Das Örtliche Scraper

Scrape business listings from dasoertliche.de, Germany's phone directory.

Features

• JSON-LD extraction — Reliable structured data from ItemList with LocalBusiness entries
• 25 results per page with automatic pagination
• Phone numbers — Direct from JSON-LD structured data
• Geo coordinates — Latitude/longitude for each business
• Optional detail pages — Website URLs, email, fax, opening hours, logo, ratings, social links
• HTTP-only — Low memory (~128MB), fast execution
• Pay-per-result pricing via Apify billing events

Input

Field	Type	Default	Description
keyword	string	required	Business type (e.g., "Zahnarzt", "Restaurant")
city	string	required	City to search in (e.g., "Berlin", "München")
maxResults	number	100	Maximum results to scrape
includeDetails	boolean	false	Fetch detail pages for website/email/hours
proxyConfiguration	object	Apify proxy	Proxy settings

Output Fields

From SERP (always available)

• name — Business name
• phone — Phone number
• address — Full address
• street, city, postal_code, region — Address components
• latitude, longitude — Geo coordinates
• detail_url — Link to detail page

From Detail Page (when includeDetails=true)

• website — Business website URL
• email — Email address
• fax — Fax number
• opening_hours — Business hours
• additional_phones — Extra phone numbers
• logo_url — Business logo image URL (from Kundenlogo or itemprop image)
• ratings — Aggregated web ratings as key-value pairs, e.g. { "Google": "4.3/5 aus 3 Bewertungen" } (platforms: Google, Jameda, Sanego, Yelp, golocal, Kununu, ProvenExpert, etc.)
• social_links — Social media profile URLs as key-value pairs, e.g. { "facebook": "https://..." } (facebook, instagram, xing, linkedin, youtube)

Pricing Events

Event	Price
directory-start	$0.05
directory-serp-result	$0.003
directory-detail-result	$0.005

Enrichment add-ons

After the scrape completes, this actor can automatically trigger AI-powered extraction on every website found in the results. Each add-on runs as a separate actor and produces its own dataset.

Contact extraction

Extracts team member names, email addresses, phone numbers, positions, and departments from company websites. Powered by the Website Contact Extractor.

Enable it by setting enableContactExtraction: true and providing at least one LLM API key. The sub-actor run ID is stored in the key-value store as CONTACT_EXTRACTOR_RUN_ID.

Job listing extraction

Extracts open positions, job titles, locations, departments, and career page URLs from company websites. Powered by the Website Job Extractor.

Enable it by setting enableJobExtraction: true and providing at least one LLM API key. The sub-actor run ID is stored in the key-value store as JOB_EXTRACTOR_RUN_ID.

Browser fallback

Some company websites are built with JavaScript frameworks (React, Vue, Angular) that require a full browser to render. When enableBrowserFallback is set to true, the contact/job extractors will automatically re-scrape these sites with Playwright. This catches ~25% more sites but increases cost and runtime. Only applies when contact or job extraction is enabled.

LLM API keys

Both add-ons use LLMs to extract structured data. Provide one or more API keys. When multiple keys are provided, the system uses them in priority order with automatic fallback:

1. Gemini (recommended) -- Best quality-to-cost ratio. Free tier includes 1M tokens/min. Get a key at Google AI Studio.
2. Groq (optional) -- Ultra-fast inference. Get a key at Groq Console.
3. OpenRouter (optional) -- Access to 100+ models. Get a key at OpenRouter.

One key is sufficient. With multiple keys, if the primary provider hits a rate limit, the system falls back to the next available provider automatically.

Support

Found a bug or have a feature request? Open an issue.