⭐ AI-Powered German Impressum Scraper – B2B Legal Data Extraction (Fast Cheerio)

An ultra-fast, AI-enhanced B2B domain enrichment tool engineered specifically to discover, parse, and structure Impressum (Legal Notice / Imprint) data from websites across Germany 🇩🇪, Austria 🇦🇹, Switzerland 🇨🇭, and international stores.

By combining lightweight, high-concurrency Cheerio HTML parsing with an advanced LLM processing engine (Groq), this scraper cuts through website noise to deliver clean, structured company, owner, contact, and statutory compliance data.

---

🔥 Why Choose This Scraper?

• ⚡ Extreme Speed (Cheerio-Powered): Runs without heavy browser overhead (Puppeteer/Playwright), processing 100+ domains in just 2–3 minutes with minimal compute cost.
• 👔 100% Accurate Decision Maker Detection (AI Mode): Pinpoints Managing Directors (Geschäftsführer, Vorstand) for corporations and Business Owners (Inhaber) for sole proprietorships (Einzelunternehmen) with zero false positives.
• 🤖 Dual-Engine Architecture: Toggle between ultra-fast Regex pattern matching for contact/legal IDs and Premium AI parsing for decision makers and complex layouts.
• 🔗 Advanced Multi-Tier Link Discovery: Discovers imprint links across footers, sub-footers, attribute links (button[data-href], [data-url], [data-link]), inline onclick redirects, and <template> elements.
• 🎯 Intelligent Route Fallback: If a homepage lacks a clear footer link (common on Shopify and headless stores), the scraper sequentially probes high-probability candidate routes: /policies/legal-notice, /pages/impressum, /impressum, /imprint, and /rechtliches.
• 🛡️ Clean Data Filters: Built-in safeguards reject payment badge noise (American Express, Google Pay), SVG class artifacts (VR 3Lq), and promotional banner collisions.
• ⏱️ Zero-Hang Architecture: Features a built-in real-time watchdog that immediately terminates the container the second all input domains finish—preventing idle compute costs.
• 🎁 Zero API Keys Required: We cover all AI token costs! You do not need your own Groq or OpenAI API key.

---

💰 Pricing & Pay-Per-Event (PPE) Model

You only pay for successful domain extractions. If an Impressum page cannot be found or fails to load, you are charged $0.00.

Extraction Mode	Base Price / Domain	Bronze Tier (-10%)	Silver Tier (-15%)	Gold Tier (-20%)
Standard (Regex)	$0.0030	$0.0027	$0.0025	$0.0024
Premium (AI Enabled)	$0.0045	$0.0040	$0.0038	$0.0036

💵 Cost Calculation Examples (Standard Rate)

URLs Processed	Regex Mode Cost	AI Enabled Cost
100 Domains	$0.30	$0.45
200 Domains	$0.60	$0.90
500 Domains	$1.50	$2.25
1,000 Domains	$3.00	$4.50

---

⚙️ Input Configuration

Parameter	Type	Required	Default	Description
startUrls	Array	Yes	["marley.de"]	List of target websites or domains (with or without https:// / www.).
enableAi	Boolean	No	false	When true, activates Groq LLMs for verified Decision Maker and legal entity extraction. When false, uses ultra-fast Regex.
aiModel	String	No	openai/gpt-oss-20b	Model selection: openai/gpt-oss-20b (Recommended), openai/gpt-oss-120b, qwen/qwen3.6-27b, groq/compound.
maxConcurrency	Number	No	10	Concurrency limit (1–25) to ensure stability and prevent rate limits.

---

🏢 Output Data Schema

Field Name	Type	Description	Example
originalUrl	String	Target website requested	https://monoka.de
impressumUrl	String	Discovered Impressum / Legal Notice URL	https://monoka.de/pages/impressum
companyName	String	Legal company name or registered store brand	Monoka
decisionMakerName	String | null	Extracted Managing Director(s) or Business Owner (Populated via AI Mode)	Patrick Hauser
decisionMakerRole	String | null	Legal German role title (Geschäftsführer, Inhaber, etc.) (Populated via AI Mode)	Inhaber
address	String | null	Full physical street address, postal code, and city	In der Breite 67, 79224 Umkirch, DE
emails	Array	List of verified contact email addresses	["info@monoka.de"]
phones	Array	Clean international telephone numbers	["+495031530"]
registerCourt	String | null	Official Commercial Register Court (Amtsgericht)	Amtsgericht Oldenburg
registerNumber	String | null	Commercial register ID (HRB, HRA, VR)	HRB 209987
taxId	String | null	Official European VAT ID (USt-IdNr.)	DE340943599
socialLinks	Object	Extracted social media profile URLs	Facebook, Instagram, LinkedIn, YouTube, TikTok, X
metadata	Object	Page Title, Meta Description, and H1 Header	SEO metadata of the website

