The barometer of brand visibility across ChatGPT, Gemini, Perplexity and Meta AI — from Europe to Bangkok. 5,347 citations analysed, 1,057 players, 8 sectors — and a ranking by the KVS score.
When a decision-maker, a patient or a parent asks an AI “what's the best option?”, four engines — ChatGPT, Gemini, Perplexity, Meta AI — shape the answer. We queried these AIs across 394 queries, extracted 5,347 citations covering 1,057 players and 8 sectors, from Europe to Bangkok, then computed a proprietary score: the KVS.
Three companies reach a KVS of 100/100 in their sector. Common thread: perfect coverage across all three AIs, not just one.
Unlike SEO, where US giants crush the rankings, the AIs mostly cite European players on European queries.
Multi-AI coverage is the single most decisive factor. 91% of the top 40 is cited by all three models simultaneously — the new standard.
Paris concentrates 40% of SaaS leaders (Mistral, Doctolib, Contentsquare, Dataiku). London polarises fintech (Revolut, Wise, Checkout.com).
Companies valued at over €500M appear in none of the queries. AI visibility isn't correlated with size — but with structure.
Training corpus, Schema.org, brand authority and primary sources are replacing backlinks as the dominant signal.
A single firm dominates both “consumer” and “market” queries. The Anglo-Saxon Magic Circle caps below 18: a massive premium for native French-language players.
SeLoger + Meilleurs Agents capture 71% of citations. No local agency appears on “best agency in [city]”: the ground is wide open.
21 rank-1 mentions out of 25. The sector cleanly separates the “user” layer (Doctolib, Kry, Qare) from the “market” layer (hospitals, clinical AI tools).
Howard Avocats owns Paris employment law (16 answers out of 24). But 64% of the 232 firms cited appear only once — and as soon as the question turns to price, citation drops to 17%.
NIST, Patana and ISB lock in the recommendations. Meanwhile, Perplexity cites a school that doesn't exist 21 times — the “Dulwich ghost”, spread by SEO aggregators.
64 first-place spots across 83 patient queries, on all 4 AIs. Behind it, Kamol monopolises gender-reassignment surgery and 7 ghost queries cite no clinic at all.
A snapshot of the top per sector. The KVS weighs coverage across the panel's AIs, the average citation rank and volume.
| Sector | Cited leader | KVS | Followed by |
|---|---|---|---|
| SaaS / Tech | Mistral AI | 100 | Doctolib · Contentsquare · Dataiku |
| Finance / Fintech | Revolut | 100 | Adyen (96.0) · Stripe · Wise |
| Law · France (business) | Bredin Prat | 100 | Gide (64) · Magic Circle (<18) |
| Real estate | SeLoger | 100 | Meilleurs Agents (87.6) · Idealista · Notaires de France |
| Health · Europe | Doctolib | 100 | Kry · Qare · hospitals · clinical AI tools |
| Law · Paris & Lyon | Howard Avocats | 100 | Facchini (Lyon, 100) · Delsart (86.0) · Hoche (44.3) |
| International schools · Bangkok | NIST | 100 | Bangkok Patana (85.7) · ISB (82.0) · Shrewsbury (51.0) |
| Health & aesthetics · Bangkok | Bumrungrad | 100 | Yanhee (79.0) · Kamol (78.8) · Wansiri (59.8) |
Full ranking of the top 40 brands per sector, with the AI-by-AI breakdown, in the PDF report.
On European SaaS/Tech queries, the French ecosystem dominates. Six of the top ten citations are French.
| # | Company | Country | KVS |
|---|---|---|---|
| 1 | Mistral AI | 🇫🇷 France | 100.0 |
| 2 | Doctolib | 🇫🇷 France | 90.8 |
| 3 | Odoo | 🇧🇪 Belgium | 74.2 |
| 4 | Contentsquare | 🇫🇷 France | 72.9 |
| 5 | Qonto | 🇫🇷 France | 66.8 |
| 6 | Dataiku | 🇫🇷 France | 65.1 |
Stripe, Typeform and Pipedrive — global SEO champions — cap between KVS 41 and 49 on European queries. The LLMs reproduce the corpus of the European tech press, not the Google ranking.
Every company above KVS 40 is cited by ChatGPT, Gemini and Perplexity alike. Single-engine players cap at 30-40 points, whatever the product and the funding.
The head captures the essentials: beyond the top 12, dozens of players earn a single mention and vanish. One-off citations build nothing.
