OpenTable Scraper — Reviews, Restaurants & Availability

OpenTable Scraper — Reviews, Restaurants & Availability

Extract structured restaurant data and individual diner reviews from OpenTable, plus availability and listings. Works in two modes:

1. City search — give a city slug (san-francisco, new-york, chicago) and optional cuisine keyword; the actor returns up to N restaurants with full profile data.
2. URL list — supply a list of OpenTable restaurant profile URLs for targeted scraping.

Why this scraper is best-in-class

Most OpenTable scrapers give you either reviews or listings. This one gives you all three in one run:

• Individual reviews — reviewer, overall rating, food / service / ambience / value sub-ratings, noise level, full text, dined date vs review date, and helpful-vote counts. (Most competitors only return the review text and a single overall star — we return the full breakdown.)
• Restaurant listings — name, address, phone, cuisine, price band, geo, hours, aggregate rating + review count.
• Availability & reservations — reservation acceptance + slots.

Reviews are extracted from the same page as the restaurant record, so they add no extra requests and don't slow the run down.

---

Output shape

Every run produces two kinds of rows, distinguished by record_type, and accessible via two dataset views (Restaurants and Reviews):

• record_type: "restaurant" — one row per restaurant. Also carries a nested reviews[] array and a reviews_extracted count, so you can consume reviews either as flat rows or nested.
• record_type: "review" — one flat row per individual review, linked to its parent via restaurant_id + restaurant_name.

---

What you get per restaurant

Field	Example
name	Thirsty Bear
slug	thirsty-bear-san-francisco
url	https://www.opentable.com/r/thirsty-bear-san-francisco
description	Full restaurant description text
cuisine	["Spanish", "Tapas / Small Plates", "Brewery"]
cuisine_str	Spanish, Tapas / Small Plates, Brewery
price_range	$30 and under
rating	3.9
review_count	615
accepts_reservations	false
street_address	661 Howard St.
city	San Francisco
state	CA
zip_code	94105
country	United States
address	661 Howard St., San Francisco, CA, 94105
latitude	37.78565
longitude	-122.399734
phone	+1 415-974-0905
opening_hours	["Monday 11:30am–10:00pm", ...]
menu_url	https://www.opentable.com/...
reviews_extracted	10
reviews	nested array of review objects (see below)
restaurant_id	2
scraped_at	2026-06-07T12:00:00Z

---

What you get per review

Each review is emitted both as a nested item inside the restaurant record and as its own flat row (record_type: "review"):

Field	Example
restaurant_id	2
restaurant_name	Thirsty Bear
restaurant_slug	thirsty-bear-san-francisco
review_id	OT-2-1804022929-150018263055
reviewer_name	JoeF
reviewer_initials	JF
reviewer_metro	San Francisco
reviewer_review_count	12
overall_rating	5
food_rating	5
service_rating	5
ambience_rating	4
value_rating	5
noise_level	MODERATE
review_text	Great craft beers. Tapas and paella were very good.
dined_date	2018-05-25
review_date	2018-05-27
helpful_count	0
helpful_up / helpful_down	0 / 0
has_photos	false
party_size	null*
occasion	""*
scraped_at	2026-06-07T12:00:00Z

* party_size and occasion are not exposed in OpenTable's public review payload; the columns are kept for a stable schema and returned empty.

Coverage note: OpenTable embeds roughly the 10 most-recent reviews on each restaurant page. The scraper reads those in the same request used for the restaurant record (no extra cost). Set maxReviewsPerRestaurant to cap how many are returned (default 20 = all embedded). Set includeReviews: false to skip reviews entirely.

---

Inputs

City search mode (most common)

{
  "city": "san-francisco",
  "cuisine": "italian",
  "maxItems": 100,
  "includeReviews": true,
  "maxReviewsPerRestaurant": 20
}

• city — City slug matching OpenTable's URL format. Examples: san-francisco, new-york, chicago, los-angeles, miami, seattle, boston, denver, austin.
• cuisine — Optional filter keyword. Examples: italian, sushi, french, mexican, thai, indian, american.
• maxItems — Maximum number of restaurant records to return (default 10, max 5000).
• includeReviews — Also extract individual reviews (default true).
• maxReviewsPerRestaurant — Cap on reviews per restaurant (default 20; OpenTable embeds ~10 per page).

URL list mode

{
  "restaurantUrls": [
    "https://www.opentable.com/r/thirsty-bear-san-francisco",
    "https://www.opentable.com/r/foreign-cinema-san-francisco"
  ]
}

Supply any list of OpenTable /r/{slug} URLs. The actor scrapes each one and returns the full record.

---

How it works

OpenTable uses a behavioral JavaScript challenge. This actor uses a managed proxy layer to reliably access OpenTable pages without any setup on your end.

Restaurant discovery (city search mode): OpenTable publishes a public sitemap index containing ~90,000 restaurant URLs. Each URL encodes the city name in the slug (e.g., thirsty-bear-san-francisco). The actor filters this list by your city input, then fetches each matching profile page.

Data source: Every OpenTable restaurant profile page embeds (a) a JSON-LD @type=Restaurant block and (b) an inline application-state blob containing the most-recent reviews with full sub-ratings. Both are maintained by OpenTable, parsed from the same page in a single request — the most reliable and stable extraction method, with no fragile CSS class selectors.