---

🧩 Examples: AI Mode vs. Regex Mode

Example 1: Premium AI Mode (enableAi: true)

Best for: Complete B2B enrichment where you need verified Decision Makers (Geschäftsführer, Inhaber), sole proprietorship stores, or complex legal paragraphs.

Input:

{
  "startUrls": [
    "https://monoka.de/",
    "https://www.dr-johanna-budwig.de/"
  ],
  "enableAi": true,
  "aiModel": "openai/gpt-oss-20b",
  "maxConcurrency": 10
}

Output:

[
  {
    "originalUrl": "https://monoka.de/",
    "impressumUrl": "https://monoka.de/pages/impressum",
    "companyName": "Monoka",
    "decisionMakerName": "Patrick Hauser",
    "decisionMakerRole": "Inhaber",
    "address": "In der Breite 67, 79224 Umkirch, Deutschland",
    "emails": [
      "info@monoka.de"
    ],
    "phones": [],
    "registerCourt": null,
    "registerNumber": null,
    "taxId": "DE340943599",
    "socialLinks": {
      "instagram": "https://www.instagram.com/monoka.de",
      "youtube": "https://www.youtube.com/@Monoka./videos",
      "tiktok": "https://www.tiktok.com/@monoka.de"
    },
    "metadata": {
      "title": "Produkte für dein Traum Dekolleté - für jede Form und Größe",
      "description": "Trägerlos, Rückenfrei oder eine größere Oberweite...",
      "h1": "Monoka"
    }
  },
  {
    "originalUrl": "https://www.dr-johanna-budwig.de/",
    "impressumUrl": "https://www.dr-johanna-budwig.de/policies/legal-notice",
    "companyName": "Dr. Johanna Budwig GmbH",
    "decisionMakerName": "Franz Nowotny",
    "decisionMakerRole": "Geschäftsführer",
    "address": "An den Kolonaten 2-4, 26160 Bad Zwischenahn, Deutschland",
    "emails": [
      "kontakt@dr-johanna-budwig.de"
    ],
    "phones": [
      "+494413906300"
    ],
    "registerCourt": "Amtsgericht Oldenburg",
    "registerNumber": "HRB 209987",
    "taxId": "DE300469959",
    "socialLinks": {
      "facebook": "https://www.facebook.com/Dr.Johanna.Budwig",
      "instagram": "https://www.instagram.com/dr.johannabudwig",
      "youtube": "https://www.youtube.com/user/DrJohannaBudwig",
      "tiktok": "https://www.tiktok.com/@dr.johannabudwig"
    },
    "metadata": {
      "title": "Dr. Johanna Budwig",
      "description": "Dr. Johanna Budwig",
      "h1": "Frühstücksöl Limette & Orange"
    }
  }
]

---

Example 2: Ultra-Fast Standard Mode (enableAi: false)

Best for: High-volume bulk runs where maximum extraction speed and lowest cost per lead are your top priorities. Contact details, addresses, and statutory IDs are extracted via hardened Regex.

