OpenStreetMap POI Extractor - OSM Places & Business Leads avatar

OpenStreetMap POI Extractor - OSM Places & Business Leads

Pricing

from $0.52 / 1,000 poi extracteds

Go to Apify Store
OpenStreetMap POI Extractor - OSM Places & Business Leads

OpenStreetMap POI Extractor - OSM Places & Business Leads

Extract places from OpenStreetMap (OSM) with the Overpass API: restaurants, hotels, shops, hospitals and 20+ categories, or any brand or name, by city + radius or bounding box. Get address, phone, email, website, socials, brand and hours. Lead filters: with phone, email, or no website.

Pricing

from $0.52 / 1,000 poi extracteds

Rating

0.0

(0)

Developer

Automation Lab

Automation Lab

Maintained by Community

Actor stats

0

Bookmarked

112

Total users

24

Monthly active users

3 days ago

Last modified

Categories

Features

Limited permissionsMulti-tenantServer interface

Share

OpenStreetMap POI Extractor for Local Business and POI Data

Extract points-of-interest (POIs) from OpenStreetMap using the free Overpass API — no API key needed. Search a place or brand by name, or find restaurants, hotels, hospitals, shops, parks, and 20+ other categories by location name + radius or precise bounding box. Get name, coordinates, one-line address, phone, email, website, Facebook and Instagram, brand, operator, cuisine, opening hours, an openstreetmap.org link and, optionally, all OSM tags for every place. Contact filters build lead lists: only places with a phone, email or website, or only places without a website.

Best first run: one category in one city

Start with one location (for example Berlin, Germany) and one category such as restaurants, hotels, cafes, or hospitals. Use a modest radius first so you can verify addresses, coordinates, contact fields, opening hours, and raw OSM tags before expanding to multiple categories or a larger bounding box.

Pricing clarity: estimate cost by expected POI count; a small city/category test is the best way to see how many places OpenStreetMap has for your area.

What does OpenStreetMap POI Extractor do?

OpenStreetMap POI Extractor queries the Overpass API — a free, read-only API for the OpenStreetMap dataset — to find and extract points-of-interest in any area worldwide. Just enter a location name (e.g., "Vienna, Austria") and a search radius, pick a category, and the actor returns structured data for every matching place.

You can search for restaurants, hotels, hospitals, parks, museums, shops, banks, gyms, and more. Advanced users can write custom Overpass QL queries for any OSM tag combination. All output fields are structured and flat — ready for CSV/JSON export, Google Sheets import, or database ingestion.

Who is it for? OpenStreetMap POI Extractor use cases

📊 Researchers and data analysts who need geospatial datasets without spending money on premium APIs. Build datasets of restaurants, green spaces, healthcare facilities, or educational institutions for any city or region worldwide. Export to CSV and analyze in Excel, Python, or R.

🏙️ Urban planners and academics studying city infrastructure, accessibility, or service distribution. Extract wheelchair-accessible venues, healthcare coverage, school density, or park availability across administrative boundaries.

📱 App developers and GIS professionals who need POI seed data for maps, routing apps, or location services. OpenStreetMap's global coverage and community-maintained tags make it an excellent alternative to commercial POI databases.

💼 Market researchers and business analysts identifying competitor locations, mapping service gaps, or analyzing local market saturation. Find all coffee shops, hotels, or pharmacies in a target area without expensive data purchases.

🤖 Automation teams building pipelines that periodically refresh local business data for dashboards, reports, or alerting systems.

Why use OpenStreetMap POI Extractor?

  • ✅ No API key required — Overpass API is 100% free and public
  • ✅ Global coverage — OSM data exists for every country and city
  • ✅ Place and brand name search — find names such as Starbucks or Central Library without writing a query
  • ✅ 20+ preset categories — restaurants, hotels, hospitals, parks, shops, banks, and more
  • ✅ Custom queries — write any Overpass QL query for advanced use cases
  • ✅ Geocoding included — enter a city name, no need to look up coordinates
  • ✅ Structured output — name, coordinates, full address, phone, email, website, socials, brand, cuisine, opening hours
  • ✅ Lead filters — keep only places with a phone, email or website, or only places without a website; you pay only for places that pass
  • ✅ HTTP server mode — answer one small-area lookup per request from your app or AI agent (see Server API)
  • ✅ No login, no browser, no proxy — pure HTTP actor, fast and lightweight
  • ✅ Apify platform benefits — schedule runs, export CSV/JSON/Excel, connect to 5,000+ apps

What data can you extract?

