Google Maps Business Scraper

Find real business records from public Google Maps pages. Use a text search to discover places, or provide specific Google Maps place URLs for direct extraction. The actor runs cookielessly in a fresh browser context and streams each record to the dataset as soon as it is verified.

What this actor does

• Searches Google Maps for a query, optionally combined with a location.
• Scrapes one or more public place URLs directly.
• Extracts visible business details such as name, category, address, phone, website, rating, review count, hours, coordinates, place ID, and photos when Google exposes them.
• Deduplicates records using an observed place ID or canonical Maps URL.
• Retries transient page failures with a bounded optional proxy fallback.
• Snapshots public result-card fields before detail navigation; if a detail page is slow or unavailable, emits that observed card as a conservative fallback instead of losing the result or inventing detail-only fields.
• Omits unavailable fields instead of returning null placeholders, guessed values, or fabricated businesses.

Output per record

Each business row normally includes:

Field	Meaning
recordType	business
searchTerm	The search that found this business, so batch runs stay traceable
name	Business name shown by Google Maps
categoryName	Visible category information, when available
address	Public address
phone	Public phone number, when shown
website	Public website, when shown
totalScore	Visible rating, when rated
reviewsCount	Visible review count
priceLevel	Visible price information, when shown
plusCode	Public Plus Code, when shown
location	Observed latitude and longitude
latitude / longitude	The same observed coordinates as plain columns, for spreadsheets and CRM imports
placeId	Canonical Google Place ID (the ChIJ... token), when Google exposes one
featureId	Legacy public Google Maps feature ID (the 0x...:0x... token), when Google exposes one. It is emitted separately and is never relabelled as a placeId
kgmid	Observed Knowledge Graph identifier (the /g/... token in the canonical URL)
locatedIn	Parent place when Google shows "Located in:" — useful for shops inside malls, stations, and buildings
placeAttributes	Google's own published attribute chips, exactly as shown, e.g. LGBTQ+ friendly, Check-in time: 3:00 PM, Check-out time: 12:00 PM
hasStreetView / streetViewCount	Set when Google exposes a Street View control for the place
fuelPrices	Fuel grades and their published prices for a gas station, e.g. [{"grade": "Regular", "price": "$4.79"}]. Only a grade with a directly observed price is emitted; stale marks a price Google flags as over 24 hours old.
hotelStars / hotelStarRating	Star class as Google labels it, plus the numeric value
checkInTime / checkOutTime	Published check-in and check-out times
priceOfferAmount / priceOfferCurrency	The nightly price Google currently shows, and its currency
priceOfferDateRange	The dates that price applies to
priceOfferFreeCancellation	Directly observed, localized free-cancellation note from the target hotel's booking offer; filter labels such as "Free cancellation only" are not emitted
hotelAmenities	Named amenities with whether Google marks each one available, e.g. [{"name": "Pool", "available": true}]; localized badge text is preserved exactly as Google displays it, including explicit unavailable entries
description	Editorial description visible on the place page
hoursStatus / hours	Visible business hours, when available
isOpenNow	Live open/closed state Google shows for the place, when available
ratingHistogram	Visible rating distribution, when available
serviceOptions	Visible service-option labels, such as dine-in, takeout, or delivery
amenities	Visible header amenity labels; localized text and Google's availability wording are preserved exactly, and the field is omitted when that source surface is unavailable
menuLink	Public menu link, when shown
bookingLinks	Public booking, order, or reservation links, when shown
photosSample	Public image URLs visible on the place page
photoCount	Number of photos Google reports for the place, when available
popularTimesSummary	Published popular-times text, when Google exposes it
claimed	Set to false only when Google itself shows a "Claim this business" control
permanentlyClosed / temporarilyClosed	Set to true only when Google itself marks the place permanently or temporarily closed
googleMapsUrl	Canonical or observed Maps URL
sourceUrl	URL used as source context
scrapedAt	UTC scrape timestamp

Fields are conditional. A business without a website, phone, rating, hours, or photos simply omits that field.

Input

Field	Type	Description
mode	enum	search (default), placeUrls, or placeIds
searchQuery	string	Search terms such as coffee shops in Brooklyn; used in search mode
location	string	Optional location appended to the query, such as London, UK
placeUrls	list of strings or URL objects	Full public Google Maps place URLs; required in placeUrls mode. Common copied quotes, tracking parameters, country domains, and short links are normalized.
placeIds	list of strings	Canonical Google Place IDs beginning with ChIJ; required in placeIds mode and resolved through Google's public URL API. Feature IDs/CIDs are not accepted as Place IDs.
maxResults	integer, 1–100000	Requested ceiling per search. Records are emitted until the public feed ends, the ceiling for that search is reached, or the bounded runtime deadline expires. With a batch, every search gets its own allowance, so one busy term cannot starve the others.
searchTerms	list of strings	Run several search terms in one go, for example ["coffee shops", "dentists"]. Each term is searched in every location you provide.
locations	list of strings	Run the search across several places in one go, for example ["New York, NY", "Austin, TX"].
minRating	number, 1–5	Keep only businesses rated at or above this value. A business with no published rating does not qualify, so it is skipped.
maxRating	number, 1–5	Keep only businesses rated at or below this value.
minReviews	integer	Keep only businesses with at least this many reviews.
maxReviews	integer	Keep only businesses with at most this many reviews.
openNow	boolean	Keep only businesses Google currently shows as open.
hasWebsite	boolean	Keep only businesses that publish a website on their listing.
hasPhone	boolean	Keep only businesses that publish a phone number on their listing.
nameFilter	string	Keep only businesses whose name contains this text.
excludeCategories	list of strings	Skip businesses whose category contains any of these words.
includeClosed	boolean	When off, permanently closed businesses are skipped. On by default.
language	enum	Maps interface language; defaults to English
proxyConfiguration	object	Optional proxy override. By default the actor starts with Apify Residential (US); direct access is only a bounded fallback if the Residential route cannot be set up or does not return a usable page.

