2GIS Scraper — Phones, Emails & Reviews | Парсер 2ГИС

Turn any 2GIS search — "кофе" in Moscow, "стоматология" in Kazan, "coffee" in Dubai — into a clean lead list with phone numbers, emails, websites and VK / Telegram / WhatsApp / Instagram links, plus address, rating, opening hours, coordinates and, if you want them, reviews. No login, no 2GIS API key. Export as JSON, CSV or Excel, or pull results through the Apify API.

Парсер 2ГИС без API-ключа. Выгружает организации из 2ГИС по городу и поисковому запросу: телефоны, e-mail, сайты, ссылки на ВКонтакте, Telegram, WhatsApp, адрес, район, рубрики, рейтинг, часы работы, координаты и отзывы. До 10 000 организаций на один запрос, а с «Deep search» — больше (разбивка по районам). Выгрузка в Excel, CSV или JSON.

What does 2GIS Scraper do?

You give it cities and search queries, exactly as you would type them into 2GIS. For every city × query it walks the 2GIS search results and opens each place's card to collect its contacts. Every place becomes one row:

• Contacts: phones (dialable, e.g. +79160663469), emails, website, all websites, and socials — one link per network (VK, Telegram, WhatsApp, Odnoklassniki, Instagram, Facebook, YouTube, TikTok, Max, Dzen, Rutube, LinkedIn, Pinterest…).
• Place: name, address, district, city, categories, coordinates (lat/lon), opening hours per weekday, photo count, chain (orgId/orgName), 2GIS link.
• Reputation: rating, number of reviews, and optionally the reviews themselves (author, stars, text, date, likes), newest first.

In a test run of 300 Moscow coffee shops, 86% had a phone, 90% a social link, 69% a website and 4% a public email — contacts are only there when the business publishes them in 2GIS.

Works for Moscow, Saint Petersburg, Novosibirsk, Yekaterinburg, Kazan and the other Russian cities on 2GIS, plus cities served from 2GIS's country sites such as Almaty, Astana, Tashkent, Bishkek, Minsk and Dubai (use the city slug from the 2GIS link, e.g. almaty from 2gis.kz/almaty).

Use cases

• B2B sales and lead generation — every dentist, car service or beauty salon in a city with phone, email and website, ready for your CRM or cold-calling list.
• Messenger and social outreach — agencies and SMM freelancers collect Telegram, WhatsApp and VK links of local businesses in one export.
• Market research and expansion planning — count competitors by district before opening a new location, map them with coordinates, compare ratings.
• Reputation monitoring — pull the newest reviews of your branches or your competitors' every week.
• Data enrichment — add verified addresses, hours, categories and geo-coordinates to an existing list of companies.
• Analysts and researchers — complete datasets of a business category across a whole city, beyond the 10,000-result limit of one 2GIS search.

How much does it cost?

You pay only for the places and reviews saved to your dataset (pay per event). Duplicates are never charged.

Event	What you pay for	Free plan	Paid plans (from)
place	one place with its contacts	$2.00 per 1,000	$1.50 per 1,000
review	one review (only when Include reviews is on)	$0.50 per 1,000	$0.30 per 1,000

Examples:

• 1,000 coffee shops in Moscow with contacts cost $2.00 on the Free plan and $1.50 on a Gold+ plan.
• 500 dental clinics with up to 20 reviews each (500 places + 10,000 reviews) cost $1.00 + $5.00 = $6.00 on the Free plan and $0.75 + $3.00 = $3.75 on a Gold+ plan.

The Apify Free plan includes $5 monthly credit — enough for about 2,500 places with contacts every month. Use Max items to put a hard cap on any run.

How to use 2GIS Scraper

In Apify Console:

1. Open 2GIS Scraper and click Try for free.
2. Enter one or more Cities (moscow, spb, kazan, almaty… or a 2GIS city link).
3. Enter one or more Search queries (кофе, стоматология, автосервис…). Every city × query pair is searched.
4. Set Max places per city/query (default 30). Turn on Include reviews if you need them.
5. Click Start. Places appear in the dataset while the run is still going.
6. Download the results as Excel, CSV or JSON from the Storage tab.

