{"@context":"https://schema.org","@type":"BlogPosting","headline":"Google AI Mode Hotel Study 2026: Where Do Hotel Links Actually Go?","description":"4,000 AI Mode queries reveal: 79% of hotel links go to Google Business Profiles, only 3.6% to OTAs.","datePublished":"2026-02-03","dateModified":"2026-02-03","url":"https://nicolassitter.com/research/google-ai-mode-hotel-study-2026","category":"research","keywords":["Google AI Mode hotels","Google AI Mode clicks","hotel search AI","GBP hotel visibility"],"articleSection":"Research","wordCount":5100,"readTime":"21 min","articleBody":"February 2026Google AI Mode\n\n# Google AI Mode Hotel Study 2026:Where Do Hotel Links Actually Go?\n\n**TL;DR:** In a study of 4,000 Google AI Mode queries across 8 cities, we found that **79% of hotel links stay within Google** via Business Profiles, while OTAs receive just **3.6% of clicks** despite accounting for 46.6% of reference citations. The Local Pack appears in only 6.5% of queries but sends 81% of its traffic directly to hotel websites. For hotels, this means Google Business Profile optimization is critical for AI Mode visibility, while OTA presence mainly influences which hotels get mentioned — not where users actually book.\n\nNS\n\nNicolas Sitter\n\nAI Search Researcher · Published February 3, 2026\n\n4,000\n\nQueries\n\n84,329\n\nSources\n\n6,130\n\nHotel Links\n\n1,146\n\nUnique Hotels\n\n[Get Your Hotel's AI Visibility Report](/contact)[Read the Report](#hotel-links)\n\n[Summary](#executive-summary)[1\\. Hotel Links](#hotel-links)[2\\. Local Pack](#local-pack)[3\\. Sources](#sources)[4\\. Hotels](#hotels)[Methodology](#methodology)[FAQ](#faq)\n\n## Executive Summary\n\nGoogle keeps the click. Hotels still win.\n\n79% of hotel links stay within Google (via Google Business Profiles). Hotels get 16.3% direct — but since most GBP listings link back to the hotel, the effective direct access rate is ~93%.\n\nOnly 3.6% go to OTAs.\n\nYet OTAs dominate source citations — the small source links AI Mode uses. Intermediaries account for 46.6% of all citations. This means AI Mode _learns from OTAs_ but _sends traffic to Google and hotels_.\n\nLocal Pack appears in only 6.5% of queries but when it does, 81% of its links go directly to hotel websites — making it the most hotel-friendly surface.\n\nSection 1\n\n## Where Do Hotel Links Go?\n\nOf 6,130 hotel mentions users can interact with in AI Mode responses:\n\n**GBP = Google Business Profile** — the hotel's listing on Google Maps/Search showing address, reviews, photos, and booking links. When users click a hotel link in AI Mode, 79% land on this Google-owned page first.\n\n~93%\n\nEffective Direct Access Rate\n\nHotel website direct (16.3%) + GBP listings that link to the hotel (~97% of 79.1%) = **~93%** of hotel links ultimately reach the hotel.\n\n79.1%\n\nvia GBP\n\n16.3%\n\nDirect\n\n3.6%\n\nOTA\n\nhotel-link-destinations\n\n#### Hotel Link Destinations\n\nHotel Link Destinations\n\nDestination\n\nCount\n\nShare\n\nGoogle (GBP)\n\n4,851\n\n79.1%\n\nHotel website (direct)\n\n1,016\n\n16.6%\n\nOTA\n\n221\n\n3.6%\n\nMeta/review site\n\n42\n\n0.7%\n\nOTAs get almost nothing from hotel links — only 3.6%. Google keeps 79% within its own ecosystem (GBP), and hotels get 16.3% direct. The OTA dominance is only in the source citations, not in the actual hotel cards users click.\n\n### Two Surfaces, Two Behaviors\n\nAI Mode displays hotels in two distinct surfaces with very different link behaviors.