01 / overview

Google's map of the world — now queryable next to yours

Places Insights puts one of the largest, most detailed catalogs of physical places in the world directly into BigQuery — 300M+ places, 480 place types, refreshed monthly. For the first time, you can query Google's own place data in the same statement as your own sales, real estate, or customer data. It describes what places are — their category, hours, rating — not who visited them or when.

See what's included Coverage & pricing

in one line

It is: a structured attributes dataset for millions of physical places, queryable directly in BigQuery.
It is not: a foot-traffic, visits, or movement product — that's a different dataset entirely.

Places Insights overview docs

places_insights.places300M+ rows

place_idSTRING

primary_typeSTRING

ratingFLOAT64

price_levelINT64

wheelchair_accessibleBOOL

···· 65 more columns

0places worldwide

0place types

0attributes

0countries covered

02 / what's included

What's included — and what's not

Every row describes one place, not a category of interest — the same fields whether it's a coffee shop or a cardiology clinic. That breadth is what makes it possible to analyze businesses, new markets, and the physical landscape around them with one consistent schema, whether you're looking at restaurants, hospitals, or auto shops.

Included

480 place types
70+ attributes — price, ratings, hours, parking, accessibility
Brands table (US, Canada, UK, Australia)

Not included

Visit counts
Dwell time
Busyness / foot traffic
Demographics
Individual-level data

About Places Insights data

sample place types

RestaurantHospital Auto repairGrocery store GymPharmacy

sample attributes

Price levelRating HoursParking Wheelchair accessibleDelivery

03 / data source

Google's own dataset, not a resold panel

This is the same underlying place data behind Google Maps — Google's own curated Places dataset for over 300 million establishments worldwide, not a third-party feed licensed and repackaged. It's generally available, and still growing: expect more attributes and more countries over time.

key facts

sourced from Google Maps' curated places data, covering 300M+ establishments worldwide
refreshed monthly
Generally available since September 2025 — attribute and country coverage keep expanding

This is the first time Google's own place data can be queried in the same breath as yours — joined, filtered, and exported entirely on your terms. It lives in your BigQuery project, not inside someone else's app, so nothing else decides what you can do with the results.

GA announcement Google Cloud Architecture Center

Google Mapsreal-world place signals

Curated Places datasetmonthly refresh

Places Insights (BigQuery)generally available

Joins your BigQuery datastandard SQL joins, no ETL

Looker StudioBigQuery ML / Vertex AIYour GCP data

04 / extend with maps platform

Not just a table — a launchpad into Maps Platform

Places Insights is the foundation: category, hours, rating, brand, at scale. When a specific place needs more than the base attributes, bridge straight into the same APIs that power Google Maps itself — no separate vendor, no separate data model.

Places API & Place DetailsEvery Place ID in the dataset bridges directly to the Places API — live photos, current hours, and reviews beyond what's in the static table.

Geocoding API — building outlines & entrancesTurn any address into an exact building footprint and entrance point — the physical shape of a location, not just a lat/long pin.

Places API docs Building outlines & entrances

pairs well with

Places API Place Details Geocoding API Building outlines & entrances Looker Studio BigQuery ML / Vertex AI

example queries

What a query actually looks like

Two examples on a QSR/burger-chain use case: where the open ground is for a new site, and how a brand's locations stack up on reputation against named competitors. Click run on either card — these are mocked queries, not a live connection.

whitespace / competitive density

competing QSR locations near a candidate site

SELECT
  candidate.site_name,
  COUNT(p.place_id) AS competitors_within_2mi
FROM candidate_sites AS candidate
JOIN places_insights.places AS p
  ON ST_DWithin(p.geometry, candidate.geometry, 3218)
WHERE p.primary_type = 'fast_food_restaurant'
GROUP BY candidate.site_name
ORDER BY competitors_within_2mi ASC

Query complete · 0.4s elapsed · 3 rows

site	competitors / 2mi
Route 9 & Elm	2
Westgate Plaza	5
Downtown transit hub	11

reputation benchmarking

rating and price level vs. named competitors

SELECT
  brand_name,
  ROUND(AVG(rating), 2) AS avg_rating,
  ROUND(AVG(price_level), 1) AS avg_price_level
FROM places_insights.places
WHERE metro_area = 'Austin-Round Rock'
  AND brand_name IN ('Brand A', 'Competitor B', 'Competitor C')
GROUP BY brand_name

Query complete · 0.3s elapsed · 3 rows

brand	avg rating	avg price
Brand A	4.3	2.0
Competitor B	4.1	2.0
Competitor C	3.9	1.0

Query the dataset directly Places Count functions

05 / use cases and industries

Where Places Insights gets used

From forecasting demand to picking the next site, here's a sample of how teams put the dataset to work — and the industries built on it. Query results drop straight into a Looker Studio dashboard in minutes.

Sales forecastingModel demand for a new or existing location using the density, quality, and mix of nearby businesses as inputs.

Marketing & media targetingBuild geo-targeted audiences and campaign zones around real category and brand data, not estimated boundaries.

Site selection & whitespaceFind low-competition areas for a new location before committing to a lease.

Competitive benchmarkingCompare ratings, pricing, and hours against named competitors in a market.

Trade area profilingSee the mix of nearby businesses and amenities around any location.

Brand & franchise auditingVerify hours, pricing, and attributes stay consistent across every location.

Data enrichmentJoin place attributes directly onto your own location, CRM, or logistics data via SQL.

Accessibility reportingAudit wheelchair-accessibility coverage across a market or portfolio.

Architecture Center: analyze site performance Architecture Center: visualize in Looker Studio

industries

Retail & QSR Real estate Financial services Insurance Logistics & delivery Hospitality & travel Telecom Government & urban planning Franchising

06 / privacy and delivery

The delivery model is the privacy control

Places Insights ships two ways, and each carries its own built-in privacy floor rather than a bolt-on policy. Either way, no query can isolate a single place's data.

direct query

BigQuery data exchange listing
results under 5 places omitted, not rounded

count functions

40m x 40m minimum search area
always returns a value, including zero

BigQuery analysis rules (aggregation threshold)

SQL query ≥ 5 places?else omitted Result

Count function ≥ 40x40m?min. search area Result (0 ok)

07 / vs. a competitor

Differences, not a verdict

Every row below is a real, verifiable difference — most reflect strengths of Places Insights' single-source, BigQuery-native model.

	Places Insights	Competitor
Native BigQuery SQL access	✓	✕
Single Google-sourced data	✓	✕
Monthly data refresh	✓	✕
No per-query credit cost	✓	✕
Structured, queryable brands table	✓	✕
Consistent attribute schema on every record	✓	✕
Cross-vendor category standardization	✕	✓

Competitor capabilities, pricing, and coverage vary by data source and configuration — confirm current specifics for whichever offering you're evaluating against. Full row-by-row detail also lives in chat as markdown.

08 / coverage and pricing

Where it's available, and what it costs

Coverage is sold per country — 50 separate datasets, not one global dataset — and here's what each costs beyond the free sample.

coverage

50 separate datasets — one per country, each its own BigQuery listing
free sample: one city per country, city-only

cost

sample dataset — free, per country
each country's full dataset — paid, contact sales, no rate card

Supported locations reference

50 separate country datasets make up the full paid catalog — each subscribed individually as its own BigQuery listing