FieldDescriptionExample
osmIdOpenStreetMap element ID76514684
osmTypeElement type (node/way/relation)"node"
namePlace name"Pizza Bizi"
categorySearch category used"restaurants"
latLatitude48.2096466
lonLongitude16.3736843
osmUrlLink to the place on openstreetmap.org"https://www.openstreetmap.org/node/76514684"
address.streetStreet name"Rotenturmstraße"
address.houseNumberHouse number"4"
address.cityCity/town"Wien"
address.postcodePostal code"1010"
address.countryCountry code"AT"
fullAddressOne-line address (null when no address is tagged)"Rotenturmstraße 4, 1010 Wien, AT"
phonePhone number"+43 1 5133705"
emailEmail address (email or contact:email)"office@example.at"
websiteWebsite URL"https://www.pizzabizi.at/"
facebook / instagramSocial profiles from contact:* tags"https://www.facebook.com/example"
brandChain brand"Starbucks"
operatorCompany or body that runs the place"NHS"
cuisineCuisine tag"italian;pizza"
openingHoursOSM opening hours string"Mo-Su 11:00-24:00"
wheelchairWheelchair accessibility"yes" / "no" / "limited"
extrasAll other OSM tags (optional){"cuisine": "italian"}

How much does it cost to extract POIs from OpenStreetMap?

OpenStreetMap POI Extractor uses pay-per-event (PPE) pricing — you only pay for what you extract.

EventFREE tierBRONZESILVERGOLD
Run start$0.005$0.005$0.0043$0.00375
Per POI extracted$0.0008$0.00072$0.00064$0.00052

From December 3, 2026 the price moves onto the standard plan curve: run start $0.005 on every plan, and per POI $0.00072 (FREE and BRONZE), $0.000648 (SILVER), $0.000576 (GOLD and above). The FREE per-POI price goes down; SILVER and GOLD go up slightly.

Real-world cost examples:

  • 50 restaurants in a city center (~2 km radius): ~$0.045
  • 500 hotels in a country capital: ~$0.365
  • 1,000 shops in a major shopping district: ~$0.725

Free plan estimate: Apify's free $5/month credit gets you approximately 6,000 POIs on the FREE tier — plenty for research projects and testing.

How to extract POIs from OpenStreetMap

  1. Go to OpenStreetMap POI Extractor on Apify Store
  2. Click Try for free to open the actor
  3. Enter a location name (e.g., "Berlin, Germany") in the Location field
  4. Set a search radius in kilometers (e.g., 5 km for a city center)
  5. Select a category, or enter an optional Place or brand name to search named POIs instead
  6. Optionally set Max POIs to limit results for testing
  7. Click Start to begin extraction
  8. Download results as JSON, CSV, or Excel from the dataset

Example inputs for different use cases:

// Find all restaurants in central Paris (2 km radius)
{
"location": "Paris, France",
"radiusKm": 2,
"category": "restaurants",
"maxItems": 200
}
// Extract hotels in a precise area using bounding box
{
"boundingBox": "48.18,16.33,48.24,16.42",
"category": "hotels"
}
// Find named Starbucks locations around Vienna (case-insensitive)
{
"location": "Vienna, Austria",
"radiusKm": 5,
"keyword": "Starbucks",
"maxItems": 50
}
// Custom Overpass query for bicycle parking in Vienna
{
"category": "custom",
"customQuery": "[out:json][timeout:60];\n(\n node[\"amenity\"=\"bicycle_parking\"](48.18,16.33,48.24,16.42);\n);\nout center;"
}

Input parameters

ParameterTypeDefaultDescription
locationstring—City or address to search around. Geocoded automatically via Nominatim.
radiusKminteger2Search radius in km around location. Min 0, max 100.
boundingBoxstring—Precise area as "south,west,north,east" coordinates. Alternative to location+radius.
keywordstring—Case-insensitive literal match against the OSM name tag. When set, it replaces the category preset for this run.
categorystring"restaurants"Category preset. See full list below; ignored when keyword is set.
customQuerystring—Full Overpass QL query. Used when category is "custom"; cannot be combined with keyword.
maxItemsinteger0Max POIs to extract. 0 = unlimited.
includeAllTagsbooleanfalseInclude all raw OSM tags in extras field.
requirePhonebooleanfalseKeep only places with a phone number.
requireWebsitebooleanfalseKeep only places with a website.
requireEmailbooleanfalseKeep only places with an email address.
onlyWithoutWebsitebooleanfalseKeep only places with no website (lead lists). Cannot be combined with requireWebsite.

