2GIS Places Scraper

2GIS Places Scraper — Addresses, Contacts and Ratings

2GIS Places Scraper searches 2GIS — the leading maps and business directory across Russia, the CIS, Central Asia and the Middle East — and returns every matching place as structured JSON: name, address, GPS coordinates, rubrics/category, rating and review counts, opening hours, and (opt-in) phone/email/website/social contacts, reviews and photos. No parsing, no HTML, no 2GIS account. Give it a search term and a city, hit Run, and get a ready-to-use dataset.

What is 2GIS Places Scraper?

2GIS Places Scraper is an Apify Actor that runs real search queries against 2GIS's live site (e.g. "ресторан" in "Москва") and extracts a normalized record for every business that comes back, across 19 country domains and 13 interface languages. It does not require a 2GIS account, login, or API key — it reads the same public pages a site visitor sees, through a headless Chromium browser with automatic proxy escalation. It's built for lead-generation teams, local-market researchers, directory builders and developers who need 2GIS business data as clean JSON instead of scraping HTML by hand.

What 2GIS place data is publicly available to scrape?

Everything a visitor sees on a 2GIS place card and search results page — name, category, address, coordinates, rating, reviews and listed contact details — is visible without signing in. What's gated is content 2GIS only serves inside its own authenticated app session or to the business owner managing the listing.

Data Category	Publicly Available	Restricted
Name, rubrics/category, brand info	Yes	—
Address, postcode, GPS coordinates	Yes	—
Rating score, review count	Yes	—
Review text, ratings, dates, business's official replies	Yes	—
Phone, email, website, social links shown on the card	Yes	—
Opening hours	Yes	—
Photo album cover image + total photo count	Yes	—
Every individual photo inside a gallery beyond each album's cover	—	Authenticated 2GIS app session only
Owner analytics / lead-management dashboard	—	Business-owner login

2GIS Places Scraper only returns publicly visible data — what any visitor sees. Nothing behind a login wall.

What data can I extract with 2GIS Places Scraper?

Each place record combines identity and location fields with rating, contact, media and review data, matching the exact JSON keys the Actor pushes to the dataset — not just the default table view.

📍 Identity, address and category fields

Field Name	Description
id	2GIS branch/firm numeric ID
title	Place name
shortName	Short display name
extension	Legal-entity suffix (e.g. "ООО")
url	Canonical 2GIS place page URL
rubrics	Array of 2GIS category names assigned to the place
category	Raw poi_category value from 2GIS, when present
summary	Best available description — ad tagline, full description, description or annotation, in that priority order
postcode, country, region, city, district	Address components
street, houseNumber	Parsed street and house number
address	Full formatted address string
location	{ lat, lng } — real scraped coordinates
working_hours	Per weekday (Mon–Sun), each { from, to } for the first schedule interval of the day
nearestStations	Array of { name, distance, comment } — nearby metro/transport stops
attributeGroups	Flat list of listed amenities/attributes (e.g. "Wi-Fi", card payment)
brand	{ name, shortName, extension, branchCount, rating: { score, reviews } } — chain/brand-level info
detailsType	2GIS's own entity type for the record (always "branch" — non-business entities are filtered out before this point)
mainPhotoUrl	Cover photo URL of the primary photo album

⭐ Ratings, contacts and review fields

Field Name	Description
totalScore	Place's overall rating
reviewsCount	Total review count
ratingCount	Number of reviews that carry a star rating
phoneText, phoneValue	Phone numbers as displayed / as raw dialable values — only when includeContacts: true
email, website	Contact email(s) and website(s) — only when includeContacts: true
socials	{ vk, telegram, whatsapp }, each an array — only when includeContacts: true
mediaCount	2GIS's real total photo count across all albums — only when maxMediaPerPlace > 0
media	Array of { url, count }, one entry per photo album cover, capped at maxMediaPerPlace
reviewsMeta	{ branchId, branchRating, branchReviewsCount, totalCount, ratedCount, unratedCount, fetchedCount, providers, sortBy, ratings, locale, sentimentFilter } — present only when maxReviewsPerPlace > 0
reviews	Array of review objects (see below) — present only when maxReviewsPerPlace > 0

Each entry in reviews[] carries id, provider, rating, text, dateCreated, dateEdited, isRated, likesCount, commentsCount, officialAnswer ({ id, dateCreated, orgName, text, logoUrl } or null), emojis, trustFactors, and — only when includeReviewerData: true — a user object with { id, publicId, name, reviewsCount, photoUrl }.