\n\ntwo-surfaces-comparison\n\n#### Link Type by Surface\n\nAI Overview81% of mentions\n\n4,968 hotel mentions\n\nGBP4,812\n\nDirect75\n\nMeta42\n\nOTA39\n\nGBP dominates — Google acts as intermediary between user and hotel\n\nLocal Pack19% of mentions\n\n1,162 hotel mentions\n\nDirect (hotel website)941\n\nOTA182\n\nGoogle39\n\n81% supplier rate — most hotel-friendly surface\n\n### Source Citations\n\nSeparate from hotel links. These are the inline source links AI Mode uses to support its answers. OTAs dominate here — but users don't interact with these the same way.\n\n46.6%\n\nIntermediaries\n\nTripAdvisor, OTAs, meta\n\n21.2%\n\nSuppliers\n\nHotel websites\n\n19.9%\n\nGoogle Properties\n\nMaps, Travel, etc.\n\nSection 2\n\n## Local Pack Appearance\n\nHow often do map results appear in AI Mode responses?\n\n**What is the Local Pack?** The map-based results panel showing 3-4 nearby hotels with ratings, prices, and direct booking links. Unlike the AI Overview (which links to GBP), the Local Pack often links directly to hotel websites — making it the most hotel-friendly surface in AI Mode.\n\n6.5%\n\nof queries triggered a local pack\n\n### By Query Type\n\nWe tested 5 types of hotel queries to see which trigger the Local Pack most often:\n\n-   **Generic:** \"Best hotels in Paris\" — broad city-level queries\n-   **Boutique:** \"Best boutique hotels in Paris\" — style-filtered\n-   **Star Filter:** \"Best 3-star hotels in Paris\" — category-filtered\n-   **Neighborhood:** \"Best hotels in Paris Marais\" — location-specific\n-   **Persona:** \"Best hotels in Paris for couples\" — intent-based\n\nlocal-pack-by-query-type\n\n#### Local Pack Appearance by Query Type\n\nLocal Pack Rate by Query Type\n\nQuery Type\n\nLocal Pack Rate\n\nBoutique\n\n10%\n\nGeneric\n\n8.5%\n\nPersona\n\n6.1%\n\nStar Filter\n\n6%\n\nNeighborhood\n\n1.6%\n\nNeighborhood queries have the lowest local pack rate (1.6%) despite being the most location-specific. Boutique queries trigger it most often (10.0%).\n\n### By City\n\nWe tested the same prompts across 8 cities: Paris, London, New York, Barcelona, Berlin, Lisbon, Vienna, and Bordeaux. Local Pack appearance varies significantly by destination — London triggers it most (14.2%), Bordeaux least (1.2%).\n\nlocal-pack-by-city\n\n#### Local Pack Appearance by City\n\nLocal Pack Rate by City\n\nCity\n\nLocal Pack Rate\n\nLondon\n\n14.2%\n\nLisbon\n\n9.2%\n\nParis\n\n8.4%\n\nBarcelona\n\n5.8%\n\nBerlin\n\n5.4%\n\nNew York\n\n5.2%\n\nSection 3\n\n## Source Citations\n\n84,329 inline source citations — the small source links AI Mode uses to support its answers.\n\n### Sources by Type\n\nsources-by-type\n\n#### Source Citations by Type\n\nSource Citations by Type\n\nSource Type\n\nCount\n\nShare\n\nMetasearch/Reviews\n\n24,651\n\n29.2%\n\nDirect Website\n\n17,847\n\n21.2%\n\nGoogle Properties\n\n16,777\n\n19.9%\n\nOTAs\n\n14,627\n\n17.3%\n\nEditorial/Media\n\n9,214\n\n10.9%\n\nUGC (Reddit, etc.)\n\n1,213\n\n1.4%\n\n### Top 10 Domains\n\ntop-10-source-domains\n\n#### Top 10 Source Domains\n\nTop 10 Source Domains\n\nRank\n\nDomain\n\nType\n\nCitations\n\nShare\n\nRuns In\n\n#1\n\ngoogle.com\n\nGoogle\n\n16,770\n\n19.9%\n\n1,863\n\n#2\n\ntripadvisor.com\n\nMeta\n\n13,891\n\n16.5%\n\n3,422\n\n#3\n\nbooking.com\n\nOTA\n\n3,269\n\n3.9%\n\n1,792\n\n#4\n\nagoda.com\n\nOTA\n\n2,766\n\n3.3%\n\n1,629\n\n#5\n\ncntraveler.com\n\nEditorial\n\n2,463\n\n2.9%\n\n1,750\n\n#6\n\nexpedia.com\n\nOTA\n\n2,299\n\n2.7%\n\n1,604\n\nTripAdvisor appears in 3,422 of 4,000 runs (86%) — the most consistently cited source. Google cites itself in 19.9% of all sources.