ORIAS Intermediary Scraper

Bulk-extract intermediary data from the French ORIAS register — the official register of insurance, banking, and finance intermediaries. Two modes: scrape specific profiles by SIREN, or automatically discover and enrich all intermediaries in one or more categories (COA, CIF, AGA, MIA, COBSP, and more).

Built for: Lead generation · Market research · Compliance checks · CRM enrichment · Insurance & finance prospecting

---

What does ORIAS Intermediary Scraper do?

The ORIAS Intermediary Scraper extracts publicly available data from ORIAS — the Organisme pour le Registre des Intermédiaires en Assurance, Banque et Finance, under the supervision of the French Treasury. All insurance, banking, and finance intermediaries operating in France must be registered.

The Actor works in two modes:

Mode	How it works	Best for
urls	Provide specific SIRENs or ORIAS URLs	Enrich a known list
categories	Select categories → Actor discovers all SIRENs automatically, then scrapes every profile	Full register extraction

In category mode, the Actor uses two optimised discovery strategies:

• CSV download (COA, CIF, COBSP): ORIAS publishes official CSV exports — the Actor downloads and parses them to get up to 27,000 active SIRENs in seconds
• Paginated search (AGA, MA, MAL, MIA, MOBSP, MOBSPL, MIOBSP): the Actor submits the ORIAS advanced search form and paginates through all results using session-based navigation

---

What data does it extract?

Category	Fields
Identity	Company name, commercial name (enseigne)
Contact	Email, phone, website, full address
Address	Street address, city, zipcode, country
Registration	ORIAS number, registration status
Legal	SIREN, legal form, RCS, NAF code
Associations	Linked companies with SIREN, name, address, profile URL

Fields marked "non renseigné" (not provided) on ORIAS are returned as null.

---

Input

Mode 1 — Scrape specific SIRENs / URLs

{
  "mode": "urls",
  "startUrls": [
    "https://orias.fr/home/showIntermediaire/000032630",
    "483472817"
  ],
  "maxConcurrency": 5,
  "delayBetweenRequests": 1
}

startUrls accepts full ORIAS profile URLs or raw SIREN numbers (auto-padded to 9 digits).

---

Mode 2 — Discover and scrape by category

{
  "mode": "categories",
  "categories": ["COA", "MIA", "COBSP"],
  "maxConcurrency": 5,
  "delayBetweenRequests": 1
}

Available categories:

Group	Code	Full name	Discovery method	~Active count
Insurance	AGA	Agent général d'assurance	Paginated search	~12
Insurance	COA	Courtier d'assurance ou de réassurance	CSV download	~27,000
Insurance	MA	Mandataire d'assurance	Paginated search	~14
Insurance	MAL	Mandataire d'assurance lié	Paginated search	~1
Insurance	MIA	Mandataire d'intermédiaire d'assurance	Paginated search	~240
Banking	COBSP	Courtier en opérations de banque	CSV download	~5,800
Banking	MOBSP	Mandataire non-exclusif OBP	Paginated search	~53
Banking	MOBSPL	Mandataire exclusif OBP	Paginated search	~2
Banking	MIOBSP	Mandataire d'intermédiaires OBP	Paginated search	~60
Finance	CIF	Conseiller en investissement financier	CSV download	~7,700
Finance	ALPSI	Agent lié de PSI	Paginated search	—
Finance	IFP	Intermédiaire en financement participatif	Paginated search	—

To scrape the entire register:

{
  "mode": "categories",
  "categories": ["COA", "CIF", "COBSP", "AGA", "MA", "MAL", "MIA", "MOBSP", "MOBSPL", "MIOBSP"]
}

→ ~41,000 intermediaries discovered and enriched automatically.

---

All input parameters

Parameter	Type	Default	Description
mode	String	"urls"	"urls" or "categories"
startUrls	Array	—	[mode=urls] ORIAS URLs or SIREN numbers
categories	Array	—	[mode=categories] Category codes to enumerate
maxRequestsPerCrawl	Integer	0	Max profiles to scrape (0 = unlimited)
maxConcurrency	Integer	5	Parallel requests (1–10)
delayBetweenRequests	Integer	1	Seconds between requests (recommended: 1–3)

---

Output example

{
  "url": "https://orias.fr/home/showIntermediaire/000032630",
  "name": "ANTOINE BOUTARD",
  "website": null,
  "email": "antoine.boutard.courtage@laposte.net",
  "phone": "0241597249",
  "phoneE164": "+33241597249",
  "address": "5 rue des Fontaines",
  "city": "DOUE EN ANJOU",
  "zipcode": "49700",
  "country": "France",
  "addressFull": "5 rue des Fontaines 49700 DOUE EN ANJOU France",
  "commercialName": null,
  "status": "Personne physique",
  "registrationStatus": "Inscrit 1",
  "oriasNumber": "23004891",
  "siren": "000 032 630",
  "legalForm": "Entrepreneur individuel",
  "nafCode": "6622Z",
  "rcs": "ANGERS",
  "associations": [
    {
      "siren": "908 554 702",
      "name": "VOTRASSO",
      "address": "33 rue de la République",
      "city": "LYON 02",
      "zipcode": "69002",
      "country": "France",
      "url": "https://orias.fr/home/showIntermediaire/908554702"
    }
  ]
}

Download in JSON, CSV, Excel, or HTML from the Dataset tab.

