I Asked ChatGPT to Recommend Restaurants in Lyon — Here's Who Made the List
Lyon is the gastronomic capital of France. But when AI decides where to eat, the results might surprise you. We tested how ChatGPT, Perplexity, and Google AI recommend restaurants in Lyon — and what separates the winners from the invisible.
The experiment
Lyon has over 4,000 restaurants. It's the birthplace of modern French cuisine, home to legendary bouchons, Michelin-starred institutions, and a thriving new generation of bistronomie spots. Competition for diners is fierce.
We wanted to know: when someone asks an AI assistant "Where should I eat in Lyon?" or "Recommend a good restaurant in Lyon" — who shows up? And more importantly, why do they show up while thousands of others don't?
We ran multiple queries across ChatGPT, Perplexity, and Google AI Overviews. We varied the prompts: casual ("best restaurants in Lyon"), specific ("traditional bouchon in Vieux Lyon"), and intent-based ("romantic dinner in Lyon for anniversary"). Here's what we found.
The results: who AI recommends
Across all three AI platforms, a clear pattern emerged. The same handful of names appeared repeatedly, while the vast majority of Lyon's restaurants were completely absent.
Consistently recommended (appeared across multiple AI platforms):
- ✓Paul Bocuse (L'Auberge du Pont de Collonges) — Legendary name, massive web presence, mentioned in thousands of articles worldwide
- ✓Les Halles de Lyon Paul Bocuse — Referenced in nearly every "things to do in Lyon" travel guide
- ✓Café Comptoir Abel — One of Lyon's oldest bouchons, featured extensively in travel blogs and food media
- ✓Daniel & Denise — Strong review presence on multiple platforms, frequently mentioned in "best bouchon" roundups
- ✓Le Bouchon des Filles — Consistently praised across review platforms with a clear brand identity
Rarely or never mentioned by AI:
- ×Restaurants with great food but minimal web presence beyond a basic website
- ×Newer establishments that haven't yet accumulated reviews and editorial mentions
- ×Restaurants that rely primarily on word-of-mouth or physical foot traffic
- ×Establishments with inconsistent information across platforms (different names, outdated hours, missing menus)
Why these restaurants? The 5 patterns we found
After analyzing the results across dozens of queries, we identified five common factors that separated the AI-recommended restaurants from the invisible ones.
1. Volume and diversity of mentions
Every restaurant that AI recommended consistently appeared in multiple types of content: travel blogs, food review sites, local guides, international press, and user-generated content. It wasn't just reviews — it was editorial coverage, listicles, and social media mentions. The AI doesn't count backlinks; it absorbs mentions from everywhere.
2. Consistent identity across platforms
The winners had the same name, same description, same positioning on Google Maps, TripAdvisor, TheFork, their own website, and social media. When AI cross-references these sources, consistency builds confidence. Restaurants with mismatched names (e.g., "Chez Pierre" on Google but "Restaurant Pierre Martin" on their website) create confusion that AI resolves by... recommending someone else.
3. Review volume AND quality on multiple platforms
Having 500 Google reviews matters. But having 500 Google reviews plus 300 TripAdvisor reviews plus a strong TheFork rating matters far more. AI synthesizes across platforms. A restaurant with great reviews on only one platform has a thinner signal than one that's consistently praised everywhere.
4. A clear specialty or positioning
When we asked for "traditional bouchon lyonnais," the AI recommended restaurants that clearly positioned themselves as bouchons. When we asked for "romantic dinner," it picked places described as intimate and elegant across their web presence. The restaurants that tried to be everything — bouchon, bistro, gastro, pizza — got recommended for nothing.
5. Recent, fresh content signals
The AI showed a preference for restaurants with recent reviews (last 6 months), recently updated websites, and recent editorial mentions. A restaurant that was heavily covered in 2019 but has had no new content since then was less likely to be recommended than one with steady, ongoing mentions.
The bigger picture: this isn't just about restaurants
We used Lyon restaurants as a case study, but the same patterns apply to every local business in every industry. Hotels, estate agents, law firms, dental practices, wedding venues, salons — the rules are the same:
- AI recommends businesses it "knows" well from diverse sources
- Consistency across platforms is more important than perfection on one
- Clear positioning beats trying to appeal to everyone
- Freshness matters — AI favours businesses with recent activity
- Most businesses are completely invisible to AI — the opportunity is wide open
What restaurant owners (and any local business) can do today
- 1
Test it yourself — right now
Open ChatGPT or Perplexity and ask: "Recommend a good [your type of business] in [your city]." See if you appear. Note who does. This takes 60 seconds and the results will be eye-opening.
- 2
Unify your brand across every platform
Same name, same description, same hours, same menu, same photos on Google Maps, TripAdvisor, TheFork, your website, Instagram, and Facebook. Inconsistency is the single biggest AI visibility killer for local businesses.
- 3
Actively pursue editorial mentions
Reach out to local food bloggers, city guides, and travel writers. A single mention in a well-read blog can significantly boost your AI visibility. Offer a complimentary tasting — it's the best marketing ROI you'll ever get.
- 4
Get reviews on multiple platforms
Don't just focus on Google. Encourage customers to review you on TripAdvisor, TheFork, Yelp, and any industry-specific platforms. AI cross-references all of them.
- 5
Own your positioning — be specific
"The best traditional bouchon in Vieux Lyon" is a position AI can work with. "A restaurant in Lyon serving various cuisines" is not. Pick your niche and make sure every platform reflects it.
The window is still open
Here's the good news: AI search is still early. Most businesses haven't optimised for it at all. In Lyon, out of 4,000+ restaurants, only a handful are being consistently recommended by AI. That means the competition to get recommended is far less fierce than traditional SEO or paid advertising.
But this window won't stay open forever. As more businesses catch on and start optimising their AI presence, the early movers will have built an insurmountable advantage — because AI learns and reinforces its own recommendations over time.
The question isn't whether AI search matters for your business. It's whether you'll be one of the businesses it recommends.
Want to know how AI search sees your business?
Our AI PR + SEO Growth Guide audits your visibility across ChatGPT, Perplexity, and Google AI Overviews — with a step-by-step action plan to get recommended.
Get Your Audit — $49One-time payment · Delivered in 48 hours · No subscription