---

Pricing

Two pay-per-event prices:

• Per restaurant record delivered.
• Per review record delivered.

Reviews come from the same page as the restaurant (no extra request), so review pricing reflects the data value, not extra compute. No per-query or setup fees.

You do not need to supply your own proxy or API key. Access is fully handled by the actor.

---

Usage examples

Lead generation for restaurant industry vendors

Scrape all Italian restaurants in NYC to build a targeted outreach list for a restaurant POS system or food delivery service:

{
  "city": "new-york",
  "cuisine": "italian",
  "maxItems": 500
}

Restaurant data aggregation

Build a dining guide for a specific city by scraping all available restaurants:

{
  "city": "chicago",
  "maxItems": 1000
}

Monitoring specific restaurants

Track data for a known set of competitor restaurants:

{
  "restaurantUrls": [
    "https://www.opentable.com/r/nobu-new-york",
    "https://www.opentable.com/r/eleven-madison-park-new-york",
    "https://www.opentable.com/r/per-se-new-york"
  ]
}

---

Supported cities

Any city where OpenTable has restaurant listings. The city input must match the format used in OpenTable slugs. Common formats:

• san-francisco
• new-york
• los-angeles
• chicago
• miami
• seattle
• boston
• denver
• austin
• las-vegas
• washington-dc (use washington)
• portland
• atlanta
• philadelphia

---

Technical notes

• Anti-bot: handled via managed proxy (baked in, no buyer setup needed).
• Data stability: JSON-LD @type=Restaurant + inline application-state reviews — maintained by OpenTable, much more stable than scraped HTML. Reviews use the rich state blob (sub-ratings, dined date, helpfulness) with a JSON-LD review fallback.
• Reviews: ~10 most-recent reviews per restaurant, extracted from the same page (no extra request).
• Rate: ~1–2 restaurants/second (network-bound by managed proxy response times).
• Coverage: ~90,000 restaurant URLs across all US cities in sitemap.

---

Output example

Restaurant row (record_type: "restaurant", with nested reviews[]):

{
  "record_type": "restaurant",
  "restaurant_id": "2",
  "name": "Thirsty Bear",
  "slug": "thirsty-bear-san-francisco",
  "url": "https://www.opentable.com/r/thirsty-bear-san-francisco",
  "description": "ThirstyBear is the first & only Certified Organic brewery in San Francisco...",
  "cuisine": ["Spanish", "Tapas / Small Plates", "Brewery"],
  "cuisine_str": "Spanish, Tapas / Small Plates, Brewery",
  "price_range": "$30 and under",
  "rating": 3.9,
  "review_count": 615,
  "accepts_reservations": false,
  "street_address": "661 Howard St.",
  "city": "San Francisco",
  "state": "CA",
  "zip_code": "94105",
  "country": "United States",
  "address": "661 Howard St., San Francisco, CA, 94105",
  "latitude": 37.78565,
  "longitude": -122.399734,
  "phone": "+1 415-974-0905",
  "opening_hours": ["Monday 11:30am–10:00pm", "Friday 11:30am–11:00pm"],
  "menu_url": null,
  "reviews_extracted": 10,
  "reviews": [ { "...": "see review row below" } ],
  "scraped_at": "2026-06-07T12:00:00Z"
}

Review row (record_type: "review"):

{
  "record_type": "review",
  "restaurant_id": "2",
  "restaurant_name": "Thirsty Bear",
  "restaurant_slug": "thirsty-bear-san-francisco",
  "review_id": "OT-2-1804022929-150018263055",
  "reviewer_name": "JoeF",
  "reviewer_initials": "JF",
  "reviewer_metro": "San Francisco",
  "reviewer_review_count": 12,
  "overall_rating": 5,
  "food_rating": 5,
  "service_rating": 5,
  "ambience_rating": 4,
  "value_rating": 5,
  "noise_level": "MODERATE",
  "review_text": "Great craft beers. Tapas and paella were very good. Good service.",
  "dined_date": "2018-05-25",
  "review_date": "2018-05-27",
  "party_size": null,
  "occasion": "",
  "helpful_count": 0,
  "helpful_up": 0,
  "helpful_down": 0,
  "has_photos": false,
  "scraped_at": "2026-06-07T12:00:00Z"
}

---

FAQ

Do I need to provide my own proxy or API key? No. Access is fully handled by the actor. No external credentials needed.

How much does it cost to scrape 1,000 restaurants? At the standard rate, 1,000 restaurant records cost $X. Compute costs are charged to your Apify account at standard Apify platform rates.

Is this legal? Does it violate OpenTable's Terms of Service? This actor scrapes publicly accessible restaurant profile pages. The data extracted (name, address, phone, cuisine, hours) is factual business information. Users are responsible for complying with OpenTable's Terms of Service and applicable data protection laws in their jurisdiction. The actor should not be used to spam restaurants, harvest personal data, or automate reservations.

Which cities are supported? Any city where OpenTable has restaurant listings. The city input must match OpenTable's URL slug format (e.g., san-francisco, new-york, los-angeles). See the Supported cities section above.

What if my city returns 0 results? Check that the city slug matches exactly what OpenTable uses in URLs. For example, Washington D.C. uses washington not washington-dc. If unsure, look up any restaurant in that city on OpenTable and copy the city portion of the URL slug.

Can I request custom data fields or bulk pricing? Yes — contact us via the actor Issues tab.