🕓 Run metadata fields

Field Name	Description
searchString	The query term this row was found under
searchUrl	The 2GIS search URL used
pageNumber	Search-result page the place was discovered on
scrapedAt	ISO-8601 UTC timestamp of extraction

🤖 Add-on: Need additional 2GIS data?

Reviews, reviewer profiles and photo albums are already built into this Actor as opt-in toggles (maxReviewsPerPlace, includeReviewerData, maxMediaPerPlace) rather than separate scrapers — turn them on in one run instead of chaining tools. If you're also tracking the same locations on other platforms, pair this with Google Maps Reviews Scraper or Facebook Reviews Scraper to cross-reference reputation across map providers.

How does 2GIS Places Scraper differ from the official 2GIS API?

2GIS publishes an official Places API that returns only id, name and type by default — every additional field (address, contacts, hours, structure info) is opted into via a fields parameter, and some of it requires sales-team approval. Access starts with a demo key from 2GIS's Platform Manager, capped at 1,000 requests per service for one month; production use means purchasing a subscription whose price is negotiated directly with a 2GIS account manager, not published.

Feature	2GIS Places API	2GIS Places Scraper
Access	API key from Platform Manager, demo or paid subscription	Runs on Apify — no 2GIS account or key
Default response	id, name, type only	Full place record (identity, address, rating, hours) by default
Contacts, reviews, photos	Some fields gated behind sales approval	Toggled by input parameters (includeContacts, maxReviewsPerPlace, maxMediaPerPlace)
Usage limits	Demo key: 1,000 requests/service/month; paid tiers by subscription	Bound by the Actor's own crawl of maxItems per search term
Pricing	Negotiated with a 2GIS account manager	Apify's platform pricing for this Actor
Setup time	Sign up, create a key, negotiate a subscription	Run the Actor with a JSON input

Use the official API when you need a direct, long-term commercial data relationship with 2GIS. Use 2GIS Places Scraper when you want structured place data immediately, without a 2GIS account or a sales conversation.

How to use 2GIS Places Scraper

2GIS Places Scraper runs on the Apify platform — no separate signup with 2GIS.

1. Open the Actor's page in Apify Store and click Try for free (or Run, if you already have an Apify account).
2. Provide query — one or more search phrases, exactly as you'd type them into the 2GIS search box (this is the only required input).
3. Set locationQuery for the city/area, and optionally narrow results with rubricIds, the filterHas*/filterCreatedRecently/filterWifi toggles, filterAvgPriceMin/filterAvgPriceMax, or filterRating.
4. Click Start to launch the run.
5. Download results from the Output tab as JSON, CSV, Excel, HTML, XML or RSS, or stream them via the Apify API.

How to scale to bulk place extraction

query accepts an array, so one run can search several terms — each term's results are collected and saved as its own batch once that term finishes. maxItems applies per search term, not per run, so ["ресторан", "кафе", "бар"] with maxItems: 50 can return up to 150 places in a single run. There's no separate URL-list input — this Actor works from search terms and filters, not a pre-built list of place links.

What can you do with 2GIS place data?

• A lead-generation agency building call lists for a city uses phoneText and website to reach every business in a rubric that filters on filterHasSite: true.
• A local SEO consultant benchmarking a niche uses totalScore and reviewsCount across every competitor returned by one query to size the competitive field.
• A franchise analyst mapping a chain's footprint uses brand.branchCount and brand.rating.score to compare a brand's locations city by city.
• A reputation manager uses reviews[].rating alongside reviews[].officialAnswer to check which negative reviews a business has (or hasn't) responded to.
• An AI agent or RAG pipeline indexes summary, rubrics, address and reviews[].text as retrievable context for a local-business chatbot or lead-qualification agent, with location.lat/location.lng for distance-based ranking.

How does 2GIS Places Scraper handle rate limits and blocking?

Requests start direct (no proxy). On a hard failure, or a confirmed empty result on page 1, the search phase escalates to a datacenter proxy and then a residential proxy, sticking with residential once it succeeds. A homepage visit runs before each search/detail browser context to establish a normal-looking session, since 2GIS's anti-bot layer reacts to a cold, referrer-less deep link into /search. The detail-page pool (which loads individual place pages) has its own, independent escalation ladder — if every place in a batch fails on the tier that worked for search, that pool is rebuilt on a fresh context at the next proxy tier and retried, up to twice. Each place page also gets up to 3 retry attempts before being given up on.