Through the API:

1. Get your API token in Apify Console → Settings → Integrations.
2. Call the actor with your input (see Integrations and API) and read the dataset items from the response.

Input

Field	Type	Default	Description
cities	array of strings	— (required)	City slugs (moscow, spb, novosibirsk, almaty, dubai…) or 2GIS city links (https://2gis.ru/kazan).
queries	array of strings	— (required)	What you would type into 2GIS search: кофе, стоматология, автосервис, салон красоты.
includeReviews	boolean	false	Also save the reviews of every place, one row per review, newest first.
maxResultsPerQuery	integer	30	Places kept from each city × query, up to 100,000.
deepSearch	boolean	false	Go past the 10,000-places limit of one 2GIS search by repeating it district by district (see Limits).
maxReviewsPerPlace	integer	20	Review cap per place when reviews are on.
maxItems	integer	1000	Hard cap on rows (places + reviews) for the whole run, which also caps the cost.
proxyConfiguration	object	Apify Proxy	Proxy settings; recommended for large runs.
debug	boolean	false	Verbose log.

Example — every coffee place in Moscow, as deep as 2GIS goes:

{
    "cities": ["moscow"],
    "queries": ["кофе"],
    "maxResultsPerQuery": 20000,
    "deepSearch": true,
    "maxItems": 20000
}

Example — dentists in two cities with their latest reviews:

{
    "cities": ["moscow", "spb"],
    "queries": ["стоматология"],
    "includeReviews": true,
    "maxResultsPerQuery": 100,
    "maxReviewsPerPlace": 20,
    "maxItems": 5000
}

Output

One row per place ("type": "place") and, with reviews on, one row per review ("type": "review"). A place from a real run on 2026-11-02:

{
    "type": "place",
    "placeId": "70000001064534699",
    "name": "Comme à Paris, французская кофейня-кондитерская",
    "city": "Москва",
    "citySlug": "moscow",
    "address": "Ленинградский проспект, 36",
    "district": "Аэропорт",
    "categories": ["Кофейни", "Пекарни", "Кондитерские изделия"],
    "rating": 4.8,
    "reviewsCount": 24,
    "phones": ["+79160663469"],
    "emails": ["info@commeaparis.ru"],
    "website": "http://commeaparis.ru",
    "websites": ["http://commeaparis.ru"],
    "socials": {
        "vkontakte": "https://vk.com/commeaparis",
        "whatsapp": "https://wa.me/79104520508",
        "telegram": "https://t.me/commeaparismsk"
    },
    "hours": {
        "Mon": "08:00-22:00",
        "Tue": "08:00-22:00",
        "Wed": "08:00-22:00",
        "Thu": "08:00-22:00",
        "Fri": "08:00-22:00",
        "Sat": "09:00-22:00",
        "Sun": "09:00-22:00"
    },
    "lat": 55.792065,
    "lon": 37.561501,
    "photosCount": 26,
    "orgId": "70000001018413368",
    "orgName": "Comme à Paris, французская кофейня-кондитерская",
    "purpose": null,
    "url": "https://2gis.ru/moscow/firm/70000001064534699"
}

A review:

{
    "type": "review",
    "placeId": "70000001113359188",
    "reviewId": "312671295",
    "author": "Айша Болатбек",
    "rating": 5,
    "text": "Отличная кофейня, вернусь ещё",
    "date": "2026-11-02T18:22:09.436602+07:00",
    "likes": 0,
    "provider": "2gis",
    "isVerified": false
}

Field notes:

• phones are in international format, ready to dial or import into a CRM.
• website is the business's own site; links to social networks that 2GIS lists as a "website" are moved to socials. websites lists every site the business published.
• socials keys are the network names (vkontakte, telegram, whatsapp, odnoklassniki, instagram, facebook, youtube, tiktok, max, dzen, rutube, linkedin, pinterest, twitter…), one link each. In CSV/Excel they become columns such as socials/telegram.
• emails, website and socials are empty when the business did not publish them.
• rating and reviewsCount are for this branch (not the whole chain). Places of the same chain share orgId.
• hours keys are Mon…Sun; a day with a break looks like "10:00-13:00, 14:00-20:00".
• A review's text is empty when the author left only stars; date is the last edit date.