Paris concentrates 40% of the SaaS/Tech leaders cited by the AIs (Mistral, Doctolib, Contentsquare, Dataiku, Algolia…). French DeepTech is better documented in training corpora than people assume — the fruit of intense press coverage since 2020. For a challenger, head queries (“best CRM”, “best project-management tool”) are locked in by years of accumulated corpus: the lever isn't one more feature list, but becoming the textual definition of a precise workflow — the Doctolib effect — documented in top-tier tech media.
European fintech is polarised by the City. Adyen holds the Dutch seat; Stripe runs its European operations from Dublin.
| # | Company | Country | KVS |
|---|---|---|---|
| 1 | Revolut | 🇬🇧 United Kingdom | 100.0 |
| 2 | Adyen | 🇳🇱 Netherlands | 96.0 |
| 3 | Stripe | 🇮🇪 Ireland | 83.2 |
| 4 | Wise | 🇬🇧 United Kingdom | 75.4 |
| 5 | Qonto | 🇫🇷 France | 69.4 |
| 6 | Pennylane | 🇫🇷 France | 42.1 |
Revolut (100), Adyen (96.0), Stripe (83.2), Wise (75.4), Qonto (69.4) — number 6, Pennylane, already drops to 42.1, and the curve keeps falling.
Stripe takes rank 1 more often than anyone (17 times) yet stays third. Revolut is cited 41 times across the panel: being present everywhere weighs more than winning somewhere.
258 unique companies cited for 40 FR + EN queries. Hundreds of players earn one mention and disappear — only repetition across all three engines scores.
Beyond the Paris–London–Berlin trio, Amsterdam (Adyen, Mollie), Stockholm (Klarna, Tink) and Barcelona (Typeform, Factorial) produce champions. Italy, though, is absent from the top 30 — a deficit of public documentation, not of talent. For a fintech outside the top 5, taking on neobank or payment queries is a corpus war lost in advance: the lever is to own a precise financial job-to-be-done — expense management, treasury, revenue-based financing — across all three engines at once. That corpus is thin, and a KVS of 20-30 is reachable there within two quarters.
Across 30 queries in two layers — 15 “consumer” and 15 “market” — a single firm dominates both. The Anglo-Saxon giants cap out on French-language queries.
| # | Company | Country | KVS |
|---|---|---|---|
| 1 | Bredin Prat | 🇫🇷 France | 100.0 |
| 2 | Gide Loyrette Nouel | 🇫🇷 France | 64.0 |
| 3 | Darrois Villey Maillot Brochier | 🇫🇷 France | 31.8 |
| 4 | Freshfields | 🇬🇧 United Kingdom | 29.9 |
| 5 | Captain Contrat (legaltech) | 🇫🇷 France | 20.5 |
Bredin Prat (100) dominates Gide Loyrette Nouel (64.0) and Darrois Villey (31.8). The fifth most-cited player is a legaltech — Captain Contrat (20.5).
Linklaters (17.8) and Clifford Chance (14.6) cap below 18: their corpus lives in Chambers and Legal 500, not in the French-language sources the LLMs read.
Firms visible on a single layer — consumer or market — cap at KVS 12-20. Cross-layer visibility is multiplicative, not additive.
For a firm ranked between 10 and 30, the lever isn't more generalist content — that's saturated. It's capturing a vertical niche (healthcare M&A, AI, energy, international succession) with dense output, marked up with LegalService + Attorney + FAQPage. A niche firm goes from KVS 5 to 40+ in 9 months and overtakes full-service firms four times its size.
SeLoger and Meilleurs Agents (AVIV group) together take nearly three quarters of AI mentions in French real estate. No brick-and-mortar agency appears before rank 14, with a KVS below 8.
| # | Company | Country | KVS |
|---|---|---|---|
| 1 | SeLoger | 🇫🇷 France | 100.0 |
| 2 | Meilleurs Agents | 🇫🇷 France | 87.6 |
| 3 | Idealista | 🇪🇸 Spain | 44.3 |
| 4 | Notaires de France | 🇫🇷 France | 43.1 |
| 5 | Rightmove | 🇬🇧 United Kingdom | 27.1 |
SeLoger (100) and Meilleurs Agents (87.6) — same owner — absorb nearly three quarters of mentions. No physical agency or network before rank 14, below KVS 8.
Local agencies with hundreds of Google reviews are textually non-existent to the engines: they publish no structured content about their own markets.