Examples

Example: search for local businesses

{
  "mode": "search",
  "searchQuery": "coffee shops",
  "location": "Brooklyn, NY",
  "maxResults": 20,
  "language": "en"
}

Example: many search terms across several cities, filtered

{
  "mode": "search",
  "searchTerms": ["coffee shops", "bakeries"],
  "locations": ["Austin, TX", "Dallas, TX"],
  "minRating": 4.0,
  "minReviews": 25,
  "hasWebsite": true,
  "maxResults": 50
}

This searches all four term-and-city combinations in one run, keeps only well-reviewed businesses that publish a website, and drops anything that does not qualify.

Example: search internationally

{
  "mode": "search",
  "searchQuery": "寿司",
  "location": "東京 日本",
  "maxResults": 10,
  "language": "ja"
}

Example: scrape specific place URLs

{
  "mode": "placeUrls",
  "placeUrls": [
    "https://www.google.com/maps/place/GRIND+THE+NYC+COFFEE+SHOP+%26+BAGEL+HOUSE/@40.7590839,-73.9918101,17z/data=!3m1!4m6!3m5!1s0x89c259125a931087:0x3d471d4575240620!8m2!3d40.7590839!4d-73.9918101!16s%2Fg%2F11kgnrjx7t"
  ],
  "maxResults": 1
}

Example: scrape canonical Place IDs

{
  "mode": "placeIds",
  "placeIds": ["ChIJabtoT7hlwokRLECK0SW3_8I"],
  "maxResults": 1
}

Use cases

• Local market and competitor research.
• Business directory and location-data projects.
• Prospecting by category and geography.
• Store or branch discovery from a controlled list of place URLs.
• Enriching a dataset with public ratings, hours, websites, and coordinates.

Cookieless and high-volume behavior

No Google login, cookies, browser profile, or authenticated session is required. The actor starts with direct public access and uses bounded retries; a blocked or unavailable page is not converted into a fake empty business. If Google serves its intermittent limited-view shell while preserving the target's contact anchors, the actor tries up to two fresh anonymous browser contexts for both direct place URLs and search-detail hydration. A complete render wins without merging fields across contexts; if every attempt remains degraded, the richest single observed row is retained with the upstream limited-view condition.

Search mode snapshots each public result card before opening detail pages. A successfully hydrated detail record is preferred; when the detail budget is exhausted or navigation fails, the actor may emit a card-level record containing only fields visibly present on that card (for example name, rating, review count, category, price, description, address, and the observed Maps URL). Detail-only fields such as phone, website, Place ID, coordinates, hours, photos, and service options are omitted from that fallback row unless they were obtained from the place page. This keeps result coverage and field provenance explicit.

Records are pushed progressively, so memory does not grow with the complete result set. Direct URL mode accepts large batches up to the schema limit. A single text search is still limited by the public result feed Google exposes; use the Google Maps Area Scanner for systematic geographic coverage beyond one search window.

FAQ

Are all fields always present? No. Google publishes different fields for different businesses and regions. Unavailable values are omitted.

Does the actor invent ratings, review counts, or identifiers? No. Values are emitted only when observed on the public page or canonical URL.

Can I scrape more than 120 search results? You can request up to 100,000. Google commonly exposes around 120 or fewer cards in one public feed; the actor forwards the requested ceiling and follows the feed until source exhaustion or its bounded runtime deadline. For systematic geographic coverage, use the Area Scanner or provide a large placeUrls list in direct mode.

What happens with duplicate URLs? Records are deduplicated using an observed place ID or canonical URL; a name-and-address fallback is used only when no stable identifier is exposed.

What happens if Google limits the page? The actor keeps already-emitted records and reports an honest terminal status. It never pads the dataset to reach maxResults.

Related Google Maps actors

Actor	Use it for
Google Maps Reviews Scraper	Reviews, reviewer details, photos, and owner responses
Google Maps Photos Scraper	Place photo URLs and visible photo metadata
Google Maps Business Hours Scraper	Regular and special business hours
Google Maps Popular Times Scraper	Directly published popular-times values
Google Maps Email Extractor	Public website email and social-link enrichment
Google Maps Area Scanner	Bounding-box, circle, and polygon coverage
Google Maps Leads Scraper	Search plus optional website contact enrichment
Google Maps MCP Server	Search and review tools for AI assistants
Google Maps Directions Scraper	Public route and travel-mode information
Google Maps Geocoding Scraper	Address and coordinate lookup
Google Maps Similar Places Scraper	Related-place discovery
Google Maps Menu Scraper	Public menu items and prices
Google Maps Nearby Scraper	Places near a coordinate
Google Maps Place List Scraper	Curated lists and list/search URLs
Google Maps Timezone Scraper	Timezone data for coordinates

Use Google Maps data responsibly and follow applicable terms and laws.