Contact filters run before maxItems: with maxItems: 50 and requirePhone: true you get up to 50 places that all have a phone number, and you are charged only for those. | overpassEndpoint | string | — | Custom Overpass API server. Defaults to overpass-api.de. |

Available categories: restaurants, cafes, bars, fast_food, hotels, shops, supermarkets, hospitals, pharmacies, schools, universities, parks, museums, attractions, banks, parking, fuel, post_offices, police, fire_stations, gyms, cinemas, all_amenities, custom

Output examples

Single restaurant record:

{
"osmId": 76514684,
"osmType": "node",
"name": "Pizza Bizi",
"category": "restaurants",
"lat": 48.2096466,
"lon": 16.3736843,
"address": {
"street": "Rotenturmstraße",
"houseNumber": "4",
"city": "Wien",
"postcode": "1010",
"country": "AT",
"countryCode": "AT",
"state": null
},
"phone": "+43 1 5133705",
"website": "https://www.pizzabizi.at/",
"openingHours": "Mo-Su,PH 11:00-24:00",
"wheelchair": "no",
"extras": {}
}

Hospital record with extras enabled:

{
"osmId": 499319990,
"osmType": "node",
"name": "NHS Soho Walk-In Centre",
"category": "hospitals",
"lat": 51.5145964,
"lon": -0.132073,
"address": { "street": null, "city": null, ... },
"phone": null,
"website": null,
"openingHours": null,
"wheelchair": null,
"extras": {
"healthcare": "clinic",
"healthcare:speciality": "minor_injuries",
"operator": "NHS"
}
}

Tips for best results

  • 🎯 Start small — use maxItems: 50 to preview results before extracting thousands of POIs
  • 📍 Be specific with location names — "Paris, France" works better than just "Paris"
  • 🗺️ Use bounding box for precise areas — copy coordinates from bbox.2live.de or bboxfinder.com
  • 🏷️ Enable includeAllTags for research — OSM tags contain cuisine type, seating capacity, email, social media, accessibility details, and more
  • ⚡ Large areas return more results — a 10 km radius in a major city may return 5,000+ restaurants; use maxItems to cap costs
  • 🔁 Schedule runs to detect new or closed venues over time (Apify's built-in scheduler works great for monthly refreshes)
  • 🔍 Custom queries let you filter by multiple tags — e.g., only vegan restaurants (amenity=restaurant + diet:vegan=yes)

Integrations

OpenStreetMap POI Extractor → Google Sheets Extract all restaurants in your city and pipe results directly to Google Sheets via the Apify → Google Sheets integration. Refresh weekly to detect new openings. Build a live competitor map without any manual data entry.

OpenStreetMap POI Extractor → Slack/Discord alerts Schedule a daily run to detect new hospitals, clinics, or healthcare facilities in your target area. When new POIs appear, trigger a Slack webhook to notify your research team automatically.

OpenStreetMap POI Extractor → Make (Integromat) / Zapier Connect extracted POI data to your CRM, send location data to field sales teams, or auto-populate a property management system with nearby amenities. The flat JSON output maps directly to most form fields.

Batch multi-city extraction via Apify API Use the API to queue extractions for 50 cities at once. Each run targets a single city; your pipeline aggregates results into a master dataset for cross-market analysis.

Scheduled accessibility audits Run the actor monthly with category: "all_amenities" and includeAllTags: true to track changes in wheelchair accessibility tags across a neighborhood. Use this for urban planning or disability advocacy research.

API usage — Using the Apify API

Node.js

import { ApifyClient } from 'apify-client';
const client = new ApifyClient({ token: 'YOUR_APIFY_TOKEN' });
const run = await client.actor('automation-lab/openstreetmap-poi-extractor').call({
location: 'Vienna, Austria',
radiusKm: 2,
category: 'restaurants',
maxItems: 100,
});
const { items } = await client.dataset(run.defaultDatasetId).listItems();
console.log(`Extracted ${items.length} POIs`);

