Yelp Business Search & Review Scraper

Search Yelp by keyword and location, or paste specific business page URLs, and get back structured business profiles with ratings, categories, contact details, and the customer reviews shown on each business page.

What this actor does

• Searches Yelp for every keyword × location pair, or scrapes specific business URLs directly (both can be combined)
• Returns business name, rating, review count, price range, categories, phone, address, and website
• Collects the customer reviews shown on the business page with star rating, date, text, photos, and Useful/Funny/Cool reaction counts
• Loads more search results automatically up to your configured limit (at most 10 result pages per query)
• If the input is unusable or Yelp blocks every attempt, emits a single status record instead of failing the whole run
• Every field below is always present on each business record; fields the business page doesn't show come back as an empty string or empty array rather than being left out

Output per business

• directUrl — full Yelp URL of the business page
• bizId — Yelp business slug identifier
• name — business name
• description — business description from the page's metadata
• categories — array of business categories (e.g. "Pizza", "Italian")
• type — business type from structured data (e.g. "Restaurant")
• phone — business phone number
• reviewCount — total number of reviews
• aggregatedRating — average star rating (e.g. "4.5")
• priceRange — price level (e.g. "$", "$$", "$$$")
• cuisine — cuisine type, for restaurants
• website — business website URL
• images — array of business image URLs
• address — object: streetAddress, addressLocality, addressRegion, postalCode, addressCountry
• reviews — array of review objects (empty when reviewLimit is 0), each with:
  • date — review date
  • rating — star rating (1–5; a number or a numeric string depending on how the page delivered it)
  • text — review text
  • photos — array of photo URLs attached to the review
  • isUsefulCount, isFunnyCount, isCoolCount — reaction counts

Status record

Instead of businesses, the dataset holds a single record with type: "yelp_blocked", reason, message and scrapedAt when:

• reason: "invalid_input" — the proxy is switched off, or neither searchTerms nor directUrls was given
• reason: "upstream_error" — no business pages could be found or loaded (usually a Yelp block); this record may also list up to 10 requestedUrls

Input

Field	Type	Default	Description
searchTerms	array of strings	–	Search queries (e.g. "pizza", "plumber", "hair salon")
locations	array of strings	–	Locations to search (e.g. "New York, NY"); each term is combined with each location. Without locations, terms are searched with no location; locations without terms do nothing
directUrls	array	–	Direct Yelp business page URLs to scrape instead of searching
searchLimit	integer	10	Maximum business results per term × location pair (1–100)
reviewLimit	integer	5	Maximum reviews per business (0–50; set 0 to skip reviews). Only reviews on the business page itself are read, so fewer may come back
runTimeoutSecs	integer	1800	Wall-clock cap on the run, in seconds (60–7,200); lower it for quick test runs, raise it for large searches
proxy	object	{"useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"]}	Required. Residential proxy configuration; Yelp blocks non-residential traffic

At least one of searchTerms or directUrls is required, along with proxy.

Example: keyword search across two cities

{
  "searchTerms": ["pizza", "sushi"],
  "locations": ["New York, NY", "San Francisco, CA"],
  "searchLimit": 5,
  "reviewLimit": 3,
  "proxy": { "useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"] }
}

Example: scrape specific business pages

{
  "directUrls": [
    { "url": "https://www.yelp.com/biz/prince-street-pizza-new-york-2" },
    { "url": "https://www.yelp.com/biz/joes-pizza-new-york" }
  ],
  "reviewLimit": 10,
  "proxy": { "useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"] }
}

Example: business lookup without reviews

{
  "directUrls": [{ "url": "https://www.yelp.com/biz/prince-street-pizza-new-york-2" }],
  "reviewLimit": 0,
  "proxy": { "useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"] }
}

Use cases

• Market research — survey competitor businesses, ratings, and customer feedback in an area
• Lead generation — collect business contact details for outreach campaigns
• Location intelligence — gauge business density and quality across neighborhoods
• Review monitoring — track a specific business's reviews over time with scheduled runs
• Customer sentiment analysis — aggregate review text for trend or keyword analysis

FAQ

Do I need a proxy? Yes. A residential proxy is required — Yelp reliably blocks data-center traffic, so the proxy input must stay on the RESIDENTIAL group.

How many businesses can I scrape in one run? Up to searchLimit businesses per term × location pair (max 100 each), plus every URL in directUrls, until runTimeoutSecs runs out. Keeping a single run to roughly 100 businesses and splitting larger jobs across runs is the most reliable approach.

What happens if Yelp blocks the run? If no business could be collected, the actor exits cleanly with a single status record (type: "yelp_blocked", reason: "upstream_error") instead of failing, so downstream pipelines can detect the block and retry. Filter on type to separate it from business records.

Why are some review or business fields empty? Yelp doesn't expose every field on every business or review — a field comes back as an empty string or array when the business page doesn't actually show it.

Can I scrape every review for a business? No. The actor reads only the reviews shown on the business page itself, in Yelp's default order, up to reviewLimit; it does not page through a business's full review history.

What locations are supported? Any location Yelp itself supports. Use "City, State" for US locations (e.g. "New York, NY") or a city name for international locations.

What's the difference between searching and using directUrls? Searching with searchTerms + locations discovers businesses matching a query; directUrls skips discovery and scrapes the exact business pages you provide — faster when you already know which businesses you want.

What happens if a run hits runTimeoutSecs? The actor stops starting new fetches at that mark, pushes everything already collected, and exits cleanly rather than timing out mid-fetch.

Is this affiliated with Yelp? No. This is an independent, third-party actor that reads Yelp's public business pages; it isn't affiliated with or endorsed by Yelp Inc.