AI visibility
Schema markup every European lodge needs
The structured data that decides AI citation.
Colin HarrisonFounder26 August 2026
Quick answer
Independent European lodges and small hotels need six core schema.org JSON-LD types deployed to be reliably extractable by AI engines. These include:
- LodgingBusiness on the homepage with full address, geo, and amenities
- Room schema per room category
- Offer schema per published rate
- FAQPage on the most-asked questions
- Person schema on any named host or chef bio
- BreadcrumbList on every interior page.
Together, these tell ChatGPT, Gemini, Claude, Perplexity, and any other AI engine what kind of business you are, what's bookable, what it costs, what travellers ask, who runs it, and where each page sits in the site hierarchy. Deployment takes an afternoon, but citation offers years in return.
Key takeaways
- Six core schema.org JSON-LD types cover ~80% of the AI extraction surface for accommodation: LodgingBusiness, Room, Offer, FAQPage, Person, BreadcrumbList.
- FAQPage schema is the single highest-leverage type because it directly mirrors the question-and-answer shape AI engines render in their responses.
- Person schema on named-host or named-chef bios cites measurably better than anonymous "About Us" grids — and most independent operators skip it.
- An afternoon to deploy for a single-property B&B; a day or two for a multi-cabin cluster with a restaurant attached. The compounding return is years.
Why schema is the foundation
Schema.org JSON-LD is the structured-data layer that tells AI engines what your page is about in unambiguous, machine-readable terms. A page without schema asks the model to guess; a page with the right schema hands the model an answer. Schema is one layer of the structural fix for AI visibility — the foundation that llms.txt, citation blocks, and named-author content build on top of.
For independent lodges, the six schema types below cover roughly 80% of the AI extraction surface. Get them deployed correctly and the citation rate from ChatGPT, Gemini, Claude, and Perplexity for buyer-intent regional queries lifts measurably within four to eight weeks of indexing.
The deploy is finite work — an afternoon for a single-property B&B, a day or two for a multi-cabin cluster with a restaurant attached.
1. LodgingBusiness — on the homepage
What does LodgingBusiness schema do for a hotel?
LodgingBusiness schema is the root schema for any accommodation property. It tells AI engines what category of business you are, where you are, and what the headline offer looks like — the single block everything else hangs off.
{
"@context": "https://schema.org",
"@type": "LodgingBusiness",
"name": "Buttermere Crag Cabins",
"description": "Five-cabin off-grid retreat in the western Lake District National Park.",
"url": "https://buttermerecragcabins.co.uk",
"telephone": "+44 1234 567890",
"email": "[email protected]",
"address": {
"@type": "PostalAddress",
"streetAddress": "Crag Lane",
"addressLocality": "Buttermere",
"addressRegion": "Cumbria",
"postalCode": "CA13 9XX",
"addressCountry": "GB"
},
"geo": {
"@type": "GeoCoordinates",
"latitude": 54.5410,
"longitude": -3.2730
},
"priceRange": "£165–£295 per cabin per night",
"amenityFeature": [
{"@type": "LocationFeatureSpecification", "name": "Wood-burning stove", "value": true},
{"@type": "LocationFeatureSpecification", "name": "Outdoor bath", "value": true},
{"@type": "LocationFeatureSpecification", "name": "Dog-friendly", "value": true}
],
"petsAllowed": true,
"starRating": {"@type": "Rating", "ratingValue": "5"}
}The fields that matter most for AI extraction are name, address, geo, priceRange, amenityFeature, petsAllowed. Get those six right, and you're 90% of the way there.
2. Room — one per room category
What does Room schema do for a hotel?
A separate Room (or HotelRoom) entity per category lets AI engines distinguish between the family lodge and the couples' cabin when answering "best dog-friendly cabin in the Lakes for two."
{
"@context": "https://schema.org",
"@type": "HotelRoom",
"name": "Crag View Cabin",
"description": "Two-person cabin with private outdoor bath and uninterrupted fell views.",
"occupancy": {
"@type": "QuantitativeValue",
"maxValue": 2
},
"bed": {
"@type": "BedDetails",
"numberOfBeds": 1,
"typeOfBed": "King"
},
"amenityFeature": [
{"@type": "LocationFeatureSpecification", "name": "Wood-burning stove", "value": true},
{"@type": "LocationFeatureSpecification", "name": "Outdoor bath", "value": true},
{"@type": "LocationFeatureSpecification", "name": "Floor area", "value": "32 m²"}
]
}Deploy one block per room or cabin category, embedded on its own page, not the homepage or a shared “rooms” overview.
3. Offer — per published rate
What does Offer schema do for a hotel?
It's what AI engines pull when a traveller asks about price. Without it, the model is reading marketing prose and guessing at numbers.
{
"@context": "https://schema.org",
"@type": "Offer",
"name": "Crag View Cabin — March to October midweek",
"price": "165",
"priceCurrency": "GBP",
"priceSpecification": {
"@type": "UnitPriceSpecification",
"price": "165",
"priceCurrency": "GBP",
"unitText": "PER NIGHT"
},
"availability": "https://schema.org/InStock",
"validFrom": "2026-03-01",
"validThrough": "2026-10-31"
}Deploy one Offer per distinct rate-period combination. AI engines rely heavily on priceSpecification and validThrough for "what does it cost in [month]" questions.
4. FAQPage — on FAQ and detail pages
What does FAQPage schema do for a hotel?
This is the single highest-leverage schema type for AI citation, because it directly mirrors the question-and-answer shape AI engines render in their responses.
This is the block of code that tells ChatGPT, Claude, Gemini, and Perplexity exactly which question-and-answer pairs to quote back to a traveller, word for word, when they ask something your FAQ already answers. Get this right consistently, and your answer is what shows up in the AI's response.