⚠️ 2GIS runs an active countermeasure against automated traffic. This Actor gets through it under typical conditions, but on a heavily-loaded IP range some individual place pages within a run can still be lost even after escalation — those are logged and simply skipped, never fabricated.

⬇️ Input

query is the only required field. Everything else narrows or enriches the search.

Parameter	Required	Type	Default	Description
query	Yes	array of strings	—	One or more search phrases, like in the 2GIS search box (e.g. ресторан, стоматология)
locationQuery	No	string	"Москва"	City, district, street or landmark. Ignored if customGeolocation is set
maxItems	No	integer	10	Max places per search term. 0 tries to take all available (internally capped at 500 per term)
domain	No	string (enum, 20 values incl. auto)	"auto"	Which national 2GIS site to use — 2gis.ru, 2gis.kz, 2gis.ae, 2gis.uz and 15 more; auto defaults to Russia
language	No	string (enum, 14 values incl. auto)	"auto"	Preferred interface language for the scraped page
includeContacts	No	boolean	false	Also load phones, email, website and socials when shown on the card
maxReviewsPerPlace	No	integer (0–99999)	0	Max reviews per place. 0 skips reviews entirely (faster)
reviewSortBy	No	string (enum: trust, date_edited)	"trust"	How fetched reviews are ordered
reviewRatings	No	string (enum: all, negative, positive)	"all"	Keep only reviews of the chosen sentiment (negative ≤2, positive ≥4)
includeUnratedReviews	No	boolean	false	Include text-only reviews with no star rating
includeReviewerData	No	boolean	false	Store the reviewer's public profile fields (name, avatar). Personal data — collect only with a lawful basis
maxMediaPerPlace	No	integer (0–99999)	0	When >0, adds mediaCount and up to this many photo-album cover URLs per place
rubricIds	No	array of strings (select, 1,706 categories)	[]	Narrow results to selected 2GIS category codes
filterHasGoods	No	boolean	false	Only places with a priced menu/catalogue
filterHomeDelivery	No	boolean	false	Only venues offering delivery
filterTakeaway	No	boolean	false	Only places offering takeaway
filterHasSite	No	boolean	false	Only listings with a website
filterHasPhotos	No	boolean	false	Only places with photos
filterPaymentCard	No	boolean	false	Only places accepting card payment
filterAvgPriceMin	No	integer	—	Minimum average bill (food-service places)
filterAvgPriceMax	No	integer	—	Maximum average bill (food-service places)
filterCreatedRecently	No	boolean	false	Only recently-added places
filterWifi	No	boolean	false	Only places with Wi-Fi
filterRating	No	string (enum, 5 star-tier tokens)	—	Keep places at or above the selected rating tier
sortBy	No	string (enum: rating, opened_time, name)	"rating"	How saved records are ordered, applied Actor-side to the already-scraped data (2GIS's site ignores a sort URL parameter). opened_time keeps the original crawl order — 2GIS does not publish a place's opening date
customGeolocation	No	object (GeoJSON Polygon/MultiPolygon)	—	After scraping, keep only places whose real coordinates fall inside this shape. Overrides locationQuery as the geographic filter
enableGlobalDataset	No	boolean	false	Remember place IDs saved in previous runs of this Actor (persisted store) and skip them instead of re-saving duplicates

Example input

{
  "query": ["стоматология", "автомойка"],
  "locationQuery": "Новосибирск",
  "maxItems": 40,
  "domain": "auto",
  "language": "auto",
  "includeContacts": true,
  "maxReviewsPerPlace": 25,
  "reviewSortBy": "trust",
  "reviewRatings": "negative",
  "includeUnratedReviews": false,
  "includeReviewerData": false,
  "maxMediaPerPlace": 2,
  "rubricIds": [],
  "filterHasSite": true,
  "filterPaymentCard": true,
  "sortBy": "rating",
  "enableGlobalDataset": false
}

⬆️ Output

Every place is pushed to the Actor's dataset as one typed JSON object, with a consistent field set across runs. Download it from the Output tab as JSON, CSV, Excel, HTML, XML or RSS, or pull it with the Apify API/SDK.

