Google Maps scraper for place search, business data and reviews

A Google Maps scraper reads what Google Maps shows publicly and returns it as data instead of a map. The Roketfy Data Engine splits this into three endpoints that match the three things people look at on Maps: a result list, a single business profile, and that business's reviews.

`/v1/google/maps-search` returns a ranked list of places for a query, with ratings, addresses, place ids and CIDs. `/v1/google/maps-place` returns one Google Business profile in full: categories, hours, attributes, rating distribution and review topics. `/v1/google/maps-reviews` returns review text, rating, timestamps, images and owner responses for a place.

The three are designed to be chained. Search gives you identifiers, and the place and review endpoints accept those identifiers, so you never have to match a business by name twice.

Start with `/v1/google/maps-search`. It takes a `keyword`, which is the same text you would type into Google Maps, and a required `location` that anchors the search to a country. Put the business type and the area in the keyword, for example "coffee roasters in Portland", and you get back the places Google ranks for it.

Each call covers one query and returns up to 700 places, so a lead list for a region is a loop: one call per business type and area. Store the place id or CID of every result as your primary key. Names and addresses change and can repeat across branches. The ids do not, which makes deduplicating overlapping searches straightforward.

Search costs 25 credits per 100 results, and a `depth` parameter sets how many results a call returns (100 by default, up to 700). If you need more than the list view gives you for a given lead, call the place endpoint only for the rows that pass your own filter, such as a minimum rating, to keep the credit cost in line with the leads you actually keep.

`/v1/google/maps-place` returns a single Google Business profile: its categories, opening hours, attributes, rating distribution and the review topics Google groups its reviews under. It costs 40 credits per call.

You identify the place through the `keyword` parameter in one of three ways: the business name, `cid:<id>`, or `place_id:<id>`. The identifier forms are the reliable choice. A name can resolve to a different branch depending on the location, while a CID or place id from a previous maps-search call always points at the same profile.

Yes. `/v1/google/maps-reviews` returns the reviews for a place, with text, rating, timestamps, images and owner responses, up to 4490 in one task. A `depth` parameter sets how many you want and `sort_by` sets the order, for example newest first for ongoing monitoring.

Unlike search and place, the reviews endpoint runs through the task queue: you submit it with `POST /v1/tasks` and poll for the result instead of waiting on a live request. That suits deep pulls, which take longer than a normal HTTP call.

Reviews are billed at 10 credits per ten reviews returned, so a shallow pull of recent reviews for many places costs far less than a full history for each. If reviews are your main interest, the Google Maps reviews API page goes into depth, sorting and multi-branch monitoring in more detail.

Browser extensions and desktop scrapers drive a real Maps session on your own machine. That works for a one-off export, but it ties the job to a person, a browser and an IP address, and the output is usually a spreadsheet you then have to clean and load somewhere.

A Google Maps scraper API fits the cases where the data feeds a system: a CRM enrichment job, a nightly rank check, a dashboard. You make an HTTP request from your own code, get JSON with stable identifiers back, and schedule it however you like. There is no infrastructure to run on your side and nothing to keep open.

Each endpoint has a fixed credit cost: 25 credits per 100 results for a place search, 40 for a full business profile, and 10 per ten reviews. Credits are bought as pay-as-you-go packages, and failed requests are not charged.

Authentication is one Bearer token on every request. The same three endpoints are available as MCP tools, so an AI agent can search for places, open a profile and read reviews using the same token and credit balance. Parameters and the full list of location values are in the docs at data.roketfy.com/docs.