Discussions
Exact query params supported on /reviews?
/retrieve-task/ – Supported Parameters
Do we store resolved IDs/URLs for customers (Lite vs Plus behavior), or must clients persist them?
The API operates statelessly. Clients are responsible for persisting their respective profile_key values.
For non-Google publishers (Yelp, DoorDash, Uber Eats, TripAdvisor, Facebook, OpenTable, etc.), what identifiers do we accept?
Full business page URLs for each platform:
Yelp: "https://www.yelp.com/biz/business-name-location"
Facebook: "https://www.facebook.com/BusinessPage/"
TripAdvisor: "https://www.tripadvisor.com/Restaurant_Review-..."
Do we accept Google Place ID directly for Google, or do we require the full URL?
Place IDs are not accepted; complete Google Maps URLs are required. For instance: "profile_key": "https://www.google.com/maps/place/Starbucks/@37.7928442,-122.4068649,17z/data=!3m2!4b1!5s0x808580619b2a98c1:0x31e52be8bab26a09!4m6!3m5!1s0x8085808bb2bcdb49:0x5375a88b57ff689b!8m2!3d37.79284!4d-122.40429!16s%2Fg%2F11d_c_86jf?entry=ttu&g_ep=EgoyMDI1MDkyNC4wIKXMDSoASAFQAw%3D%3D"
Besides accepting a direct publisher URL/ID, do we expose a /resolve endpoint (input: name, address, phone; output: canonical IDs/URLs per publisher)?
The API is designed to accept publisher URLs, which should be provided via the profile_key parameter.
What is the recommended polling frequency for Lite (e.g., hourly/daily/weekly)? Any hard minimum interval (e.g., “not more frequent than every X seconds”)?
While hourly polling is acceptable, webhooks are the preferred solution for data retrieval, eliminating the need to submit requests. Upon notification via webhook, the data for the specific task can be retrieved.
Concurrent tasks allowed per key and per location/publisher?
The platform runs 60–80 concurrent workers per node. Under elevated load, it horizontally scales by adding another instance with equivalent concurrency capacity.
What are the default rate limits per API key?
This endpoint is subject to rate limiting to ensure fair usage and system stability. The default rate limit is 180 requests per 60 seconds.