Restaurants, bars, caterers and delivery kitchens appear in AI answers through a dense layer of local sources: review platforms, table reservation and food delivery apps, gastronomic guides, local directories, tourist offices and social networks. Questions are hyper-local and practical, so an AI assistant recommends the places whose cuisine, prices, opening hours and atmosphere are stated clearly and consistently across these sources.
TripAdvisorTripAdvisorBiggest movers in October 2026
Rises
- TripAdvisor10.32 to 17.09+6.77 pt
Yelp0.44 to 1.81+1.37 pt
Trustpilot0.59 to 1.36+0.77 pt
Falls
Facebook2.47 to 1.77-0.70 pt
Most cited domains in AI answers, Restaurants
9 of the 10 most cited domains in Restaurants (geneve.com, petitfute.com and ubereats.com) do not appear among the 50 most cited domains of the whole panel.
How to read it: each bar is a domain; its length is its number of citations per 100 answers over the selected period and filters.
Domains ranked by citations per 100 answers, all periods combined.
| Rank | Rank change | Domain | Citations | per 100 answers | Share of citations | 12 months |
|---|---|---|---|---|---|---|
| 1 | Up 1 | TripAdvisor tripadvisor.com | 1'325 | 17.26 | 3.94% | Trend: 25.68 in Jul 2026, 17.36 in Oct 2026, high 25.68. |
| 2 | Down 1 |  geneve.com | 959 | 12.50 | 2.85% | Trend: 8.14 in Jul 2026, 15.85 in Oct 2026, high 15.85. |
| 3 | Up 3 |  petitfute.com | 668 | 8.70 | 1.99% | Trend: 14.32 in Jul 2026, 8.23 in Oct 2026, high 14.32. |
| 4 | Down 5 |  ubereats.com | 643 | 8.38 | 1.91% | Trend: 17.05 in Jul 2026, 5.40 in Oct 2026, high 17.05. |
| 5 | Up 3 |  thefork.fr | 580 | 7.56 | 1.72% | Trend: 15.37 in Jul 2026, 4.87 in Oct 2026, high 15.37. |
| 6 | Up 3 |
| 389 | 5.07 | 1.16% | Trend: 4.56 in Jul 2026, 6.49 in Oct 2026, high 6.49. |
| 7 | Up 4 |  thefork.ch | 377 | 4.91 | 1.12% | Trend: 4.06 in Aug 2026, 8.15 in Oct 2026, high 8.15. |
| 8 | out of top |  just-eat.ch | 300 | 3.91 | 0.89% | Trend: 7.44 in Jul 2026, 5.73 in Sep 2026, high 7.44. |
| 9 | Up 5 |  privateaser.com | 248 | 3.23 | 0.74% | Trend: 3.70 in Sep 2026, 6.53 in Oct 2026, high 6.53. |
| 10 | Up 4 |  ge.ch | 236 | 3.07 | 0.70% | Trend: 3.37 in Jul 2026, 3.21 in Oct 2026, high 3.43. |
| 11 | Down 1 |  chezphilippe.ch | 230 | 3.00 | 0.68% | Trend: 4.00 in Aug 2026, 3.55 in Oct 2026, high 4.00. |
| 12 | Down 6 |  mariages.net | 216 | 2.81 | 0.64% | Trend: 5.63 in Sep 2026, 3.85 in Oct 2026, high 5.63. |
| 13 | new |  pagesjaunes.fr | 208 | 2.71 | 0.62% | Trend: 4.00 in Jul 2026, 3.55 in Oct 2026, high 4.00. |
| 14 | new |  caveduboucher.ch | 207 | 2.70 | 0.62% | Trend: 4.56 in Jul 2026, 2.87 in Oct 2026, high 4.56. |
| 15 | out of top |  kerzesto.com | 173 | 2.25 | 0.51% | Trend: 4.38 in Aug 2026, 5.14 in Sep 2026, high 5.14. |
| 16 | out of top |  tacos-tacos.ch | 160 | 2.08 | 0.48% | Trend: 4.00 in Jul 2026, 2.67 in Sep 2026, high 4.00. |
| 17 | new |
| 158 | 2.06 | 0.47% | 5.96 in Oct 2026. |
| 18 | out of top | Facebook facebook.com | 152 | 1.98 | 0.45% | Trend: 3.79 in Jul 2026, 2.47 in Sep 2026, high 3.79. |
| 19 | new |
| 131 | 1.71 | 0.39% | 4.94 in Oct 2026. |
| 20 | out of top |  lacrepeduperegege.fr | 129 | 1.68 | 0.38% | Trend: 3.56 in Aug 2026, 3.60 in Sep 2026, high 3.60. |
michelin.com
tripadvisor.be
tripadvisor.chRankings are rebuilt from monthly rankings: the counts at the end of a list can be minimums. Industry: the business of the analyzed website, classified by Sorank. Data since July 2026. Rank change: rank in the October 2026 monthly list against September 2026 (new: absent from the September 2026 list). A dash: fewer than 1'000 answers in one of the two months in this scope, or absent from the October 2026 list. 12 months: monthly rate per 100 answers; gaps when the domain is out of the monthly list, hollow points for months under 1'000 answers.
