Mapzen Places

Find out more about a location with the Mapzen Places API. Whether it’s a continent, a nation, a region, a county, a locality, a neighbourhood, or even a venue, we can tell you about its geometry, its relationships, and return useful metadata.

Get well-defined placetypes with admin hierarchies as plain-text GeoJSON records with stable numeric IDs. All the data is openly licensed and ready to be used in commercial projects.


  • Plain-text GeoJSON records with stable numeric IDs
  • A finite set of placetypes and a common set of ancestors for all records
  • Perform spatial queries without running a spatial database
  • Worldwide admin hierarchies from continent to neighbourhood
  • Constantly improving language coverage, focused on 20 commonly referenced languages
  • All open data, with licensing suitable for commercial projects
  • Multiple concordances, meaning we can “holds hands” with Wikipedia, Wikidata and other datasets
  • Records may be updated, superseded or deprecated, but never removed or replaced



We’re gathering, creating, and aggregating the shapes of places collected from government and authoritative data sources with open data licensing. In addition, we support multiple geometries for (and opinions about) any single location.

All the data is available as convenient plain-text GeoJSON files that can be readily displayed and processed in Tangram, QGIS, and other GIS tools.


Every record knows where it is and what it is in, and has ‘ancestors’ all the way up from neighbourhoods to continents.

As an example, here is the hierarchy for Zanzibar:

  continent: 102191573, (Africa)
  country: 85632227 (Tanzania)
  region: 85679699 (Zanzibar West)
  county: 1108693173 (Mjini)
  locality: 890435609 (Zanzibar)

Using the Mapzen Places API, you can download some or all of the descendants of a place into a single GeoJSON file – get the counties for a region, postal codes for a locality, or regions for a continent.


All records have a common set of properties that can be supplemented with data specific to that place. We’ve integrated and aggregated useful information about places including translations, statistics, label centroids, and bounding boxes from a variety of sources.


The great thing about unique IDs is there are so many to choose from. For any record with an ID, we can add concordances, which lets you look up more things not stored in Mapzen Places but hold hands with it. For example, San Francisco:

  dbpedia: San_Francisco
  freebase: en.san_francisco
  factual: 08cb9cb0-8f76-11e1-848f-cfd5bf3ef515
  fips:code: 667000
  geonames: 5391959
  geoplanet: 2487956
  library of congress: n79018452
  new york times: 9223372036854776000
  quattroshapes: 240388
  wikidata: Q62
  wikipedia: San Francisco


Venues are points of interest like businesses, restaurants, and parks. You can gather venues belonging to a place, and fetch tags of other properties. For example, you can get restaurants by neighbourhood, airports by nation, garages by county.

Venues are always changing, and a database of venues will always be a work in progress. Building an open, global datsaset of venues is a hard problem, but that’s what we’re doing.

You can also use the Mapzen Places API to supplement and improve your own dataset. Get polygons and points and administrative hierarchies for your venues, all with stable, unique IDs. Use concordances and hierarchies to bring your data to the next level.

We invite you to concord or even contribute your data to Mapzen Places via Who’s On First. We’d love to work with you!


All data in Mapzen Places ranges from CC-0 to CC-BY at its most restrictive, and is suitable for commercial projects (with appropriate attribution).

Get Started

Sign up for an API key to get started with Mapzen Places.


Read through our documentation to get started with Mapzen Places.

Learn more about the Who’s On First Data project here.