France pairs a thin ownership layer with an aggressive acquisition market. On the confirmed-active .fr business book, hosting-relationship concentration — hosting competitors only, with CDN, SaaS-builder and DNS-infrastructure actors excluded — is an HHI of 1,984, about 5 effective owners of the customer relationship, and the single largest control-plane actor alone accounts for 35.0% of the book. What that ownership does not buy is a quiet front door. The market word is hébergement.
“Concentration” is three different questions, and they have three different answers. Hosting-relationship HHI (1,984) measures economic market power — who owns the customer relationship — counting hosting competitors only. Hosting-infrastructure HHI (1,588) measures where the workloads actually run. Control-plane HHI (1,586) measures the combined ecosystem, and deliberately includes CDN, SaaS-builder and DNS-infrastructure actors alongside hosting groups. For a market-power reading — who can raise prices — the first is the one that counts, and it is the number this page leads with.
Read as market power, France runs on about 5 effective owners of the customer relationship; read as infrastructure, the same book spreads across about 6.3 effective hosting infrastructures — a 1.3× gap between who runs the servers and who holds the contract: the market is thinner at the contract than at the rack. Shares follow the same split: the top three owners hold 63.9% of customer relationships, while the top three infrastructures carry 51.8% of the workloads, and on the wider control plane — which also counts CDN, SaaS-builder and DNS actors — the top three come to 56.1%. Three denominators, three numbers; quoting one against another is the error these pages are built to avoid.
Holding the installed base is not the same as owning the next customer. 4 ownership groups hold page-one positions on the head search term, 88% of tracked storefronts acquire on teaser pricing, and the price the average French customer meets (€1.86/mo) sits below the storefront median (€2.50) — the largest platforms are buying acquisition, not harvesting it.
All three readings are drawn from the confirmed-active .fr business book (1,836,200 classified domains) and are ownership-folded, which makes them a lower bound. They do not share one base: only the control-plane reading is measured on that full figure, while the infrastructure and relationship readings are each renormalised over their own actor set, so their shares must never be read against it. How the bands and the three readings work.
Across the 530,466 confirmed-active French business sites we HTTP-enrich and fingerprint: 28.3% carry e-commerce tooling, 44.1% run analytics, and 5.5% sit on a detected website builder. Off the web layer, 9.5% of the full book runs modern hosted email.
Presence, not billing; the builder share is a floor. How technology detection works.
Of the 1,695,269 mail-carrying business domains, 89.6% publish SPF but only 32.6% do so with a spoof-rejecting hard fail; 47.3% publish DMARC, of which 7.5% enforce with p=reject and 32.0% publish monitoring-only p=none. Adoption is not enforcement — the gap between publishing a record and actually rejecting forged mail is where the risk lives.
Across 8 tracked French storefronts, the median entry-tier price is €2.50/mo ex-VAT. Weight the same storefronts by how many businesses each platform actually serves and the price the average French customer meets is €1.86/mo — below the storefront median — the biggest platforms here price below the field.
— a floor, since storefronts that never disclose a renewal price sit in the denominator.
Fewer than five tracked French storefronts publish a measurable renewal price, so we do not state a measured market renewal cliff — a median over three storefronts would be an anecdote, not a statistic. Renewal visibility is better than the storefront count suggests: storefronts serving 60.1% of captured French customer relationships publish their renewal price on the page. At storefront grain — one vote per storefront, the 1 that display one plain recurring price with no teaser mechanics counted at face value (1.0×) — the structure-adjusted median is 4.5× across 5 storefronts. That is a merchandising statistic, not the customer picture: flat-displaying storefronts are often numerous but small.
The tracked set represents 68.7% of France's control-plane customer relationships. How prices are captured, normalised and suppressed.
French businesses search hébergement, and the storefronts answer in French: mutualisé, nom de domaine, hébergeur. An English-language page competing on “hosting” is not in this market's vocabulary at all.
Measured from French storefronts, sized by how many providers use each term.
The two terms buyers search most: hébergement / hébergeur. How this is measured.
Every figure here is queryable through the HostingBrain connector. In Claude or any MCP-compatible assistant, this is the whole page in one prompt:
“Using HostingBrain, describe the France hosting market: how concentrated is it — give me the actor-pure hosting-relationship HHI and effective owners for market power, and the hosting-infrastructure and control-plane readings beside it, ownership-folded — how large is the confirmed-active business book, what do businesses run (CMS / e-commerce / builder adoption), what does entry shared hosting cost and how steep are renewal cliffs, and what is the SPF/DMARC posture? Keep it aggregate-level.”
Resolves to market_concentration, market_digitization,
technology_adoption, market_pricing and
email_security (all free tier). See definitions('hhi').
Reproduce this analysis: market_concentration,
market_digitization, technology_adoption, market_pricing and
email_security return every figure above for France and every other European market — on
the free tier. Named provider shares and price ladders: market_structure,
pricing_profile (Pro).
Ask the follow-up yourself. HostingBrain answers questions like this — with the date, denominator and caveats attached — inside Claude and any MCP-compatible assistant.