If you're using a plugin (Rank Math, All in One SEO, or Yoast), you can usually build this by adding an FAQ block to the page and letting the plugin generate the schema for you — no code needed. Rank Math and All in One SEO currently offer the most complete schema support of the major options.
If you're deploying it manually, here's the JSON-LD:
{
"@context": "https://schema.org",
"@type": "FAQPage",
"mainEntity": [
{
"@type": "Question",
"name": "Are dogs allowed at Buttermere Crag Cabins?",
"acceptedAnswer": {
"@type": "Answer",
"text": "Yes. Up to two well-behaved dogs per cabin with a £30 per stay supplement. Towels, a bowl, and treats are provided in each cabin."
}
},
{
"@type": "Question",
"name": "What's the minimum stay?",
"acceptedAnswer": {
"@type": "Answer",
"text": "Two nights midweek, three nights for stays including a Saturday. Single-night stays available November to February only."
}
}
]
}Build the FAQ around the questions travellers actually ask before booking. Twelve to twenty questions cover most lodges. Finding out which FAQs to add can be done using tools like AlsoAsked.
5. Person — on any named-host or chef bio
What does Person schema do for a hotel?
AI engines reward named human authority. A "Meet the Owners" page with a Person schema attached to each owner cites measurably better than an anonymous "About Us" page.
{
"@context": "https://schema.org",
"@type": "Person",
"name": "Alice Whitelaw",
"jobTitle": "Owner and Head Chef",
"worksFor": {
"@type": "LodgingBusiness",
"name": "The Pine House"
},
"description": "Trained at St John in Clerkenwell. Twelve years cooking in restaurants across the UK and France before opening The Pine House in 2021.",
"image": "https://thepinehouse.co.uk/people/alice.jpg"
}This is the schema type most independent operators skip — and it's one of the easiest citation lifts available.
6. BreadcrumbList — on every interior page
What does BreadcrumbList schema do for a hotel?
It tells AI engines where the page sits in the site hierarchy. It's a small signal, but it costs almost nothing to deploy and adds context that compounds across the site.
{
"@context": "https://schema.org",
"@type": "BreadcrumbList",
"itemListElement": [
{"@type": "ListItem", "position": 1, "name": "Home", "item": "https://example.com/"},
{"@type": "ListItem", "position": 2, "name": "Cabins", "item":
"https://example.com/cabins/"},
{"@type": "ListItem", "position": 3, "name": "Crag View Cabin", "item":
"https://example.com/cabins/crag-view/"}
]
}Deploy on every interior page only; skip the homepage.
FAQs
Do I need a developer to deploy schema markup?
Sometimes. If your site runs on WordPress with a modern SEO plugin (Rank Math, Yoast, or SEOPress), you can configure most of these schema types through the admin without touching code. If you're on Squarespace, Wix, or a custom-built static site, you'll need to either drop the JSON-LD into the page head manually or get a developer to add it once. The work is bounded — three to six hours for a typical lodge once the property data is gathered.
Will Google Search Console show whether the schema is working?
Yes, partially. Google's Rich Results Test tells you whether the JSON-LD is valid and whether Google can read it. It does not tell you whether AI engines (ChatGPT, Claude, Gemini, Perplexity) are extracting from it — those engines have separate crawlers and their own opaque internal weighting. Validate first with the Google tool; then test extraction by running buyer-intent prompts in the AI engines themselves and watching what they cite.
What's the difference between LodgingBusiness and Hotel schema?
Hotel is a subtype of LodgingBusiness. Use Hotel only if you operate a property with multiple rooms under one roof, daily housekeeping, and front-desk operations. For a self-catering cabin cluster, a glamping site, a B&B, or a guest-house, use LodgingBusiness directly — it's the broader, more accurate category. The wrong subtype isn't catastrophic, but the right subtype is more extractable.
Should I add schema to my blog posts and landing pages too?
Yes — Article schema for blog posts (with Person schema attached for the author), FAQPage on landing pages with question-and-answer sections, and TouristTrip or TouristAttraction schema on itinerary or destination pages. Each schema type tells AI engines what kind of content this page is. The lift comes from coverage breadth, not from over-engineering any single page.
Does schema markup expire or need maintenance?
It needs maintenance when the underlying data changes — a rate increase, a new room category, a chef changeover, a new amenity. We recommend a quarterly schema audit alongside the rate review. The validation tools (Google's Rich Results Test, Schema.org Validator) take five minutes per page and catch any drift between what's in your schema and what's on the live page.
A note on engine differences
Not every AI engine weighs this data the same way. ChatGPT leans heavily on broad web sources like Wikipedia and Reddit; Gemini and Google AI Overviews pull from Google's own index and YouTube; Perplexity favours established news publishers; Claude tends to reward long-form, named-author editorial alongside structured data.
Deploying the six schema types here gives every engine the same clean signal to work from, but don't expect identical citation behaviour across platforms — it's worth checking how your pages are actually cited on each one, not just whether the markup validates.
Validation and testing
Two tools that should be in every operator's bookmarks:
- Google Rich Results Test (search.google.com/test/rich-results) — paste a URL or block of JSON-LD; Google reports validity and shows what would render in search.
- Schema.org Validator (validator.schema.org) — broader validation against the full schema.org spec, useful for the schema types Google doesn't surface in its rich-results coverage.
Test every page after you’ve deployed the schema, re-test after any content update, and schedule a quarterly audit to catch drift.
The whole stack — six schema types, deployed once, validated once — is the structural foundation every European lodge, hotel, B&B, or tour operator needs to be AI-extractable. It is not the whole job — the citation block work, the GBP optimisation, the third-party editorial citations all matter too — but without the schema layer underneath, the rest doesn't compound.