On REIT, crowdfunding and yield queries, the leaders change: Wemo One (19.8), Iroko Atlas (18.0), EstateGuru (12.1) — absent from the consumer top 5. A field barely defended.
An agency at 0 mentions can reach KVS 25-40 in its area within 6-9 months: 12 hyper-local guides (“Buying in the 15th arrondissement”, “EPC C or D: the real price after renovation”), RealEstateAgent + LocalBusiness + FAQPage markup, 20 local-press citations via Digital PR, a monthly cadence. On a geolocated query, the LLMs prefer a precise local player to a national aggregator — when that player exists textually.
With 21 rank-1 appearances out of 25, Doctolib is the only player across the five European sectors to dominate both the consumer and the professional query. Below it, two worlds with no company in common.
| # | Company | Country | KVS |
|---|---|---|---|
| 1 | Doctolib | 🇫🇷 France | 100.0 |
| 2 | Kry / Livi | 🇸🇪 Sweden | 30.7 |
| 3 | Qare | 🇫🇷 France | 26.9 |
| 4 | Maiia | 🇫🇷 France | 14.5 |
| 5 | Medadom | 🇫🇷 France | 8.4 |
21 rank-1 appearances out of 25 citations. Doctolib is the only company across the five European sectors to win both the consumer and the professional query.
User layer: Kry/Livi (30.7), Qare (26.9), Maiia (14.5). Market layer: Karolinska (21.0), DAX Copilot (15.6), Charité Berlin (15.0), Nabla (9.5). Almost no company in both.
Nabla, Owkin and Corti are each strong on one or two engines — never all three. Disciplined multi-AI coverage takes a player from KVS 5 to 30 in six months.
Three moves depending on your position: (1) teleconsultation platform ranked 5-15 — the “user” fight is lost, pivot to a vertical (mental health, dermatology, paediatrics); (2) clinical-AI vendor — your KVS hinges on Perplexity, which rewards technical documentation and certifications (e.g. CE / health-data hosting) cited textually; (3) hospital/clinic — invisible everywhere: Hospital + MedicalBusiness page, published PROMs, press citations. 12-18 months, but very few competitors.
80 individual-client queries — 5 practice areas, 2 cities — put to Gemini, Meta AI and Perplexity: 640 citations across 232 firms. A firm is named in 75% of answers. The sector's problem isn't absence — it's volatility and concentration.
| # | Paris | KVS | Lyon | KVS |
|---|---|---|---|---|
| 1 | Howard Avocats — employment law | 100.0 | Facchini Avocat — real estate | 100.0 |
| 2 | Hoche Avocats — business | 44.3 | Delsart Avocats — IP | 86.0 |
| 3 | Kohen Avocats — employment | 39.1 | Bressand Avocat — IP | 81.3 |
| 4 | Cabinet Zenou — employment | 38.7 | Cabinet Bouchara — IP | 74.9 |
| 5 | Avi Bitton — employment | 37.2 | Erovic Avocats — employment | 57.1 |
KVS normalised to 100 per city, against the local leader.
As soon as the question turns to price, a firm's citation drops to 17% — and Perplexity names no one. The sector's biggest blind spot, and its most vacant territory.
232 firms named, but the tail is accidental: being cited once is down to luck. Recurring citation, by contrast, is a built asset — dated expertise content, a consistent entity, a structured site.
Gemini names 3.1 firms per answer, Meta AI 1.3. And in Lyon, the AI falls back on the Bar twice as often as in Paris (33 referrals vs 17): it's looking for a firm to name — the Lyon window is open.
Some firms already own their answer: Howard in Paris employment law (16 answers out of 24), Delsart in IP and Facchini in real estate in Lyon. Proof, in the data, that the spot can be taken — by practice area × city. Directories, meanwhile, scatter: no platform exceeds 5% of answers. There's no monopoly to unseat — there's a spot to occupy.
86 queries from the parental decision journey — discovery, curriculum, budget, SEN, relocation — put to ChatGPT, Gemini and Perplexity: 1,769 citations across some sixty schools, 10 intent categories.
| # | School | Rank 1 | KVS |
|---|---|---|---|
| 1 | NIST International School | 68 | 100.0 |
| 2 | Bangkok Patana School | 28 | 85.7 |
| 3 | ISB — International School Bangkok | 51 | 82.0 |
| 4 | Shrewsbury International School | 16 | 51.0 |
| 5 | St Andrews International Schools | 18 | 40.9 |
| 6 | Harrow International School Bangkok | 10 | 31.5 |
68 rank-1 citations — 2.4× more than any other school. NIST dominates both the “parents” layer (discovery, relocation) and the “proof” layer (results, comparisons): the most complete dominance profile in the market.