Example output

{
  "id": "70000001007179642",
  "title": "Example Café",
  "shortName": "Example",
  "extension": null,
  "url": "https://2gis.ru/moscow/firm/70000001007179642",
  "rubrics": ["Кафе"],
  "category": null,
  "totalScore": 4.6,
  "reviewsCount": 312,
  "ratingCount": 198,
  "summary": "Кофе, завтраки, веранда",
  "postcode": "125009",
  "country": "Россия",
  "region": "Москва",
  "city": "Москва",
  "district": "Тверской",
  "street": "Тверская",
  "houseNumber": "7",
  "address": "Тверская, 7",
  "mainPhotoUrl": "https://i5.photo.2gis.com/main/branch/32/70000001007179642/common",
  "mediaCount": 180,
  "media": [
    { "url": "https://i5.photo.2gis.com/main/branch/32/70000001007179642/common", "count": 167 },
    { "url": "https://i6.photo.2gis.com/main/geo/32/70000001007179642/view", "count": 13 }
  ],
  "location": { "lat": 55.7654, "lng": 37.6055 },
  "working_hours": { "Mon": { "from": "08:00", "to": "22:00" } },
  "nearestStations": [{ "name": "Тверская", "distance": 210, "comment": "5 мин пешком" }],
  "attributeGroups": ["Wi-Fi", "Принимают карты"],
  "brand": {
    "name": "Example",
    "shortName": "Example",
    "extension": null,
    "branchCount": 4,
    "rating": { "score": 4.6, "reviews": 312 }
  },
  "detailsType": "branch",
  "searchString": "кафе",
  "searchUrl": "https://2gis.ru/search/кафе?m=37.385264%2C55.581699%2F9",
  "pageNumber": 1,
  "scrapedAt": "2026-08-12T10:00:00Z",
  "phoneText": ["+7 495 123-45-67"],
  "phoneValue": ["+74951234567"],
  "email": [],
  "website": ["http://example-cafe.ru"],
  "socials": { "vk": ["example_cafe"] },
  "reviewsMeta": {
    "branchId": "70000001007179642",
    "branchRating": 4.6,
    "branchReviewsCount": 312,
    "totalCount": 312,
    "ratedCount": 8,
    "unratedCount": 0,
    "fetchedCount": 8,
    "providers": [{ "tag": "2gis", "isReviewable": true }],
    "sortBy": "trust",
    "ratings": "rated_only",
    "locale": "ru_RU",
    "sentimentFilter": "negative"
  },
  "reviews": [
    {
      "id": "600123456789",
      "provider": "2gis",
      "rating": 2,
      "text": "Долго ждали заказ",
      "dateCreated": "2026-06-01T10:00:00Z",
      "dateEdited": null,
      "isRated": true,
      "likesCount": 1,
      "commentsCount": 0,
      "officialAnswer": null,
      "emojis": null,
      "trustFactors": null
    }
  ]
}

phoneText, phoneValue, email, website, socials appear only when includeContacts is true. mediaCount/media appear only when maxMediaPerPlace > 0. reviewsMeta/reviews appear only when maxReviewsPerPlace > 0 and a review API key was found for that place.

How does it work?

2GIS Places Scraper drives a real Chromium browser (Playwright) to each 2GIS search results page and place page, the same pages a visitor's browser would load. It reads the page's own client-side state (window.initialState) plus DOM and HTML fallbacks to collect place IDs, then opens each place page to pull the full record. A homepage visit warms up every new browser session before the real request, and connections escalate from direct to datacenter to residential proxy tiers when 2GIS's anti-bot layer pushes back. Only data visible on 2GIS's own public pages is returned — nothing behind a login. The output schema (the JSON keys documented above) stays the same run over run, even if 2GIS changes its page layout.

Integrations

2GIS Places Scraper runs on Apify, so it works with anything that can call the Apify API or connect to Apify's MCP server.

Calling 2GIS Places Scraper programmatically

from apify_client import ApifyClient

client = ApifyClient("<APIFY_API_TOKEN>")

run = client.actor("scraper-engine/2gis-places-scraper").call(run_input={
    "query": ["ресторан"],
    "locationQuery": "Москва",
    "maxItems": 20,
})

for place in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(place["title"], place["address"])

Works in Go, Ruby, Node.js, cURL — any language that can make an HTTP request, using an Authorization: Bearer <APIFY_API_TOKEN> header against the Apify API.

MCP integration for AI agents

2GIS Places Scraper is reachable through Apify's MCP server — add it as a tool from the server's configurator (search for the Actor by name) and it becomes callable from any MCP client, including Claude Desktop, Cursor and Claude Code. Authenticate with an Apify API token or OAuth; no separate hosting needed.