Python

from apify_client import ApifyClient
client = ApifyClient(token="YOUR_APIFY_TOKEN")
run = client.actor("automation-lab/openstreetmap-poi-extractor").call(run_input={
"location": "Berlin, Germany",
"radiusKm": 3,
"category": "hotels",
"maxItems": 200,
})
items = client.dataset(run["defaultDatasetId"]).list_items().items
print(f"Extracted {len(items)} POIs")

cURL

curl -X POST "https://api.apify.com/v2/acts/automation-lab~openstreetmap-poi-extractor/runs" \
-H "Authorization: Bearer YOUR_APIFY_TOKEN" \
-H "Content-Type: application/json" \
-d '{
"location": "Tokyo, Japan",
"radiusKm": 1,
"category": "parks",
"maxItems": 50
}'

Use with AI agents via MCP

OpenStreetMap POI Extractor is available as a tool for AI assistants that support the Model Context Protocol (MCP).

Add the Apify MCP server to your AI client — this gives you access to all Apify actors, including this one:

Setup for Claude Code

$claude mcp add --transport http apify "https://mcp.apify.com/?tools=automation-lab/openstreetmap-poi-extractor"

Setup for Claude Desktop, Cursor, or VS Code

Add this to your MCP config file:

{
"mcpServers": {
"apify": {
"url": "https://mcp.apify.com/?tools=automation-lab/openstreetmap-poi-extractor"
}
}
}

Your AI assistant will use OAuth to authenticate with your Apify account on first use.

Example prompts

Once connected, try asking your AI assistant:

  • "Use automation-lab/openstreetmap-poi-extractor to find all restaurants within 2 km of the Eiffel Tower and save results to a CSV file"
  • "Extract all hospitals and clinics in Vienna, Austria using the OpenStreetMap POI extractor, then show me which ones are wheelchair accessible"
  • "Find all parks and green spaces in a 5 km radius around Central Park, New York using OpenStreetMap data"

Learn more in the Apify MCP documentation.

Yes. OpenStreetMap data is published under the Open Database License (ODbL), which permits free use, sharing, and adaptation — including commercial use — as long as you attribute OpenStreetMap contributors and share any derived databases under the same license.

This actor queries the Overpass API, which is the official read-only API for OpenStreetMap data. It is designed for exactly this type of programmatic access. The actor follows Overpass API best practices by:

  • Using a descriptive User-Agent header identifying the application
  • Respecting rate limits with delays between requests
  • Not storing OSM data beyond the actor's dataset output

For your own use case, review the OSM copyright guidelines and add attribution in any published work. This actor is intended for research, business intelligence, and application development using lawfully accessible public data.

This Actor uses no AI models. It sends location searches to public Nominatim and POI queries to public Overpass servers (or your configured endpoint). These services have no additional Actor fee; normal published start and POI charges apply. This is an independent tool, not affiliated with or endorsed by OpenStreetMap.

Inputs, logs, and output remain in your Apify run storage under your account's retention settings; you can delete those runs and datasets in Console. Public OSM output can include business contact details. Only enable all tags when needed, and do not put secrets or private personal data in location, keyword, or custom-query fields.

Failed operations send sanitized diagnostic input, exceptions, and Actor/build/run IDs to our private GlitchTip service for repair. Secret fields and URL query strings are removed, and reports are retained for 30 days. Contact the Actor maintainer through the Store Issues tab for support or diagnostic deletion requests.

FAQ

How accurate is OpenStreetMap POI data? OSM data quality varies by region. Major cities in Europe and North America have excellent coverage (often better than Google Maps for specialized places). Rural areas or less-mapped countries may have sparser data. OSM is community-maintained and updated continuously — millions of edits happen every week.

How fast does the actor run? For most queries (under 1,000 POIs), the actor completes in under 30 seconds. The Overpass API typically responds in 5-15 seconds. Larger areas with thousands of results may take 60-120 seconds due to Overpass processing time.

How much does it cost to extract 1,000 restaurants? At the FREE tier price of $0.0008/POI + $0.005 start fee: 1,000 × $0.0008 + $0.005 = $0.805. On the GOLD tier (business subscribers): 1,000 × $0.00052 + $0.00375 = $0.524. Apify's free $5/month credit gets you ~6,000 POIs on the FREE tier.

Why do some POIs have null address or phone fields? OSM is a community-contributed dataset — not all places have complete address or contact information. Many nodes have a name and coordinates but no structured address tags. Email, Facebook and Instagram are already read from both the plain and the contact:* tags; includeAllTags: true adds every other tag, such as informal address notes. Use the contact filters to keep only places that have the details you need.