The IB belongs to NIST, AP to ISB, SEN to St Andrews, boarding to Harrow. But the whole budget segment is vacant (0% Big 5 presence on fee queries), as are 12 neighbourhood queries and Thai-English bilingual.
Perplexity cites a “Dulwich College Bangkok” that doesn't exist 21 times — spread by SEO aggregators. And RIS racks up 76 mentions without a single rank 1: always cited, never recommended.
Attacking a money query locked by tri-AI consensus head-on is the most expensive path. A mid-tier school can instead become THE reference on a vacant territory — budget, neighbourhood, bilingual, national community — in 90 days: content structured by parental intent, School + FAQPage markup, consistency across site, directories and local press. The three-AI rule applies here too: King's College gets 30 mentions on Gemini vs 4 on ChatGPT — being strong on a single model mechanically caps the score.
83 patient queries — generic, aesthetic, medical tourism, trust — put to four models: ChatGPT, Gemini, Perplexity and Meta AI. 1,068 citations across 47 players; 82 of 83 queries cite at least one clinic.
| # | Player | AIs | KVS |
|---|---|---|---|
| 1 | Bumrungrad International Hospital | 4/4 | 100.0 |
| 2 | Yanhee International Hospital | 4/4 | 79.0 |
| 3 | Kamol Cosmetic Hospital | 4/4 | 78.8 |
| 4 | Wansiri Hospital | 4/4 | 59.8 |
| 5 | ID Clinic Bangkok | 4/4 | 47.0 |
| 6 | Masterpiece Hospital | 3/4 | 28.1 |
Only 17 of 47 players are cited by all four models. The rest play half a market — or a quarter. Bumrungrad, meanwhile, takes 64 first places on generic, trust and medical-tourism queries.
Strong on ChatGPT (707 points, more than Yanhee), Masterpiece is at zero on Gemini. Depending on a single model exposes half your patient flow to one algorithm.
Kamol locks gender-reassignment surgery — rank 1 on all four AIs. Conversely, 7 ghost queries (price, safety, comparisons) cite no clinic at all: the purest content opportunity in the sector.
The hospital giants rule the generic queries; specialist clinics seize the sharp procedures — ID Clinic, Wansiri. For a challenger clinic, the plan of attack fits in two lines: target the 34 blue-ocean aesthetic-medicine queries and the 7 ghost queries (price, safety, comparisons) with procedure pages marked up MedicalClinic + FAQPage and published prices — exactly where no incumbent defends its ground.
35 international-patient questions — general, procedures, success rates, price, legal framework, source markets — put to ChatGPT, Gemini, Perplexity and Meta AI. Full grid: 140 answers coded, 34 clinics scored, not a single query locked.
| # | Clinic | AI | KVS |
|---|---|---|---|
| 1 | Safe Fertility Center | 4/4 | 100.0 |
| 2 | Bumrungrad Fertility & IVF Clinic | 4/4 | 79.4 |
| 3 | Bangkok Hospital Fertility Center | 4/4 | 50.9 |
| 4 | Prime Fertility Center | 4/4 | 41.9 |
| 5 | Jetanin Institute | 4/4 | 41.3 |
| 6 | Superior A.R.T. | 3/4 | 25.4 |
On procedure queries (ICSI, egg freezing, FET…), it's Prime Fertility — a clinic, not a hospital — that leads. When the patient names their treatment, perceived expertise beats the hospital brand.
Across 35 queries, none has the same #1 on three engines or more — even the English-speaking doctor is contested. Four engines = four markets = four chances to swing for a challenger.
All five legal questions (foreigners, single women, surrogacy…) are ghost or near-empty: two cite no clinic, and aggregators (Konkai, PlacidWay) lurk alone. The purest content opening in the sector.
A duopoly rules the generic queries (Safe × Bumrungrad), but 9 of the 12 money queries carry an aggregator in the top 3, and 8 first-place spots remain ownerless. A challenger clinic's plan of attack: take the empty ground (legal block, logistics, destination comparisons) with factual, marked-up pages, then contest the soft-preference money queries where non-giants — Jetanin, Superior, LRC — already lead.
Multi-AI coverage is the most decisive signal in the KVS. A brand cited by a single AI caps at 30-40 points; those above 70 are, without exception, cited by all three.
