Local business intelligence · Google Maps · Snapshot September 6, 2026
The largest index in the catalog: 132,706,518 business listings collected from Google Maps searches and place pages, normalized to one JSON record each and stored in a search index you query directly — by keyword, category, country, state, city, rating, review volume, or as a radius around a point. Nothing here is scraped on request; every call reads the stored index.
132,706,518
One record per place_id, carrying the listing's name, categories, address, rating and review count, website, phone, coordinates and geographic hierarchy.
Snapshot September 6, 2026. Public listing fields only — no reviews text, no owner data.
39.2%
Only 39.2% of listings link to a website — 8073.2万 businesses are findable on the map but have nowhere to send you. The share is measured on the full corpus, not a sample.
59.3%
In the United States it is 59.3% with a website across 2005万 listings — better than the global figure, still four in ten without a site.
15.1%
The United States is the largest single country at 15.1% of the index ; India, China, Japan, Germany, Brazil and the UK follow. The country facet returns 50 buckets, so the long tail beyond them is real but unlabeled.
38%
Quality signals are dense: 38% carry a rating of 4.5 or higher and 7.9% have 100+ reviews — enough to filter for established, well-reviewed businesses at scale.
`has_website` is the one field with a complete two-bucket facet, so the split below is the whole corpus, not a top-N. It is the same measurement behind the Local Business on the Web study — this page is the dataset, that page is the analysis.
For the country-by-country and category-by-category breakdown, the shared-domain patterns and the US state and county maps, read Local Business on the Web, the data study built on this index.
Listings per country for the fifteen largest of the 50 the facet endpoint returns. Shares are of those 50 countries' 1.2亿 listings, not of the corpus.
| Country | Listings | Share of top 50 |
|---|---|---|
| United States | 20,050,344 | 16.2% |
| India | 12,057,906 | 9.8% |
| China | 11,171,645 | 9% |
| Japan | 6,077,012 | 4.9% |
| Germany | 5,643,582 | 4.6% |
| Brazil | 5,600,701 | 4.5% |
| United Kingdom | 5,471,803 | 4.4% |
| Mexico | 4,164,782 | 3.4% |
| Indonesia | 4,124,330 | 3.3% |
| France | 3,880,103 | 3.1% |
| Italy | 3,617,398 | 2.9% |
| Russia | 3,137,668 | 2.5% |
| Spain | 2,703,199 | 2.2% |
| Turkey | 2,365,623 | 1.9% |
| Canada | 2,073,029 | 1.7% |
The country facet is capped at 50 buckets. The 50 together hold 1.2亿 of 132,706,518 records; the remainder sits in countries beyond the cap. Filter with `country=` to work inside any one of them.
Listings per Google Maps category for the fifteen largest of the 50 the facet returns. Categories are Google's own slugs (`restaurant`, `beauty_salon`, `auto_repair_shop`); a listing can carry several.
| Category | Listings | Share of top 50 |
|---|---|---|
restaurant | 2,566,287 | 7.6% |
bus_stop | 1,292,492 | 3.8% |
beauty_salon | 1,221,934 | 3.6% |
grocery_store | 1,159,987 | 3.4% |
clothing_store | 1,105,064 | 3.3% |
pharmacy | 970,386 | 2.9% |
auto_repair_shop | 901,522 | 2.7% |
atm | 854,998 | 2.5% |
coffee_shop | 831,756 | 2.5% |
hotel | 817,779 | 2.4% |
bakery | 802,488 | 2.4% |
bar | 797,632 | 2.4% |
corporate_office | 797,002 | 2.3% |
hair_salon | 763,386 | 2.2% |
park | 762,127 | 2.2% |
269.7万 listings carry no category at all and are excluded from the shares above. The 50 largest categories account for 3393.8万 records; the long tail of niche categories is queryable by exact slug.
