How ChatGPT and AI assistants choose hotels
By the hoteldata.ai team · Updated
Travellers now ask ChatGPT, Gemini, Google AI Mode and Perplexity for hotel ideas. None of these companies publishes a ranking formula for hotels, but they do document how their assistants search the web, which crawlers they use and what site owners can control. This guide sums up what is public, with links, and what it means for an independent hotel.
The short answer
When an assistant answers a hotel question with web search turned on, it searches the web, reads pages its crawlers are allowed to fetch, and writes an answer with links to its sources. ChatGPT, for example, rewrites your question into one or more targeted searches sent to search providers and may use your approximate location for local results[1]. OpenAI says it ranks results "using multiple factors intended to help users find relevant, reliable information" and that "placement is not guaranteed"[1].
So there is no switch a hotel can flip. What a hotel can do is make sure its own website is reachable by these crawlers and states its key facts clearly, so that when an assistant does look, it finds accurate information.
What each company says publicly
ChatGPT (OpenAI). To be eligible to appear in ChatGPT search results, OpenAI says a site should allow the OAI-SearchBot crawler and make sure its host or CDN does not block OpenAI's published IP addresses[1]. OpenAI runs separate crawlers for separate purposes: OAI-SearchBot for search, GPTBot for training data, and ChatGPT-User for pages a user asks ChatGPT to visit. Each can be allowed or blocked on its own in robots.txt; for example, you can allow search while blocking training[2].
Google AI Overviews and AI Mode. Google says the normal SEO best practices apply and that there are "no additional requirements" to appear: a page must be indexed and eligible to show in Search with a snippet[3]. Both features may use "query fan-out", running several related searches to build an answer[3]. Google's list of fundamentals includes allowing crawling in robots.txt and at the CDN, making important content available as text, making sure structured data matches the visible text, and keeping your Business Profile up to date[3]. Google also states that you don't need to create new machine-readable files, AI text files or special schema.org markup for these features[3].
Gemini. Google documents a separate robots.txt token, Google-Extended, that controls whether content Google crawls may be used to train future Gemini models and for grounding in Gemini Apps and Vertex AI. It does not affect Google Search[4].
Perplexity. Perplexity says PerplexityBot is used to surface and link websites in its search results (not to train AI models) and recommends allowing it in robots.txt and in your firewall if you want to appear[5].
What this means for your hotel: a practical checklist
- Let the search crawlers in. Check that robots.txt, your hosting and any firewall or bot protection allow Googlebot,
OAI-SearchBotandPerplexityBot[2][5]. Decide separately whether you want to allow training crawlers such asGPTBotandGoogle-Extended[2][4]. - Put your key facts in plain text on your own website. Location and how to get there, room types, check-in and check-out times, pet, parking and cancellation policies, amenities, languages spoken and answers to common questions. Facts that only exist inside a booking engine, an image or a PDF are harder for any crawler to read. Google lists "making sure that important content is available in textual form" among its fundamentals[3].
- Add Hotel structured data. Schema.org has a Hotel type with properties such as
checkinTime,checkoutTime,amenityFeature,petsAllowedandstarRating[7]. Google describes structured data as a way of "providing explicit clues about the meaning of a page"[6]. Keep it consistent with the text visitors see[3]. - Keep your listings consistent. Google mentions keeping your Business Profile up to date[3]. The same name, address, phone and facts everywhere make it less likely that an assistant repeats an outdated detail.
- Optional: publish an llms.txt. llms.txt is a proposal for a short markdown summary at
/llms.txtthat AI agents can read[8]. Google says it isn't needed for its AI features[3], so treat it as a low-cost extra after the basics. Our llms.txt guide for hotels explains how. - Ask the questions your guests ask. Try "best boutique hotel near [landmark] in [city]" or "hotel in [city] that allows dogs" in a few assistants, more than once. Answers change from run to run, so look for patterns rather than a single result.
What nobody outside these companies knows
None of these companies publishes how much weight they give to reviews, prices, online travel agencies or a hotel's own website when they recommend a hotel. Be careful with any service, including ours, that promises a guaranteed AI recommendation: OpenAI itself says placement is not guaranteed[1]. What you can control is whether your facts are reachable, readable and consistent.
How hoteldata.ai helps
The free scan checks your website the way AI crawlers read it (raw HTML, structured data, robots.txt and llms.txt) and gives you an AI readiness score with recommendations. On Pro, you get Hotel JSON-LD and an llms.txt built from your profile, an install check that confirms they are live on your website, and monthly AI answer checks. Those checks are simulated traveller questions answered by Gemini with live web search, not ChatGPT or Perplexity directly, and we label them that way.
Sources
- [1]OpenAI Help Center: Searching the web with ChatGPT
- [2]OpenAI: Overview of OpenAI crawlers
- [3]Google Search Central: AI features and your website
- [4]Google Search Central: Google's common crawlers (Google-Extended)
- [5]Perplexity: Perplexity crawlers
- [6]Google Search Central: Introduction to structured data markup in Google Search
- [7]Schema.org: Hotel (type)
- [8]llmstxt.org: The /llms.txt file