Yelp Business Contact Scraper

Features

• Multi-domain support: Scrape from 32 different Yelp country domains
• Comprehensive data extraction: Business name, address, phone, website, ratings, reviews, categories, hours, amenities, and photos
• Configurable result limits: Control how many businesses to scrape per search term
• Automatic pagination: Handles multiple pages of search results
• Robust error handling: Graceful handling of missing data and page load failures
• Thread-safe operation: Concurrent crawling with proper state management
• Language-aware selectors: Optimized selectors for different language versions of Yelp

Input Parameters

Parameter	Type	Required	Default	Description
country	string	Yes	United States	The Yelp country domain to scrape from. Select from 32 supported countries.
searchTerms	array	Yes	-	List of keywords to search for (e.g., "pizza", "plumber", "cafe"). Multiple terms will be searched sequentially.
location	string	Yes	-	The city or area to search in (e.g., "New York", "Berlin", "Tokyo").
maxResults	integer	No	100	Maximum number of business listings to scrape per search term. Range: 1-200.

Example Input

{
    "country": "United States",
    "searchTerms": ["pizza", "sushi"],
    "location": "New York",
    "maxResults": 50
}

Input for German Yelp

{
    "country": "Germany",
    "searchTerms": ["restaurant", "cafe"],
    "location": "Berlin",
    "maxResults": 100
}

Input for Japanese Yelp

{
    "country": "Japan",
    "searchTerms": ["restaurant"],
    "location": "Tokyo",
    "maxResults": 25
}

Output Format

Each scraped business is stored as a JSON object with the following fields:

Field	Type	Description
01_scrapedAt	string	ISO 8601 timestamp of when the data was scraped
02_url	string	Full URL of the business detail page
03_country	string	Country/domain the business was scraped from
04_searchTerm	string	The search term used to find this business
05_searchLocation	string	The location used in the search
06_name	string	Business name
07_rating	string	Star rating (e.g., "4.5 star rating")
08_reviewCount	string	Number of reviews
09_address	string	Full business address
10_phone	string	Phone number
11_website	string	Business website URL
12_priceRange	string	Price range indicator
13_categories	array	List of business categories
14_businessHours	array	Operating hours for each day
15_amenitiesAndMore	array	List of amenities and features
16_photoUrls	array	URLs of business photos (up to 10)
17_reviews	array	Recent reviews with author, rating, date, and text

Sample Output

{
    "01_scrapedAt": "2024-01-15T10:30:00.000Z",
    "02_url": "https://www.yelp.com/biz/joes-pizza-new-york",
    "03_country": "United States",
    "04_searchTerm": "pizza",
    "05_searchLocation": "New York",
    "06_name": "Joe's Pizza",
    "07_rating": "4.5 star rating",
    "08_reviewCount": "2,847 reviews",
    "09_address": "7 Carmine St, New York, NY 10014",
    "10_phone": "(212) 366-1182",
    "11_website": "https://www.joespizzanyc.com",
    "12_priceRange": "$",
    "13_categories": ["Pizza", "Italian"],
    "14_businessHours": [
        "Mon: 10:00 AM - 4:00 AM",
        "Tue: 10:00 AM - 4:00 AM",
        "Wed: 10:00 AM - 4:00 AM",
        "Thu: 10:00 AM - 4:00 AM",
        "Fri: 10:00 AM - 5:00 AM",
        "Sat: 10:00 AM - 5:00 AM",
        "Sun: 10:00 AM - 4:00 AM"
    ],
    "15_amenitiesAndMore": ["Casual", "Good for Groups", "Take Out"],
    "16_photoUrls": [
        "https://s3-media0.fl.yelpcdn.com/bphoto/example1.jpg",
        "https://s3-media0.fl.yelpcdn.com/bphoto/example2.jpg"
    ],
    "17_reviews": [
        {
            "author": "John D.",
            "rating": "5 star rating",
            "date": "1/10/2024",
            "text": "Best pizza in NYC! The classic slice is perfection."
        }
    ]
}

Supported Yelp Domains

The actor supports the following 32 Yelp country domains:

Country	Domain
Argentina	yelp.com.ar
Australia	yelp.com.au
Austria	yelp.at
Belgium	yelp.be
Brazil	yelp.com.br
Canada	yelp.ca
Chile	yelp.cl
Czech Republic	yelp.cz
Denmark	yelp.dk
Finland	yelp.fi
France	yelp.fr
Germany	yelp.de
Hong Kong	yelp.com.hk
Ireland	yelp.ie
Italy	yelp.it
Japan	yelp.co.jp
Malaysia	yelp.my
Mexico	yelp.com.mx
Netherlands	yelp.nl
New Zealand	yelp.co.nz
Norway	yelp.no
Philippines	yelp.com.ph
Poland	yelp.pl
Portugal	yelp.pt
Singapore	yelp.com.sg
Spain	yelp.es
Sweden	yelp.se
Switzerland	yelp.ch
Taiwan	yelp.com.tw
Turkey	yelp.com.tr
United Kingdom	yelp.co.uk
United States	yelp.com

Technical Details

Technology Stack

• Runtime: Node.js
• Web Scraping Framework: Crawlee with Playwright
• Browser: Chromium (headless mode)
• Platform: Apify

Architecture

The actor uses a two-phase scraping approach:

1. LIST phase: Scrapes search result pages to collect business URLs
2. DETAIL phase: Visits each business page to extract comprehensive data

Configuration

• Navigation timeout: 120 seconds
• Request handler timeout: 180 seconds
• Maximum concurrency: 5 parallel requests
• Maximum retries: 3 per request
• Headless browser with anti-detection measures

Anti-Detection Features

• Custom HTTP headers mimicking real browser requests
• Disabled automation detection flags
• Proxy support through Apify proxy configuration

Cost Estimation

The cost depends on the number of businesses scraped and the complexity of the pages:

• Approximate compute units per 100 businesses: 0.5-1.0 CU
• Each search term generates additional requests for pagination
• Detail pages require more resources than list pages

For a typical run scraping 100 businesses with 2 search terms:

• Estimated compute units: 1-2 CU
• Estimated time: 5-15 minutes

Limitations

1. Maximum results per search term: 200 (Yelp pagination limit)
2. Rate limiting: Yelp may temporarily block requests if too many are made in a short period
3. Dynamic content: Some business information may not be available if loaded via JavaScript after initial page load
4. Regional availability: Not all businesses have complete information across all fields
5. Review limit: Only the first 5 reviews are extracted per business
6. Photo limit: Maximum of 10 photos are extracted per business

Error Handling

The actor handles errors gracefully:

• Missing selectors: Falls back to alternative selectors
• Page load failures: Retries up to 3 times
• Missing data: Fields are set to null if not found
• Failed business pages: Error is logged and scraping continues with next business

Usage Tips

1. Start with a small maxResults value (10-20) to test your search parameters
2. Use specific search terms for better results
3. Combine multiple search terms in a single run for efficiency
4. Monitor the run logs for any warnings about blocked requests