Why are results empty for my location? First check that your location name is recognized (try a more specific format like "City, Country"). Second, some areas have sparse OSM coverage — try a larger radiusKm. Third, some categories like museums have fewer POIs than restaurants — the search is working correctly, there may just be fewer mapped places in that area. An empty answer is checked on a second Overpass server; when both find nothing, the run succeeds with 0 results, says so in its status message, and charges only the start fee. Runs with invalid input or an unknown location fail before any charge.

Can I search for a specific place or brand? Yes. Set keyword together with location + radiusKm or boundingBox. It performs a case-insensitive literal match against named OpenStreetMap nodes, ways, and relations. For example, "keyword": "Starbucks" finds mapped Starbucks locations in the selected area. Category presets are ignored in keyword mode.

Can I filter by sub-categories, like vegan restaurants or 5-star hotels? Yes — use the custom category with an Overpass QL query. For example, to find vegan restaurants:

[out:json][timeout:60];
(node["amenity"="restaurant"]["diet:vegan"="yes"](LAT1,LON1,LAT2,LON2););
out center;

See the OSM Map Features wiki for all available tags.

My custom query covers a large region — how long can it run? The Actor waits as long as your query's [timeout:N] setting allows (Overpass uses 180 seconds when it is omitted), within the run's own timeout. For heavy queries, set the run timeout above the query timeout (for example 300+ seconds for [timeout:240]). If the run time is too short, the run fails with a message naming the timeout to increase. Queries longer than 5 minutes (for example [timeout:1100]) are supported as long as the run timeout allows them.

Can I query historical data with retro, [date:…] or [diff:…]? Yes, but only the main overpass-api.de server keeps history data; the public mirrors used as fallbacks do not. If the main server is busy, the Actor retries it after 10 and 30 seconds (within the run timeout); if it is still busy, the run fails with the main server's error and a note to retry later or lower [timeout].

What happens if my custom overpassEndpoint is busy or my query is invalid? A busy or briefly unavailable custom endpoint is retried once after 5 seconds; it is never replaced by another server. Leave overpassEndpoint empty to get automatic mirror fallback. An invalid query (HTTP 400) is not retried, and the server's parse errors are written to the run log.

The actor runs but returns fewer results than expected — why? If you search a very small radius (under 0.5 km) or a category with few POIs, fewer results are normal. For dense categories in large cities, increase radiusKm or use a bounding box covering the full city. Note that OSM coverage for some business types (e.g., specific shop subtypes) may be incomplete in certain regions.

Server API: find places over HTTP

OpenStreetMap POI Extractor is also a Server Actor. Besides normal runs, it answers short HTTP requests directly, without starting a run for each lookup, which suits apps and AI agents that need nearby places on demand. Copy the server URL from the Actor's Server tab (do not build it from the Actor name) and authenticate with your Apify token. Use normal runs for more than 100 places, larger areas, or custom Overpass queries.

curl -sS -X POST "$SERVER_URL/pois" \
-H "Authorization: Bearer $APIFY_TOKEN" -H "Content-Type: application/json" \
-d '{"location":"Vienna, Austria","radiusKm":1,"category":"cafes","requirePhone":true,"maxItems":10}'
Method and pathInputOutput and charge
GET /Noneready: true, the routes and limits; no lookup or charge
POST /poisJSON body with location (optional radiusKm up to 10, default 2) or boundingBox (each side at most 0.5 degrees); optional category (default restaurants) or keyword, includeAllTags, requirePhone, requireWebsite, requireEmail, onlyWithoutWebsite (same meaning as the run input), maxItems 1–100 (default 20){ok,count,results} with the same records as normal-run dataset items; one POI extracted event per returned place; no start fee

A lookup with no matching place returns 404 NO_RESULTS and is free. Each request has a 90-second deadline and a 1 MiB response cap, and one server works on at most four requests at once. Nothing is cached. Errors use { "ok": false, "error": { "code": "...", "message": "...", "retryable": false } }: 400 INVALID_INPUT or INVALID_LIMIT; 402 spending limit; 404 NO_RESULTS or unknown route; 405 wrong method; 413 body or response too large; 415 body is not JSON; 502 UPSTREAM_ERROR, GEOCODER_UNAVAILABLE or charging failure; 503 server busy or stopping; 504 UPSTREAM_TIMEOUT or TIMEOUT. Failed requests are free. Retry only when retryable is true. The Server tab shows the full OpenAPI description.