Most cited platforms in AI answers, Restaurants
TripAdvisor is the platform AI cites most in Restaurants: 16.53 citations per 100 answers, 18.5 times its rate across all sectors over the same months (0.89).
How to read it: each bar is one of the tracked platforms; its length is its number of citations per 100 answers over the selected period.
| Rank | Rank change | Platform | per 100 answers | 12 months |
|---|---|---|---|---|
| 1 | Unchanged | TripAdvisor | 16.53 | Trend: 23.79 in Jul 2026, 17.09 in Oct 2026, high 23.79. |
| 2 | Down 1 | Facebook | 2.59 | Trend: 3.79 in Jul 2026, 1.77 in Oct 2026, high 3.79. |
| 3 | Unchanged |  Instagram | 1.37 | Trend: 5.68 in Jul 2026, 0.42 in Oct 2026, high 5.68. |
| 4 | Down 1 |  Reddit | 1.29 | Trend: 3.37 in Jul 2026, 0.64 in Oct 2026, high 3.37. |
| 5 | Up 5 | Yelp | 1.04 | Trend: 0.84 in Jul 2026, 1.81 in Oct 2026, high 1.81. |
| 6 | Up 1 | Trustpilot | 0.98 | Trend: 1.26 in Jul 2026, 1.36 in Oct 2026, high 1.36. |
| 7 | Up 2 |  Wikipedia | 0.60 | Trend: 1.47 in Jul 2026, 0.15 in Oct 2026, high 1.47. |
| 8 | Down 11 |  LinkedIn | 0.26 | Trend: 0.00 in Jul 2026, 0.04 in Oct 2026, high 0.84. |
| 9 | No comparison |  TikTok | 0.16 | Trend: 0.84 in Jul 2026, 0.00 in Oct 2026, high 0.84. |
| 10 | Down 3 |  Apple | 0.15 | Trend: 0.28 in Jul 2026, 0.04 in Oct 2026, high 0.28. |
Rank change: rank among the platforms cited in October 2026 against September 2026 (new: not cited in September 2026). A dash: fewer than 1'000 answers in one of the two months in this scope, or not cited in October 2026. 12 months: monthly rate per 100 answers, hollow points for months under 1'000 answers.
AI models compared
AI assistants do not agree on the first source in Restaurants: ChatGPT, Perplexity and Claude cite TripAdvisor first; Gemini cites petitfute.com first.
How to read it: left, how the citations split between the five models; right, each model's citations per 100 answers over time.
First source of each AI model in Restaurants
| Model | First source | per 100 answers | Second source |
|---|---|---|---|
 ChatGPT | TripAdvisor tripadvisor.com | 12.48 | geneve.com |
 Gemini | petitfute.com | 14.59 | geneve.com |
 Perplexity | TripAdvisor tripadvisor.com | 31.43 | ubereats.com |
 Claude | TripAdvisor tripadvisor.com | 33.88 |
|
What diners ask AI assistants before choosing a restaurant
Restaurant prompts read like a conversation with a well-informed local. They combine a place, a moment and a constraint:
- "good Italian restaurant open on Sunday evening in Annecy"
- "restaurant with a terrace for a group of 15 near the old town"
- "gluten-free brunch in Geneva", "vegan tacos delivered tonight"
- "affordable Michelin-starred lunch menu"
- "caterer for a wedding of 120 guests in Corsica"
Restaurant owners ask AI too, about regulation and management: which licence is needed to serve alcohol, what hygiene training is mandatory, how to open a food truck. Those answers lean on government and chamber of commerce pages rather than on consumer platforms.
Which sources do AI models quote about restaurants?
- Review platforms: TripAdvisor and Yelp provide ratings and diner comments on food, service and value.
- Reservation apps: TheFork publishes menus, price ranges and booking slots for a very large number of restaurants.
- Delivery apps: Uber Eats and Just Eat expose full menus with prices, an easy source for questions about takeaway or delivery.
- Guides: the Michelin Guide for fine dining, Petit Futé and similar guides for everyday addresses.
- Local directories and tourist offices: PagesJaunes in France, local.ch in Switzerland, city tourism portals that list places to eat.
- Social networks: Facebook and Instagram, often the only place where a small restaurant posts its daily menu or a holiday closure.