The `country=United States` cut, faceted by state. Fifteen largest of the 50 the facet returns; shares are of the US cut.
| State | Listings | Share of US |
|---|---|---|
| Texas | 1,713,403 | 8.5% |
| California | 1,632,354 | 8.1% |
| New York | 1,458,875 | 7.3% |
| Florida | 1,341,176 | 6.7% |
| Pennsylvania | 920,514 | 4.6% |
| Illinois | 771,420 | 3.8% |
| New Jersey | 704,046 | 3.5% |
| Ohio | 699,922 | 3.5% |
| North Carolina | 659,429 | 3.3% |
| Georgia | 629,772 | 3.1% |
| Michigan | 604,042 | 3% |
| Virginia | 544,154 | 2.7% |
| Massachusetts | 475,629 | 2.4% |
| Tennessee | 436,559 | 2.2% |
| Colorado | 421,058 | 2.1% |
Fields are normalized across countries and languages; arrays stay arrays, coordinates are numeric, and the geographic hierarchy is resolved server-side so `state`/`county`/`city` filter consistently.
idplace_idnameurlcategory[]descriptionaddressgeo.latgeo.loncountrystatecountycitytownwebsitephoneratingreview_countpermanently_closedimagecreated_atupdated_atgeo_is_updatedcontact_is_updatedEvery parameter below is a real filter that shrinks the result set (verified when this snapshot was generated). Combine them freely; page with `page` and `page_size`.
Search — well-reviewed restaurants in one city that still have no website
curl "https://api.crawlora.net/api/v1/datasets/google-map-businesses/search?category=restaurant&city=Austin&min_rating=4.5&has_website=false&sort=review_count_desc" \
-H "x-api-key: $CRAWLORA_API_KEY"Nearby — everything within 800 m of a point, best-rated first
curl "https://api.crawlora.net/api/v1/datasets/google-map-businesses/nearby?lat=40.7580&lon=-73.9855&radius_m=800&sort=rating_desc" \
-H "x-api-key: $CRAWLORA_API_KEY"Facets — category mix of one country
curl "https://api.crawlora.net/api/v1/datasets/google-map-businesses/facets?facet=category&country=Japan" \
-H "x-api-key: $CRAWLORA_API_KEY"132,706,518 stored listings as of 2026-09-06, 20,050,345 of them in the United States. The index grows continuously as new searches and place pages are collected, so the total on this page is a snapshot and the API's own `total` field is the live count.
No. It is a discovered index — listings reached through searches and place pages — not a census. Coverage is deepest where searches concentrate (the United States, India, China, Japan, Western Europe, Brazil) and thinner elsewhere. Treat shares as descriptive of the index, not of Google Maps as a whole.
51,974,045 listings (39.2%) link to a website and 42,364,826 (31.9%) list a phone number. The website figure is measured on the whole corpus through the `website_status` facet, which is a complete partition; it matches the headline of the Local Business on the Web study.
Yes. The `nearby` endpoint takes `lat`, `lon` and `radius_m` and returns listings inside the circle, with the same category, rating and website filters as `search`. 132,233,546 listings (99.6%) carry coordinates, so nearby search covers effectively the whole index.
Partly. `permanently_closed` is set when a listing is re-verified against Google Maps and found closed; 5,608 listings carry the flag today. Re-verification is ongoing, so read the flag as a floor — a listing without it has not necessarily been re-checked. Filter with `permanently_closed=false` to exclude the known-closed ones.
The facet endpoint returns at most 50 buckets per field. `website_status` has two, so its split is the full corpus; `country`, `category` and `state` come back at exactly 50 and are truncated top-N lists, so their sums are lower bounds and the shares on this page are shares of those 50 values, not of every record. Filtering by an exact value (for example `country=Portugal`) works for any value, in the top 50 or not.
Every call reads the stored index — no live scraping happens on request, so responses are fast and priced per successful call like the rest of the Datasets API. Public listing fields only: no review text, no owner contact data beyond the listed business phone and website.
Lead lists by category and city, market-density studies by state, or a nearby-places feature — all from one REST API with a free tier and no crawl to run.