing mix.

Reddit matters for the opposite reason. It is text-rich, question-led and full of firsthand opinions, comparisons and recommendations. In other words, it is exactly the kind of content AI can read and use.

He also pointed to sources such as Wikipedia, Forbes, BuzzFeed and The Points Guy as examples of the kinds of accessible, text-led content Gen AI is more likely to pull from. By contrast, he argued that prestige outlets behind paywalls may still matter for human audiences, but often have less influence on what machines can actually access and cite.

Brands need to understand where AI is actually pulling from, not where they assume it should be pulling from.

Freshness matters

Another key point was recency.

AI platforms favor newer content. Hotels cannot update their websites once a year, publish a few campaign-led stories and assume that is enough.

They need a steady flow of current, useful information: refreshed landing pages, updated destination content, timely blog posts, relevant expert input and content that reflects changing traveler needs.

Freshness is not just digital housekeeping. It affects visibility, recommendation and booking consideration.

This cannot sit in one department

The AI shift in discovery is bigger than PR. Bigger than digital. Bigger than marketing.

PR cannot solve it by itself. Marketing cannot solve it by itself. Sales cannot ignore it. Revenue leaders need to care because it affects demand. General managers need to care because it affects positioning and competitiveness.

AI visibility sits at the intersection of reputation, content, discoverability and commercial performance.

For hotels, that means tighter alignment between communications, marketing, digital, sales and commercial strategy.

Jonny's list of 10 things you can do now to improve AI discovery for your hotel or brand

The takeaway: start now

Brentwood did not present this as a future trend. He presented it as a live shift already underway, backed by the research and data his team tracks every day.

And he did not mince words about what brands should do next. Start.

Review your website. Audit your content. Fix your YouTube descriptions. Rethink your press releases. Build content around the questions your customers are actually asking. Understand which sources are shaping AI answers in your category.

The old playbook is no longer enough. Hotels still need to market to people, of course. But as Brentwood made clear, they also need to market to the machines increasingly shaping what people see, compare and trust first.

Because increasingly, that is where the customer journey begins.