- Restaurant websites, cited when they hold a readable menu, a clear address and practical details.
Event-related platforms join the list for private dining and catering: wedding vendor directories and venue booking sites are quoted when the question is about a group, a party or a wedding meal.
Why menus, hours and dietary details decide AI recommendations
An AI assistant cannot taste food. It recommends a restaurant when it can justify the choice with verifiable facts: cuisine, price per person, opening days, dietary options, terrace, capacity, location. Sources that state those facts in structured text, like reservation and delivery apps, are naturally favoured.
This is where many restaurants lose visibility. A menu published only as a photo on Instagram or as a scanned PDF is hard for crawlers to read, and part of what is posted on social networks is hidden from search engines. The same menu written as text on the restaurant's website becomes a source a model can quote.
Wrong or contradictory facts are the other trap. If a directory still shows old opening hours, an AI may repeat them. Updating listings after every change of schedule is part of AI visibility, not only customer service.
Do all AI models recommend restaurants the same way?
No. Gemini grounds its answers in Google Search, where local results weigh heavily (Google itself is not counted as a cited source on this site). Perplexity searches the web for each question and shows the pages it used, so review and guide pages appear openly in its answers. ChatGPT cites restaurant pages mainly when it runs a web search, and Claude tends to answer with fewer explicit sources.
The tracked questions are mostly asked in French and concern several French-speaking areas, so local directories and tourism portals from more than one country can appear side by side in the ranking.
How a restaurant can earn AI citations
- Publish your menu as text on your website, with prices, dietary labels (vegetarian, gluten-free, halal) and opening hours, and keep a PDF only as a complement.
- Claim your listings on TripAdvisor, TheFork, Yelp and local directories, and use exactly the same name, address and phone number everywhere.
- If you deliver, keep your menus on Uber Eats or Just Eat current: these pages are read by AI models when people ask what they can order tonight.
- Encourage diners to mention specifics in their reviews (the terrace, the brunch, the group room) and reply by restating those details.
- Seek the editorial coverage AI models trust for "best" questions: the tourist office of your town, local guides, food journalists and city blogs.
- If you cater for events, create one page per offer (weddings, corporate lunches, private hire) and list it on wedding and venue directories. That is GEO (Generative Engine Optimization) applied to a local business: one clear page per intent.
Related industries
Sectors sharing the most domains with Restaurants among their 20 most cited, then sectors with a similar number of answers.
Frequently asked questions
How does ChatGPT choose which restaurants to recommend?
When it searches the web, ChatGPT looks for pages that match the place and criteria of the question, then names the restaurants that several sources describe positively. Review platforms, reservation apps, guides and the restaurant's own site all count. Without a web search, it relies on its training data, which favours long-established and widely reviewed places over recent openings.
Does a restaurant need to be on TheFork or Uber Eats to be cited by AI?
No, but these apps make it easier. They give each restaurant a page with a structured menu, prices and ratings, which AI models read easily. A restaurant absent from them can still be cited through its own website, review platforms, local guides and the tourist office, provided these sources state the same facts about cuisine, location and opening hours.
Can Instagram help a restaurant appear in AI answers?
Somewhat. Instagram and Facebook are cited in restaurant answers, mostly to confirm that a place exists, show its atmosphere or give recent news. But images and stories carry little readable text, and some content is not accessible to crawlers. Use social networks for freshness, and keep the core facts (menu, hours, address) in text on your website and listings.
Why does an AI give wrong opening hours for my restaurant?
Because it copied them from an outdated source. AI assistants read directories, review platforms and old pages, and if one of them still shows your former schedule, the error can resurface in answers. Update your hours everywhere you are listed, publish them in text on your website, and announce exceptional closures there as well as on social networks.
Which websites do AI assistants cite for Restaurants businesses?
TripAdvisor (17.26 citations per 100 answers), geneve.com (12.50) and petitfute.com (8.70), across 7'675 answers from ChatGPT, Claude, Gemini, Grok and Perplexity about Restaurants businesses since July 2026.
Which platforms do AI assistants cite in Restaurants?
TripAdvisor (16.53 citations per 100 answers), Facebook (2.59) and Instagram (1.37).
Method and sources
498'158 answers from ChatGPT, Claude, Gemini, Grok and Perplexity, from August 2025 to October 2026, to the questions Sorank asks every week to follow its clients' visibility (87% of them in French). A citation is a source domain given in an answer; the rate is the number of citations per 100 answers. Data snapshot of 2 November 2026.
Full methodology · Raw data (JSON)
How to cite this page: Sorank Data, “Websites AI cites most in Restaurants (Oct 2026)”, sorank.com/data, data snapshot of 2 November 2026.
Track the citations of your own site with Sorank's AI mention tracking.
