French Website to SIREN & Mentions Légales Extractor

Give it a list of French websites or domains, get back the company behind each one: SIREN, SIRET, VAT number, RCS city, share capital, emails and phones read from the site's mentions légales, then verified against official SIRENE data (legal name, registered address, NAF code, headcount, size category, latest revenue). You only pay for websites matched to a company.

French law (LCEN, art. 6) requires company websites to publish their legal identity, so for most SME and B2B sites the SIREN is right there; this actor finds the right page, extracts it and validates it.

Good for: enriching lead lists that only have a website (Apollo, Clay, Google Maps or event exports), matching domains to SIREN for a CRM, KYB / supplier onboarding, and preparing lists for French Company Data: SIRENE enrichment or French Company Financials & Ratios.

How it works

1. Loads the homepage (https, falls back to http) and respects robots.txt.
2. Follows the most likely legal links: mentions légales, informations légales, legal notice, CGV/CGU, à propos, contact, plus common paths such as /mentions-legales.
3. Extracts SIREN / SIRET (only when written next to SIREN, SIRET, RCS or immatriculée), French VAT numbers, RCS city, share capital, emails, phones. Every number is checked with the Luhn algorithm and VAT keys are recomputed, so random 9-digit numbers are not returned.
4. Picks the best SIREN (legal pages and VAT matches weigh more), then looks it up in the official Recherche d'entreprises API (INSEE SIRENE + RNE).
5. Crawls politely: one page per second per site, at most 6 pages by default, and stops as soon as a legal page with a SIREN is found.

Input example

{
  "websites": ["pennylane.com", "payfit.com", "swile.co", "qonto.com", "danone.com/fr"],
  "enrichWithSirene": true,
  "maxPagesPerSite": 6,
  "maxConcurrency": 3,
  "respectRobotsTxt": true
}

Output example

{
  "input": "swile.co",
  "domain": "swile.co",
  "status": "found",
  "siren": "824012173",
  "legalName": "SWILE",
  "vatNumber": "FR67824012173",
  "rcsCity": "Montpellier",
  "shareCapital": "90 402,50",
  "emails": [
    "hello@swile.fr"
  ],
  "phones": [
    "0974594217"
  ],
  "legalPageUrl": "https://www.swile.co/legal/company-details",
  "registryAddress": "IMMEUBLE L'ALTIS BATIMENT A @ 7CENTER 561 RUE GEORGES MELIES 34000 MONTPELLIER",
  "postalCode": "34000",
  "city": "MONTPELLIER",
  "nafCode": "66.19B",
  "legalFormCode": "5710",
  "companyCategory": "ETI",
  "headcountRange": "41",
  "creationDate": "2016-11-24",
  "isActive": true,
  "sirenCandidates": [
    "824012173"
  ],
  "sireneMatch": true
}

Status values

Status	Meaning	Charged?
found	SIREN found on the site and confirmed in SIRENE	Yes
found_not_in_sirene	Valid SIREN found but not in the public SIRENE base (or enrichment off)	No
found_sirene_unavailable	SIREN found on the site, but the government API was down after retries, so it could not be verified	No
not_found	Site crawled, no SIREN published (emails/phones still returned)	No
blocked	The site refused the request (403/429, bot protection). Try the proxy option	No
robots_disallowed	robots.txt forbids crawling	No
unreachable / invalid_url / error	DNS, TLS or timeout problems	No

Pricing

Pay per event: one charge per website matched to a SIREN confirmed in SIRENE (found). Every other status is free. Example: 1,000 domains with a 60 % match rate = 600 charged events (at $0.005 per match that is $3). See the Pricing tab for current prices.

Use with Make, Zapier, n8n, Clay or Google Sheets

• Make / n8n / Zapier: Apify app → Run an Actor (atesen-software/french-website-siren-mentions-legales) with {"websites": ["{{domain}}"]}, then Get dataset items and map siren, legalName, vatNumber to your CRM.
• API: POST https://api.apify.com/v2/acts/atesen-software~french-website-siren-mentions-legales/run-sync-get-dataset-items?token=YOUR_TOKEN with the JSON input.
• Clay / Sheets: export as CSV, or call the run-sync endpoint from an HTTP column.

FAQ

Which sites work best? French SMEs, startups, B2B companies and e-shops, which nearly always publish mentions légales. Very large consumer sites behind Cloudflare/DataDome may return blocked from cloud IPs; try a residential proxy.

Do you guess the SIREN from the company name? No. The SIREN must be published on the website; there is no fuzzy name guessing, so false matches are rare. Several candidates are listed in sirenCandidates.

Personal data? The actor returns contact emails/phones published on the site's legal and contact pages, which may include a named person's address. Process them under a lawful basis (GDPR art. 6) and honour opt-outs. Director names are not extracted.

Is this a Pappers / societe.com alternative? For domain → company matching, yes; the enrichment uses official open data (INSEE SIRENE, RNE) via the French government API.

This Actor is not affiliated with INSEE, INPI, the French government, Pappers or societe.com.