POST /web/scrape with formats.parse: true and parseParams: {"type": "prebuilt", "parser": "yelp-reviews"}. Read the result in parsed.data, with parsed.type: "yelp-reviews". See the Quickstart for authentication and SDK setup.
Don’t send custom rules with this parser. For CSS selectors and rules, use the Custom parser.
Request
Use an HTTPS Yelp business URL,yelp.com/biz/{alias}. Yelp search returns these URLs. Reviews are always sorted newest first. The optional start parameter is a zero-based offset; each page contains up to 20 reviews.
sort_by=date_desc is accepted; other sort orders and non-empty q, rr, rating, or review_filter parameters are rejected. Single-review links using hrid and tracking parameters are accepted, but those parameters are ignored: the parser returns the requested page, not one selected review.
Output
Whenparsed.success is true, the parsed output has this shape:
Review fields
text is the full review. updated_at is Yelp’s last-modified time in ISO 8601 format, not the original publication date. Review photos and owner replies are not included.
There is no total review count in pagination. A full page sets next_offset, but the next page may still be empty. An empty page returns reviews: [], business: null, and next_offset: null.
Illustrative parsed.data with one review:
Errors and request options
Unsupported URLs, sort orders, or review filters fail withINPUT_VALIDATION_ERROR; an unavailable or invalid Yelp response fails with WEBSITE_ACCESS_ERROR. Check parsed.success before reading parsed.data. A failed parse has data: null and can coexist with successful outputs. See timeouts and errors.
This parser fetches Yelp data separately from the rendered page. Content filters and browser actions do not change its results. Browser profiles are unsupported. Caching, timeouts, and zero data retention still apply.