Google Maps Extractor

🗺️ Google Maps Extractor

Turn a keyword, a city, or a Google Maps link into a dataset of public places: name, address, phone, website, rating, and hours when Maps publishes them.

Built for lead lists, local research, and location ops that need the same search again next week, not a one-off export.

✨ Features

• 🔎 Search by keyword plus a location, or pass place_id:ChIJ... when you already know the place.
• 🔗 Accept Google Maps search links, place links, cid links, maps.app.goo.gl links, and goo.gl/maps links. A URL that is not one of those fails the run and is not stored. A short link that does not open a public place fails the run the same way.
• 📍 Center a search on a city, country, postal code, or a GeoJSON point. locationQuery is kept only when that text appears on the place address. A polygon or multipolygon keeps places that fall inside that shape.
• ⭐ Filter by name match, category, minimum stars, whether a website is listed, and permanently or temporarily closed places.
• 📄 Optional place preview (scrapePlaceDetailPage) for fields that show up on the place page. A failed preview keeps the values already on the search card.
• 🌐 Language follows the language code. A two-letter countryCode is sent with the search.
• 🚫 Empty fields stay off the item. A missing closed flag is not written as false.

🚀 Why teams use this

Request-based fetches, not a browser session per place, so a production run spends its time on Maps responses. You get a normal Scrapely dataset: schedule it, call it from the API, and export. Every stored row is a place Maps or the business's own public page actually returned. A start URL that is not Google Maps fails the run, and no error row is written. A search that returns no places exits with an empty dataset.

💡 Use cases

• Pull every business for a keyword and city into a CRM or spreadsheet.
• Size a local market: count places per category, rating, and neighbourhood.
• Enrich an existing list of places with Maps coordinates, phones, and websites.
• Track new or closed places for an area over time.

🧪 Example input

{
  "language": "en",
  "locationQuery": "New York, USA",
  "maxCrawledPlacesPerSearch": 20,
  "scrapeSocialMediaProfiles": {
    "facebooks": false,
    "instagrams": false,
    "tiktoks": false,
    "twitters": false,
    "youtubes": false
  },
  "searchStringsArray": [
    "restaurant"
  ]
}

🧾 Sample output

One item from a real run of the example input, shortened. Long text is cut and lists show their first entries.

{
  "title": "Gramercy Tavern",
  "placeId": "ChIJvSQIgqFZwokRFYQbJdzceSs",
  "address": "42 E 20th St, New York, NY 10003, United States",
  "categoryName": "New American restaurant",
  "categories": [
    "New American restaurant"
  ],
  "totalScore": 4.6,
  "reviewsCount": 5067,
  "location": {
    "lat": 40.7384555,
    "lng": -73.98850639999999
  },
  "street": "42 E 20th St",
  "city": "New York",
  "postalCode": "10003",
  "state": "New York",
  "neighborhood": "Manhattan",
  "countryCode": "US",
  "phone": "+1 555 0100",
  "phones": [
    "+1 555 0100"
  ],
  "phoneUnformatted": "+1 555 0100",
  "website": "https://www.gramercytavern.com/",
  "openingHours": [
    {
      "day": "Monday",
      "hours": "11:30 am to 11 pm"
    },
    {
      "day": "Tuesday",
      "hours": "11:30 am to 11 pm"
    }
  ],
  "imageUrl": "https://lh3.googleusercontent.com/…",
  "imagesCount": 10041,
  "fid": "0x89c259a1820824bd:0x2b79dcdc251b8415",
  "cid": "3132777853886366741",
  "businessProfileId": "18235747991152489841",
  "url": "https://www.google.com/maps/search/?api=1&query=Gramercy%20Tavern&query_place_id=ChIJvSQIgqFZwokRFYQbJdzceSs"
}

📥 Input

The fields that change what this actor collects:

maxQuestions is reserved: Google Maps no longer shows questions and answers, so it returns nothing and does not affect the run. Leads, verification and social profiles never fail the run when a site has nothing to show.

{
  "searchStringsArray": ["restaurant"],
  "locationQuery": "Konstanz",
  "maxCrawledPlacesPerSearch": 8,
  "language": "en",
  "searchMatching": "all",
  "website": "allPlaces",
  "skipClosedPlaces": false,
  "scrapePlaceDetailPage": false
}

📤 Output

Each place is one dataset item. A key is present only when the Maps payload had a value. Depending on the place, that can include:

title, placeId, address, neighborhood, street, city, state, postalCode, countryCode, location, plusCode, floor, categoryName, categories, price, totalScore, reviewsCount, reviewsDistribution, reviewsTags, phone, phones, phoneUnformatted, website, menu, openingHours, additionalOpeningHours, popularTimesHistogram, popularTimesLiveText, popularTimesLivePercent, imageUrl, imagesCount, imageCategories, fid, cid, kgmid, businessProfileId, url, searchPageUrl, searchString, language, rank, scrapedAt, description, ownerDescription, subTitle, additionalInfo, claimThisBusiness, reserveTableUrl, permanentlyClosed, temporarilyClosed, wasOpenAtScrapeTime.

A hotel place can also include hotelStars, hotelDescription, checkInDate, checkOutDate, hotelAds (provider title, the price string Maps printed, googleUrl, and isOfficialSite when Maps labels that row the official site), and similarHotelsNearby. A gas station can include gasPrices with the fuel name, the price tag, and the number from that tag. popularTimesHistogram uses Su through Sa, each with hour and occupancyPercent.

When the matching add-on is on and the page has the value, an item can also include tableReservationLinks, tableReservationProviders (name and reserveTableUrl), emails, facebooks, instagrams, youtubes, tiktoks, twitters, webResults (title, url, displayedUrl, description), orderOnline, locatedIn, and parentPlaceUrl.

inputStartUrl is set for rows that came from startUrls. inputPlaceId is set for a direct place id. openingHours is a list of {day, hours}. location is {lat, lng}. Rank is the kept-place order, including directory rows.

💰 Pricing

You pay per event, plus platform usage (compute and proxy traffic).

Add-ons are charged only when you turn the option on and only for places where it returned data.

Example: 1,000 places with no add-ons cost $4.40 in event fees plus platform usage. In our test runs platform usage for a plain search was about $0.08 to $0.16 per 1,000 places. The default 20-place search costs $0.088 in event fees.

If you set a maximum cost per run, the run saves as many places as that budget pays for and then stops cleanly.