Get Yelp Reviews for a Topic
View all Yelp Business, Reviews & Restaurant Data endpoints →Request reviews associated with one topic for a specific Yelp business. Supply the business alias, 22-character business ID or Yelp business URL together with a topic name obtained from the Review Topics operation for that business. The documented result contains reviews about the selected topic and a matching snippet, with 20 reviews per page. page defaults to 1 and has a documented range of 1–1000; this endpoint does not expose a page_size input. This operation has no saved response probe, and its supplied response example is an empty object. Topic-matching reviews and supporting snippets are documented response content, not independently observed fields. Exact JSON property names, the snippet layout, pagination metadata, review identifiers and empty-result behavior are not established by the available evidence. Do not infer that it uses the general business-review response schema. Use the snippets to inspect evidence for a returned topic rather than treating a topic label as a complete account of customer experience or as a sentiment score.
API Features
Live API Demo
🔒 Sign in to try it live
Log in to call this API with your own API key. New accounts get 100 free credits — no card required.
Log in / Sign upNo active API key found. Create one in your Dashboard, then reload this page. Dashboard →
/api/project/yelp/reviews-by-topic API Response (JSON)
{} curl -X GET 'https://api.opendata-api.com/api/project/yelp/reviews-by-topic?key=<API_KEY>&business=gary-danko-san-francisco&page=1&topic=<topic>'
Query Result
Use Cases
- Select a service topic returned for a restaurant and prepare a focused evidence packet of matching passages for a service-quality review.
- Let a branch operator check whether the snippets behind a topic label actually support the wording of an internal issue summary before sharing it.
- Compare the supporting review passages for the same returned topic across selected branches, retaining each branch's topic vocabulary and avoiding sentiment-score assumptions.
- Build a human-reviewed theme investigation queue that records the requested business, topic and page, with a parser that is finalized only after the actual response layout is confirmed.
Frequently asked questions
Where should I obtain the topic value?
Use a topic name from the Review Topics operation for the same business. service is the documented example, not a guarantee that every business exposes that topic.
Is a topic the same as a keyword search?
No. topic selects a name supplied by Review Topics. This endpoint does not expose the general review endpoint's query parameter, so do not assume that arbitrary keywords will resolve to a topic.
Can I change the number of topic reviews per page?
The contract documents 20 reviews per page and exposes page but not page_size. page defaults to 1 and accepts documented values from 1 to 1000. The response size and availability have not been independently verified.
Which response fields and snippet format are verified?
None have been independently verified for this endpoint. The documented content is reviews about the selected topic with a matching snippet, but there is no saved probe and the supplied example is an empty object. Confirm actual field names, pagination metadata and snippet layout before depending on a parser.
Does a topic snippet establish overall customer sentiment?
No. A matching snippet is supporting context for a selected topic, not a documented sentiment score or proof that the topic represents every review. Inspect the returned passages and retain the business and topic context.