No-code tools (n8n, Make, LangChain)

In n8n, use the HTTP Request node pointed at this Actor's run endpoint (or the community Apify node) to trigger a run and fetch the dataset. In Make, the Apify app module lets you start a run and iterate results as part of a scenario. In LangChain, use ApifyDatasetLoader (or the Apify integration wrapper) to load a finished run's dataset directly into a document pipeline.

Is it legal to scrape 2GIS places?

Scraping publicly available business listing data — the kind 2GIS Places Scraper returns by default — is generally permitted, since it does not bypass a login or access control. 2GIS Places Scraper only returns data visible on public 2GIS pages. Because the default output is business/place data, not personal data, no personal-data regime (like GDPR) applies to a standard run; the relevant framework instead is 2GIS's own Terms of Service and applicable database-rights law, which you should review for your specific use.

⚠️ The one opt-in exception is includeReviewerData: turning it on stores a reviewer's public profile fields (name, avatar), which is personal data and does fall under GDPR/CCPA-style regimes — collect it only where you have a lawful basis. Consult legal counsel if your use case involves bulk storage of personal data.

Frequently asked questions

What 2GIS place fields does 2GIS Places Scraper return?

The top fields are title, address, location (GPS), totalScore/reviewsCount, and — when enabled — phoneText/website and reviews[]. See What data can I extract above for the full field list.

Does 2GIS Places Scraper require a 2GIS account or login?

No. It scrapes public 2GIS pages through a browser session, with no 2GIS account, password or API key needed on your side.

How many places can I extract in one run?

maxItems applies per search term, defaulting to 10; setting it to 0 tries for all available results, internally capped at 500 per term. Since query accepts multiple terms, one run can return up to maxItems × number of terms places.

What happens if a search query returns zero results?

If a search term returns nothing on page 1, the Actor treats it as a possible soft block and escalates the proxy tier (up to twice) before accepting a genuine zero-result outcome and moving to the next query term — no fabricated rows are added either way.

Can I scrape multiple 2GIS places at once?

Yes — query is an array, so one run processes several search terms, each saved as its own batch of results once that term's scrape finishes.

Does 2GIS Places Scraper work with Claude, ChatGPT and other AI agent tools?

Yes. It's reachable through Apify's MCP server for MCP-based clients like Claude Desktop and Claude Code, and callable as a plain HTTP endpoint by any agent framework via the Apify API.

How does 2GIS Places Scraper compare to other 2GIS place scrapers?

As observed on their Apify Store listings on 2026-08-16: one competitor returns 51 fields per place at $2.81/1,000 results with no review or geofencing support; another advertises 90+ fields and reviews/photos at tiered pay-per-event pricing starting at $4.50/1,000 places; a third uses a lightweight HTTP crawler (no browser) at $1.80/1,000 results with basic fields only. None of the three publicly document a results-level geofence or cross-run deduplication — both of which 2GIS Places Scraper provides via customGeolocation and enableGlobalDataset.

Does 2GIS Places Scraper return data in a format LLMs can use directly?

Yes. Output is typed, normalized JSON with consistent field names across runs — no HTML parsing or CSS selectors needed. Pass it directly to an LLM, index it into a vector store, or feed it to an agent tool.

What happens when 2GIS changes its layout or anti-bot system?

The Actor is maintained and the output schema is designed to stay stable across 2GIS site changes — field names and structure don't change just because 2GIS updates its page markup.

Can I use 2GIS Places Scraper without managing proxies or browser infrastructure?

Yes. The Actor runs its own headless Chromium browser and handles proxy tier selection (direct, datacenter, residential) and escalation automatically — you don't configure or maintain any of that yourself.

Which 2GIS place fields work best for AI training data and RAG indexing?

For RAG, index summary, rubrics, address and reviews[].text — the highest-information text fields. For training data, id, totalScore, reviewsCount and rubrics have the most consistent structure across every record, and all fields return as typed primitives (strings, numbers, booleans, arrays) rather than free-form HTML.

Related scrapers

Scraper	What it extracts
Google Maps Reviews Scraper	Reviews and ratings for Google Maps places
Facebook Reviews Scraper	Reviews and ratings for Facebook Pages

Your feedback

Found a bug or a missing field? Open an issue on this Actor's Issues tab in Apify Console, or use the Contact developer button on the Store listing — reports like these are how the Actor stays accurate as 2GIS changes.