Seba BlancoI help independent hotels sell more effectively and operate smarter by combining hotel technology with sales and marketing.
Do not index
Do not index
A few weeks ago I ran a small experiment: I asked an AI to identify hotels in Santiago, Chile with reputation problems.
The list itself wasn't the interesting part. The method was. The answer wasn't built from anything those hotels say about themselves — it was assembled from ratings, review volume, and guest comments cross-referenced across platforms.
That quietly redefines what "online reputation" means for your property. Here's what I learned, and what you can do with it.
The Experiment — and the Reframe That Changed Everything
My first prompt asked for the "worst hotels" in the city. Then I rephrased it: which hotels show the greatest reputational pain? The results shifted, and that shift is the whole lesson.
A low score built on 20 reviews is statistical noise. A decent score built on thousands of reviews — where the same complaint surfaces again and again — is a diagnosis. The AI wasn't averaging stars. It was reading for repetition.
That's the part most hoteliers miss when they obsess over their aggregate rating. The score is a number. The pattern is information. And AI-powered search is very, very good at finding patterns you'd rather it didn't.
Being Found Is Not the Same as Being Recommended
Search intent is changing shape. "Hotels in Santiago" asks for options — a list, a map, twenty tabs. But watch what travelers now type into ChatGPT or Gemini: "I'm in town three days for work. I want somewhere quiet, well located, good breakfast, and no recurring service complaints. Which one?"
That's not a request for options. It's a request for a decision. And to win a decision, showing up isn't enough — the model needs reasons to recommend you.
Where do those reasons live? Your website is one source among many. Around your property sits a huge volume of information you never wrote: reviews, guest photos, videos, forum threads, community posts. Studies of which sources AI engines cite (mostly B2B data, so don't map it one-to-one onto travel) all point the same way — the evidence about your brand is distributed, and most of it is third-party.
Your Reviews Are the Biggest Guest Study You Never Commissioned
Here's the turn that matters most: stop treating reviews as a reputation problem and start treating them as business intelligence. Your review history is the largest, cheapest, permanently running qualitative study your hotel will ever have.
The rule of thumb: one bad review is an anecdote. The same phrase repeated across ten reviews is a signal. A signal that persists for months isn't reputation anymore — it's product.
Try this today: export your last 12 months of reviews across platforms, paste them into Claude or ChatGPT, and ask for the five most repeated complaints and the five most repeated compliments, with example quotes and the point in the guest journey where each occurs. Twenty minutes. Most hotels have never done it.
8 AI Prompts to Analyze Your Hotel's Guest Reviews
Eight copy-paste AI prompts to turn 12 months of guest reviews into patterns, positioning and an operations brief — plus how to export reviews from each platform.
This Stops Being Marketing's Problem
Marketing can build an impeccable promise. Then come reception, the room, housekeeping, breakfast, response speed, and how your team makes people feel. When there's distance between what was promised and what was delivered, guests write that distance down — in public, permanently, in a format AI models synthesize effortlessly.
So when the same complaint survives quarter after quarter, it's no longer a communications issue. It's an operations decision wearing a marketing costume. The review analysis from the previous section belongs in your ops meeting, not just your marketing folder.
The Good News: the Machine Runs in Reverse Too
A hotel that does things well accumulates evidence the same way — and sometimes discovers advantages it never claimed. You think you sell location; your guests keep writing about the warmth of your team. You assume price is the objection; the pattern shows the friction sits somewhere else in the journey entirely.
That's positioning material you didn't have to invent. And it's more credible than anything you could write about yourself, precisely because you didn't write it.
The Question Worth Asking
The tempting question is "how do I show up in AI answers?" — and there's a whole cottage industry ready to reduce that to schema markup and robots.txt tweaks. Resist it.
The useful question is: what version of our brand can be built today from all the evidence that already exists about us? That version probably looks more like how guests experience you than how you describe yourself. Being found is no longer enough. You need to be an option someone has actual reasons to recommend.
Run the review-analysis prompt on your own property this week — then subscribe to the newsletter for more on how AI search is reshaping hotel marketing.
Written by
Seba Blanco
I help independent hotels sell more effectively and operate smarter by combining hotel technology with sales and marketing.