\n\nSection 4\n\n## Hotel Concentration\n\nHow concentrated are AI Mode's hotel recommendations?\n\nWe analyzed 6,082 hotel mentions across 8 cities and 5 query types. The story is nuanced: **London is highly competitive** (226 hotels, HHI 175), while **Bordeaux is dominated by 2 hotels** (Villas Foch + Yndo Hotel = 43% of mentions).\n\n6,082\n\nTotal Mentions\n\n971\n\nUnique Hotels\n\nLondon\n\nMost Competitive\n\nHHI 175 · 226 hotels\n\nBordeaux\n\nMost Concentrated\n\nHHI 1,169 · 74 hotels\n\n**What is HHI?** The Herfindahl-Hirschman Index measures market concentration. Below 1,000 = low concentration (competitive), 1,000–1,800 = moderate, above 1,800 = high (few players dominate). HHI is calculated per city AND per city×query-type combination for granular insights.\n\n### City-Level Concentration\n\nSmaller cities naturally show higher concentration (fewer hotels to recommend). London's 226 unique hotels make it the most competitive market.\n\nconcentration-by-city\n\n#### Market Concentration (HHI) by City\n\nConcentration by City\n\nCity\n\nHHI\n\nUnique Hotels\n\nTop 5 Share\n\nBordeaux\n\n1169\n\n74\n\n65.3%\n\nVienna\n\n618\n\n78\n\n45.4%\n\nLisbon\n\n463\n\n115\n\n40.1%\n\nBarcelona\n\n418\n\n97\n\n37.3%\n\nBerlin\n\n378\n\n114\n\n33.4%\n\nParis\n\n344\n\n165\n\n36.1%\n\n### City × Query Type Matrix\n\nConcentration varies dramatically by query type. **Family queries** in Berlin and New York are highly concentrated (2 hotels dominate), while **Boutique queries** in London have 88 hotels competing.\n\nconcentration-heatmap\n\n#### Concentration by City × Query Type (HHI)\n\nCity\n\nLuxury/5-star\n\nBoutique/Design\n\nBudget/Value\n\nFamily\n\nRomantic/Couples\n\nBarcelona\n\n613\n\n978\n\n1,034\n\n978\n\n964\n\nBerlin\n\n1,094\n\n566\n\n579\n\n2,141\n\n1,182\n\nBordeaux\n\n2,187\n\n2,348\n\n2,176\n\n1,645\n\n1,998\n\nLisbon\n\n1,435\n\n974\n\n2,099\n\n1,211\n\n428\n\nLondon\n\n544\n\n364\n\n665\n\n973\n\n464\n\nNew York\n\n748\n\n715\n\n906\n\n1,776\n\n838\n\nParis\n\n706\n\n650\n\n632\n\n1,068\n\n1,521\n\nVienna\n\n1,404\n\n1,474\n\n1,731\n\n1,897\n\n1,943\n\n<600 (Competitive) 600-1000 1000-1500 1500-2000 >2000 (Concentrated)\n\n### Most Concentrated Segments\n\nThese city×query combinations are dominated by 1-2 hotels. If you're competing here, you're up against entrenched favorites.\n\nMost Concentrated City × Query Type Combinations\n\nCity\n\nQuery Type\n\nHHI\n\nDominant Hotels\n\nBordeaux\n\nBoutique/Design\n\n2,348\n\nVillas Foch (35.2%), Yndo Hotel (31.6%)\n\nBordeaux\n\nLuxury\n\n2,187\n\nInterContinental Le Grand (24.1%), Villas Foch (18.9%)\n\nBerlin\n\nFamily\n\n2,141\n\nHotel Adlon Kempinski (33.3%), Hotel Telegraphenamt (30.0%)\n\nLisbon\n\nBudget\n\n2,099\n\nBlue Liberdade (32.4%), Hotel da Baixa (27.8%)\n\nVienna\n\nRomantic\n\n1,943\n\nHotel Sans Souci (33.1%), The Guesthouse Vienna (21.6%)\n\n### Most Competitive Segments\n\nThese markets are wide open. Lower HHI means no single hotel dominates — more opportunity for new entrants.