Integrations and API

Run the actor and get the results in one HTTP call:

curl -X POST "https://api.apify.com/v2/acts/innolabs~2gis-scraper/run-sync-get-dataset-items?token=YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"cities": ["moscow"], "queries": ["кофе"], "maxResultsPerQuery": 50}'

Python (pip install apify-client):

from apify_client import ApifyClient

client = ApifyClient("YOUR_TOKEN")
run = client.actor("innolabs/2gis-scraper").call(
    run_input={"cities": ["kazan"], "queries": ["стоматология"], "maxResultsPerQuery": 200}
)
for place in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(place["name"], place["phones"], place["emails"], place["socials"])

• Make, Zapier, n8n — use the Apify app/node to start the actor and send new places to Google Sheets, your CRM or a messenger.
• Scheduling — create a schedule in Apify Console (for example every Monday) to refresh your lists or collect new reviews.
• Webhooks — get notified when a run finishes and fetch the dataset automatically.
• AI agents (MCP) — Claude, Cursor and other MCP clients can call 2GIS Scraper through the Apify MCP server at https://mcp.apify.com.

FAQ

Is it legal to scrape 2GIS?

The actor collects only publicly visible business listing data — the same information anyone sees on 2gis.ru without logging in. It does not access accounts or private data. Check 2GIS's terms and the data protection laws that apply to you (for example 152-FZ in Russia or GDPR), especially before using contacts for marketing.

Do I need proxies?

Small runs usually work without one. For large runs keep the default Apify Proxy on: 2GIS rate-limits busy IP addresses and blocks some networks. If a city returns errors, try a proxy located in that country.

How many places can I get per search?

One 2GIS search returns at most 10,000 places, and the actor reads all of them if you raise Max places per city/query. For bigger categories turn on Deep search: the actor repeats the query for every district of the city and merges the results without duplicates. For "кофе" in Moscow that gave 12,677 unique places instead of 10,000.

Can I get only new places or new reviews?

Each run returns the current state of 2GIS. For an incremental feed, schedule the actor and compare placeId (places) or reviewId (reviews) with your previous export; reviews come newest first, so a small Max reviews per place is enough to catch new ones.

What happens if a city or a search fails?

The failure is logged in the run summary and the actor moves on to the next city or query; places already collected stay in the dataset. An unknown city slug is reported as "city failed". A search with no results simply returns 0 places.

Why does a place have no email or website?

The actor returns what the business published on its 2GIS card. Many small businesses list only a phone or a Telegram/WhatsApp link.

Limits

• 10,000 places per 2GIS search. 2GIS stops serving results after 10,000 for one search, even when it shows a larger total. Deep search goes past it by splitting the search by city district (measured on 2026-11-02 for "кофе" in Moscow: 10,000 places without it, 12,677 with it). Places that 2GIS does not assign to any district (a small share, mostly vending machines) are reachable only through the first 10,000.
• Deep search takes time. It reads every result page of every district — for "кофе" in Moscow about 14 minutes of search pages — on top of opening each place's card for contacts, which runs at roughly 6 places per second (300 places in 49 seconds in our test). Expect a full 12,000-place city run to take around an hour.
• Contacts depend on the business. Most places list a phone (86% of Moscow coffee shops in our test); emails are rare (4%).
• Reviews come from 2GIS's review service. If it is unreachable, places are still saved without reviews.
• Cities without districts. Smaller towns (for example Kolomna) have no city districts in 2GIS, so Deep search adds nothing there — such towns rarely reach 10,000 results anyway.