Microproduct data guideadvanced

Overture Maps Places

Global point-of-interest places for building local business, amenity, and coverage maps from a bounded extract.

Start the guide Open official source

At a glance

Difficulty
Advanced — expect specialized tools or domain knowledge
Size
Large · 0.001–50 GB
Formats
GeoParquet, GeoJSON
Access
Download
API key
Not required
Provider
Overture Maps Foundation
Updates
Monthly
Data terms
Community Data License Agreement - Permissive, Version 2.0

From source to product signal

Test a product idea in four steps

Overture Places is a global point layer published as GeoParquet. Start with DuckDB, one metro bounding box, and a 50-row limit. Do not copy an entire theme. Place names and categories can lag local reality, and GERS identifiers are not stable storefront IDs.

1

Check the setup

  • Python 3.10 or newer
  • A notebook environment such as Jupyter or Google Colab
  • DuckDB 1.1 or newer with spatial and httpfs extensions
2

Access the data

  1. 1.Read the Places guide and attribution page and stay on the Places theme.
  2. 2.Query the current release through a bounding box instead of downloading the theme.
  3. 3.Keep names, categories, and coordinates for a single metro extract.
Open official source
3

Run the Python example

Install the packages, then run the notebook cell.

python -m pip install duckdb pandas

import duckdb
import pandas as pd

con = duckdb.connect()
con.execute("INSTALL httpfs; LOAD httpfs;")
con.execute("INSTALL spatial; LOAD spatial;")
places = con.execute(
    """
    SELECT
      id,
      names.primary AS name,
      categories.primary AS category,
      bbox.xmin AS lon,
      bbox.ymin AS lat
    FROM read_parquet(
      'https://overturemapswestus2.blob.core.windows.net/release/2026-07-22.0/theme=places/type=place/*',
      filename=true,
      hive_partitioning=1
    )
    WHERE bbox.xmin BETWEEN -105.1 AND -104.8
      AND bbox.ymin BETWEEN 39.7 AND 39.8
    LIMIT 50
    """
).fetchdf()
places["retrieved_at_utc"] = pd.Timestamp.now(tz="UTC")
print(places.head())
4

Test a useful signal

Inventory places in one downtown bounding box

Test whether a 50-row Places extract can power a neighborhood amenity map.

  1. 01Confirm the extract stayed inside the bounding box and did not scan the global theme locally.
  2. 02Count places by primary category and flag missing names.
  3. 03State that monthly releases change GERS coverage and that a place record is not proof a business is open.

Dataset details

Overture Maps Foundation is a nonprofit source. Last verified 2026-08-17. Temporal coverage: monthly releases.

Theme

Geospatial & Infrastructure

Domains

GeographyCartographyLocal Economics

Data types

GeospatialTabular

Tasks

MappingSpatial AnalysisMarket Research

Geography

Global

Formats

GeoParquetGeoJSON

Provider

Overture Maps Foundation

Data terms

Community Data License Agreement - Permissive, Version 2.0