\n\nMost Competitive City × Query Type Combinations\n\nCity\n\nQuery Type\n\nHHI\n\nHotels Competing\n\nTop Hotel\n\nLondon\n\nBoutique/Design\n\n364\n\n88\n\nBroadwick Soho (11.5%)\n\nLisbon\n\nRomantic\n\n428\n\n54\n\nThe Ivens (12.8%)\n\nLondon\n\nRomantic\n\n464\n\n56\n\nChateau Denmark (10.2%)\n\nLondon\n\nLuxury\n\n544\n\n45\n\nClaridge's (13.1%)\n\nBerlin\n\nBoutique\n\n566\n\n41\n\nWilmina Hotel (14.2%)\n\n### Dominant Hotels by City\n\nThese hotels get the most total mentions in AI Mode for their city. Note: Villas Foch (Bordeaux) gets 143 mentions despite Bordeaux having the smallest market — showing just how dominant it is.\n\nTop Hotel by City\n\nCity\n\nDominant Hotel\n\nTotal Mentions\n\nBordeaux\n\nVillas Foch\n\n143\n\nLisbon\n\nThe Ivens\n\n107\n\nBerlin\n\nHotel Adlon Kempinski\n\n99\n\nVienna\n\nHotel Sans Souci Wien\n\n93\n\nBarcelona\n\nMercer Hotel Barcelona\n\n77\n\nLondon\n\nClaridge's\n\n69\n\n**Key insight:** Concentration varies wildly by segment. Bordeaux Boutique is a duopoly (HHI 2,348) while London Boutique is highly competitive (HHI 364 with 88 hotels). Before investing in AI visibility, check YOUR specific city×query-type — not just citywide averages.\n\nMethodology\n\n## Study Design\n\n### Data Collection\n\n-   40 prompts (8 cities x 5 query tiers)\n-   25 runs per prompt per country via **SERPAPI**\n-   4 country proxies: US, DE, FR, ES\n-   Engine: google\\_ai\\_mode (no\\_cache=True)\n-   Date: Feb 2–3, 2026\n-   Total: 4,000 queries\n\n### What We Measured\n\n-   Hotel link destinations (GBP, direct, OTA)\n-   AI Overview vs Local Pack surface\n-   Source citations (URL, domain, type)\n-   Local pack appearance rate\n-   Concentration metrics (HHI)\n-   Cross-tab outlier detection\n\n### Data Quality\n\n100%\n\nVariance Validation\n\n160/160 groups show variance\n\n8.4%\n\nZero-Source Runs\n\n337 of 4,000 runs\n\n119\n\nFiltered Hotel Names\n\nEmpty names removed\n\n### Query Tiers\n\nTier\n\nExample\n\n1\\. Generic\n\nBest hotels in Paris\n\n2\\. Boutique\n\nBest boutique hotels in Paris\n\n3\\. Star Filter\n\nBest 3-star hotels in Paris\n\n4\\. Neighborhood\n\nBest hotels in Paris Marais\n\n5\\. Persona\n\nBest boutique hotels in Paris for a couple\n\n### Cities\n\nBarcelonaBerlinBordeauxLisbonLondonNew YorkParisVienna\n\n### Data Source\n\nData collected via **SERPAPI** google\\_ai\\_mode engine on Feb 2–3, 2026.\n\nCache disabled (no\\_cache=True) to ensure fresh responses for each query.\n\nSERPAPI is the same API used by ChatGPT for web search.\n\nFAQ\n\n## Frequently Asked Questions\n\n### Does Google AI Mode favor OTAs over hotel websites?\n\nNo — the opposite is true. While OTAs dominate source citations (46.6% of all sources), they receive only 3.6% of actual hotel links. Google AI Mode sends 79% of hotel links to Google Business Profiles, and 16.6% directly to hotel websites. OTAs are used as information sources but not as link destinations. Hotels that optimize their Google Business Profile capture the vast majority of AI Mode traffic.\n\n### How do hotels get traffic from Google AI Mode?\n\nHotels receive traffic from Google AI Mode through two main surfaces: the AI Overview (81% of mentions) and the Local Pack (19% of mentions). In the AI Overview, hotel names link primarily to Google Business Profiles (96.8%). In the Local Pack, 81% of links go directly to hotel websites. To maximize AI Mode visibility, hotels should: 1) Claim and optimize their Google Business Profile with accurate information, photos, and reviews, 2) Ensure their website has proper schema markup, and 3) Maintain presence across review sites that AI Mode cites as sources.\n\n### What is a Google Business Profile link in AI Mode?