---

How to use

Step-by-step (category mode)

1. Open the Actor Input tab
2. Set Mode to categories
3. Add your target categories in Categories (e.g. COA, MIA)
4. Optionally set Max concurrency and Delay between requests
5. Click Start — the Actor discovers all SIRENs automatically, then scrapes each profile
6. Download results from the Dataset tab

Step-by-step (URL mode)

1. Search on orias.fr for the intermediaries you want
2. Copy the profile URLs or SIREN numbers
3. Paste into the SIREN numbers or ORIAS URLs field
4. Click Start

---

Local development

$npm install

Edit input.json at the project root, then run:

$npm run run:local

Or manually:

cp input.json storage/key_value_stores/default/INPUT.json
apify run --purge-none

Example input.json for category mode:

{
  "mode": "categories",
  "categories": ["AGA", "MIA"],
  "maxConcurrency": 3,
  "delayBetweenRequests": 1
}

Unit tests

npm test              # All tests
npm run test:scraper  # Scraper logic tests
npm run test:input    # Input validation tests

---

Timeout & Resurrection

For large crawls (30k+ URLs), the run can exceed the default timeout. The Actor uses a Request Queue that persists across runs, so you can resurrect a timed-out run to continue where it left off.

If the Actor times out (~11k/31k processed):

1. Go to the Runs tab → find your stopped run (status: Timed out)
2. Click Actions → Resurrect
3. Important: Before or when resurrecting, increase the Timeout in Input → Run options (e.g. 21600 = 6 hours, 28800 = 8 hours)
4. The Actor will resume from the remaining URLs — no re-discovery, no duplicate work

The default timeout in actor.json is 6 hours. If your plan limits runs to 1 hour, use resurrection to complete in multiple chunks.

---

Performance and cost

Scenario	Discovery	Scraping	Total time (est.)
1 SIREN (urls mode)	—	<5s	<5s
AGA (12 records)	~2s search	~30s	~35s
MIA (240 records)	~20s search	~8 min	~9 min
COA (27,000 records)	~30s CSV	~15 h	~15 h
Full register (~41,000)	~2 min	~22 h	~23 h

Estimates at maxConcurrency: 5, delayBetweenRequests: 1. ORIAS pages are lightweight HTML.

Cost uses Apify consumption-based pricing (Compute Units). Check Apify pricing for current rates.

---

Use cases

• Lead generation: Build complete lists of insurance brokers, financial advisors, or banking intermediaries by category
• Compliance: Verify ORIAS registration status, SIREN, and details at scale
• Market research: Analyze the full register by category, region, legal form, or NAF code
• CRM enrichment: Add ORIAS contact and registration data to existing records
• Due diligence: Cross-check SIREN and ORIAS numbers across an entire segment

---

Is it legal to scrape ORIAS?

Our scrapers extract only publicly available data from the ORIAS website — information that is openly displayed for regulatory transparency. ORIAS exists precisely so that anyone can verify intermediary registration.

Results may contain personal data (names, emails, addresses). Personal data is protected by GDPR. Use scraped data only for legitimate purposes and ensure compliance with applicable laws.

See our blog post on the legality of web scraping.

---

FAQ

Can I use SIREN numbers instead of full URLs?

Yes — in urls mode the Actor accepts raw SIREN numbers (1–9 digits) and auto-converts them to the correct 9-digit padded URL.

Does the Actor use an official ORIAS API?

No. ORIAS does not provide a public API for intermediary data. The Actor uses two approaches: official CSV exports published by ORIAS for bulk categories, and the public search form for others.

How many intermediaries can I scrape?

Unlimited — set maxRequestsPerCrawl to 0. The full French register has ~41,000 active intermediaries across all categories.

How do I scrape all COA brokers?

{ "mode": "categories", "categories": ["COA"] }

The Actor downloads the official COA CSV (~27,000 active SIRENs) and scrapes every profile.

What if results look incomplete?

ORIAS intermediaries are not required to fill all fields. Missing values are returned as null. The data reflects what ORIAS publicly displays.

How do I get help?

Use the Issues tab on the Actor page to report bugs or request features.

---

Support

For questions, custom solutions, or feedback, use the Issues tab on the Actor page.

---

Ready? Set mode: "categories", pick your categories, and click Start.