Input:

{
  "startUrls": [
    "https://marley.de",
    "https://www.dr-johanna-budwig.de/"
  ],
  "enableAi": false,
  "maxConcurrency": 10
}

Output:

[
  {
    "originalUrl": "https://marley.de",
    "impressumUrl": "https://marley.de/policies/legal-notice",
    "companyName": "Marley Deutschland GmbH",
    "decisionMakerName": null,
    "decisionMakerRole": null,
    "address": "Adolf-Oesterheld-Str. 28, 31515 Wunstorf, Deutschland",
    "emails": [
      "info@marley.de"
    ],
    "phones": [
      "+495031530"
    ],
    "registerCourt": "Amtsgericht Hannover",
    "registerNumber": "HRB 110009",
    "taxId": "DE811119134",
    "socialLinks": {
      "facebook": "https://www.facebook.com/MarleyDeutschland",
      "instagram": "https://www.instagram.com/marley_deutschland",
      "linkedin": "https://www.linkedin.com/company/marleydeutschland",
      "youtube": "https://www.youtube.com/@MarleyDeutschland",
      "tiktok": "https://www.tiktok.com/@marleydeutschland"
    },
    "metadata": {
      "title": "Marley Deutschland GmbH",
      "description": "— Marley Deutschland GmbH",
      "h1": "Impressum"
    }
  },
  {
    "originalUrl": "https://www.dr-johanna-budwig.de/",
    "impressumUrl": "https://www.dr-johanna-budwig.de/policies/legal-notice",
    "companyName": "Dr. Johanna Budwig GmbH",
    "decisionMakerName": null,
    "decisionMakerRole": null,
    "address": "An den Kolonaten 2-4, 26160 Bad Zwischenahn, Deutschland",
    "emails": [
      "kontakt@dr-johanna-budwig.de"
    ],
    "phones": [
      "+494413906300"
    ],
    "registerCourt": "Amtsgericht Oldenburg",
    "registerNumber": "HRB 209987",
    "taxId": "DE300469959",
    "socialLinks": {
      "facebook": "https://www.facebook.com/Dr.Johanna.Budwig",
      "instagram": "https://www.instagram.com/dr.johannabudwig",
      "youtube": "https://www.youtube.com/user/DrJohannaBudwig",
      "tiktok": "https://www.tiktok.com/@dr.johannabudwig"
    },
    "metadata": {
      "title": "Dr. Johanna Budwig",
      "description": "Dr. Johanna Budwig",
      "h1": "Frühstücksöl Limette & Orange"
    }
  }
]

(Note: To prevent false positives from payment tokens and navigation menus, decisionMakerName and decisionMakerRole are strictly extracted when enableAi: true is enabled).

---

📥 Export & Integrations

Export your structured German B2B data with a single click in any preferred format:

• Formats: CSV, Excel (.xlsx), JSON, XML
• Integrations: Direct API, Webhooks, Google Sheets, Make (Integromat), and Zapier

---

⚡ Heavy JavaScript & SPA Sites?

While Cheerio is exceptionally fast, some modern websites require a real headless browser to execute heavy client-side JavaScript (e.g. Next.js, Nuxt, React SPAs, Web Components, or Shadow DOM menus).

If you are scraping heavy, dynamic JavaScript portals, check out our full browser sibling actor:
👉 German Imprint Scraper (Playwright & AI)

---

❓ Frequently Asked Questions (FAQs)

1. Do I need my own Groq or OpenAI API key?
No! The actor includes load-balanced Groq API integration in the backend. You only pay the flat Apify Pay-Per-Event rate.

2. How does the Pay-Per-Event (PPE) model work?
You are billed only when a domain is successfully enriched. If a website is offline, blocks crawlers, or has no detectable Impressum page, you are charged $0.00 for that domain.

3. Why are Decision Makers only extracted in AI Mode?
German Impressum pages format person names in hundreds of different ways—ranging from sole proprietorship owner listings (Patrick Hauser under Shop-Hauser) to multi-member corporate boards. Regex frequently confuses navigation links, payment badges (American Express), or disclaimer text with person names. By delegating person identification exclusively to AI, we guarantee 100% precision.

4. How does the scraper discover hidden Shopify or JS links?
The scraper inspects HTML attribute links (button[data-href], [data-url]), inline onclick scripts, and <template> tags. If no link is present in the static markup, it automatically probes standard legal routes (/policies/legal-notice, /pages/impressum, /impressum, /imprint, /rechtliches).

5. Can I provide raw domains like marley.de without https://?
Yes. The actor automatically normalizes all input formats (e.g. marley.de, www.marley.de, http://marley.de) into valid HTTPS requests.

---

🧑‍💻 Developer Info

• Author: codescraper
• Email: codescraper011@gmail.com
• Platform: Apify
• Engine: TypeScript (@crawlee/cheerio)