\n\nWhen a user clicks on a hotel name in Google AI Mode's response, they're taken to the hotel's Google Business Profile (GBP) — the same listing that appears in Google Maps and Search. This page shows the hotel's address, photos, reviews, amenities, and booking options. From the GBP, users can then click through to the hotel's website or book via integrated booking partners. 79.1% of all hotel links in AI Mode go to these GBP pages, making GBP optimization essential for AI visibility.\n\n### What is the Local Pack in Google AI Mode?\n\nThe Local Pack is a map-based results panel that sometimes appears in Google AI Mode responses, showing 3-4 nearby hotels with ratings, prices, and booking links. It appears in only 6.5% of hotel queries but is highly valuable because 81% of its links go directly to hotel websites (compared to just 1.5% in the AI Overview). Boutique hotel queries trigger the Local Pack most often (10%), while neighborhood-specific queries trigger it least (1.6%). London has the highest Local Pack appearance rate (14.2%).\n\n### Which sources does Google AI Mode trust for hotel information?\n\nGoogle AI Mode cites a diverse mix of sources: Metasearch and review sites lead with 29.2% (primarily TripAdvisor at 16.5%), followed by hotel websites directly at 21.2%, Google's own properties at 19.9%, and OTAs at 17.3% (Booking.com 3.9%, Agoda 3.3%, Expedia 2.7%). Editorial sites like Condé Nast Traveler contribute 10.9%. TripAdvisor is the most consistently cited source, appearing in 86% of all queries. The key insight: being present across multiple source types increases your chances of being recommended.\n\n### How concentrated are Google AI Mode hotel recommendations?\n\nGoogle AI Mode shows a healthy long tail of hotel recommendations. Across 4,000 queries, we identified 1,146 unique hotels. The top 5 hotels captured only 7.6% of all mentions, and 51% of hotels appeared just once. The overall HHI (Herfindahl-Hirschman Index) of 45.2 indicates very low concentration — no single hotel dominates. Smaller cities naturally show higher concentration (Bordeaux: HHI 810) than larger ones (London: HHI 156), reflecting the available hotel inventory rather than AI bias.\n\n## Continue Reading\n\nExplore more Nicolas Sitter research on AI hotel search.\n\n[AI Hotel Landscape 2026](/research/ai-hotel-landscape-2026)\n\n[Anatomy of ChatGPT Search](/research/anatomy-chatgpt-hotel-search-2026)[Ranking Consistency Study](/research/ai-hotel-rankings-consistency-study-2026)[All Research](/research)","author":{"@type":"Person","name":"Nicolas Sitter","url":"https://nicolassitter.com/about","sameAs":["https://www.linkedin.com/in/nicolassitternolleau/","https://github.com/Nicositter88","https://hotelrank.ai"]},"publisher":{"@type":"Person","name":"Nicolas Sitter","url":"https://nicolassitter.com"},"image":"https://nicolassitter.com/api/og/google-ai-mode-hotel-study-2026","mainEntityOfPage":{"@type":"WebPage","@id":"https://nicolassitter.com/research/google-ai-mode-hotel-study-2026"},"tags":["Google AI Mode","Hotel Search","GBP","Click Distribution"],"sameAs":["https://hotelrank.ai/research/google-ai-mode-hotel-study-2026"],"alternateFormat":{"html":"https://nicolassitter.com/research/google-ai-mode-hotel-study-2026","json":"https://nicolassitter.com/api/post/google-ai-mode-hotel-study-2026","rss":"https://nicolassitter.com/rss.xml"},"datasets":[{"name":"summary","contentUrl":"https://nicolassitter.com/data/google-ai-mode-hotel-study-2026/summary.csv","encodingFormat":"text/csv"}]}