Optimising for ChatGPT alone mechanically caps a KVS at 40-50. Each model is an independent, cumulative access route — ignoring Gemini or Perplexity means giving up half your audience. Geographically, Paris leads SaaS (40% of leaders), London fintech, Berlin and Munich follow; Italy and the Nordics beyond Sweden are almost absent.
Several companies valued at over €500M appear in none of the 145 queries in the European panel. Three factors explain it — and what replaces them comes down to four signals.
1. Sales-led, not audience-led communication. The AIs don't read your CRM — they read your public articles, interviews and white papers.
2. No Wikipedia page. A core corpus for LLMs. No press coverage → no Wikipedia → no citation.
3. Unstructured content infrastructure. Without JSON-LD (Organization, Service, FAQPage…), content is unreadable to extraction engines.
1. Notable primary sources. A mention in the FT or Les Échos is worth more than a thousand directory backlinks.
2. Complete Schema.org. JSON-LD markup becomes a prerequisite, no longer an option.
3. Cross-source consistency. Website, Wikipedia and LinkedIn must all say the same thing.
4. Density of structured Q&A. Question-and-answer pages are the most extracted.
SEO was a war of links. GEO is a war of sentences. Whoever produces the clearest, most citable, most widely distributed sentences wins.
No black box: the method is reproducible, and that's what makes the score defensible.
Questions typical of a decision-maker, a patient or a parent in the discovery phase, submitted once per AI, in a neutral session — no history, no personalisation.
For each answer: players cited, rank in the list, mention context. 1,057 unique entities across 8 sectors, from Europe to Bangkok.
A rank 1 doesn't weigh like a rank 5, and coverage across all the panel's AIs (three in Europe, four in Bangkok) is the most decisive factor. Hence the final score.
KVS = Σ (CTRrank × weightAI) × (1 + coveragebonus)
Three combined signals: the rank achieved (a CTR weighted by position), multi-AI coverage (a +10% bonus if all the panel's models cite the brand, +3% for high partial coverage), and frequency across the panel. Normalised to 100, by sector.
Traffic. The KVS measures citation, not the click through to your site.
Product quality. Only the models' perception is assessed.
The brand long tail. The scope is deliberately sector-based and reproducible.
The full ranking of 40 brands per sector, the AI-by-AI breakdown, the European sovereignty maps and the six levers that replace SEO. Free, in exchange for your work email.
Methodology, limits, and how to read — then improve — your position.
A proprietary index normalised to 100 that measures a brand's organic visibility in ChatGPT, Gemini and Perplexity. It combines the rank achieved in each answer, multi-AI coverage and the frequency of appearance across a sector panel of queries.
ChatGPT, Gemini and Perplexity concentrate most B2B traffic in Europe: ChatGPT by volume, Gemini through its native integration into Google Workspace, Android and Chrome, and Perplexity through its adoption as a search engine. The Bangkok studies add Meta AI, dominant in South-East Asia via WhatsApp, Instagram and Facebook. Future editions may widen the panel further.
394 queries calibrated with marketing directors, buyers, lawyers, estate agents, doctors and expat families, to reflect real discovery and evaluation scenarios. Spread across eight sectors (SaaS/Tech, Finance, Business law, Paris-Lyon law, Real estate, European health, Bangkok international schools, Bangkok health), in French, English and local scenarios, asked in a neutral session.
No, it complements it. SEO captures transactional intent (“buy X”, “price of Y”); GEO/KVS addresses the upstream phase — discovery, evaluation, comparison — where decision-makers turn to the AIs. The two channels add up.
Serious, no. Urgent, yes. It means that across your sector's queries, the panel AIs never cited your brand — a deficit of structured presence in the public text corpus, not a judgement on your product. It's diagnosable and fixable in 60 to 90 days.
No. The ranking rests strictly on the real answers of the AIs to the 394 panel queries. No brand can buy its position or be excluded. It's this methodological neutrality that gives the index its value.
Three quick wins: implement complete Schema.org JSON-LD markup (Organization, Service, Offer, FAQPage) on your key pages; strengthen your Wikipedia presence via notable press sources; and structure your pages in question-and-answer format, the most extracted by the AIs.
Yes, with attribution: “Source: KAIS Barometer 2026, KaiZen AI Strategy” and a link to this page. The full PDF report includes reusable high-definition charts.
Find out your KVS score in 48h, on the queries that decide your clients.
Test my AI visibility