Talabat Reviews Scraper: Review Text, Ratings & Dates avatar

Talabat Reviews Scraper: Review Text, Ratings & Dates

Pricing

from $0.35 / 1,000 reviews

Go to Apify Store
Talabat Reviews Scraper: Review Text, Ratings & Dates

Talabat Reviews Scraper: Review Text, Ratings & Dates

Scrape public Talabat restaurant reviews: review text, star rating, reviewer first name, date and restaurant context. UAE, Kuwait, Saudi, Bahrain, Oman, Qatar, Jordan, Egypt, Iraq. Plain HTTP, no login, pay per review delivered.

Pricing

from $0.35 / 1,000 reviews

Rating

0.0

(0)

Developer

subimpact network

subimpact network

Maintained by Community

Actor stats

0

Bookmarked

3

Total users

2

Monthly active users

2 days ago

Last modified

Features

Limited permissions

Share

Talabat Reviews Scraper

Scrape public customer reviews of Talabat restaurants as structured JSON: review text, star rating, reviewer first name, review date and the restaurant context it belongs to. Talabat is Delivery Hero's food delivery marketplace in the Middle East and North Africa, and this Actor covers all nine of its country sites.

It is a reviews actor, not a menu actor: every row is one review that a customer left on a Talabat restaurant page.

What it does

  • Takes Talabat restaurant URLs (https://www.talabat.com/{country}/{slug}) or numeric vendor IDs.
  • Loads the restaurant page once to read the restaurant name, id, slug, aggregate rating, rating count and cuisines, then pages through the restaurant's written reviews.
  • Returns one dataset row per review, newest first, with the review text exactly as published, the star rating, the reviewer first name and both the raw date and an ISO date.
  • Supports UAE, Kuwait, Saudi Arabia, Bahrain, Oman, Qatar, Jordan, Egypt and Iraq.
  • Runs over plain HTTP at 256 MB, with no browser and no login. Reviews are public.

Reviews can be filtered with minRating / maxRating (for example, only 1 and 2 star feedback). Filtered rows are neither delivered nor charged.

How to use

Input example

{
"restaurantUrls": ["https://www.talabat.com/uae/mcdonalds"]
}

That is the default input: one restaurant, up to 100 reviews (one page).

FieldTypeDescription
restaurantUrlsstring[]Talabat restaurant page URLs. Query strings and trailing slashes are ignored.
vendorIdsstring[]Numeric vendor IDs (e.g. 40701), used directly. Restaurant fields are null for these rows.
maxReviewsPerRestaurantintegerReviews per restaurant, 1 to 5000. Default 100.
maxItemsintegerGlobal cap on review rows for the whole run, 1 to 100000. Default 1000.
minRatingintegerOptional. Keep only reviews with at least this many stars (1 to 5).
maxRatingintegerOptional. Keep only reviews with at most this many stars (1 to 5).
proxyConfigurationobjectDefault is no proxy, which works. Apify RESIDENTIAL proxies also work. Apify DATACENTER proxies are blocked by Cloudflare.

You need at least one restaurant URL or vendor ID. If both are empty the Actor logs that and scrapes the sample restaurant, so a bare {} run still works.

Output example

One dataset item per review. This row is the first review of https://www.talabat.com/uae/burger-hut (Burger Hut, vendor 40701):

{
"vendorId": "40701",
"restaurantId": "530",
"restaurantName": "Burger Hut",
"restaurantSlug": "burger-hut",
"country": "uae",
"restaurantUrl": "https://www.talabat.com/uae/burger-hut",
"restaurantRating": 4.7,
"restaurantTotalRatings": 1000,
"cuisines": "Mediterranean, American, Burgers",
"reviewId": "c68bb69cdf67351a40adb321ac1245a7401b98fe",
"reviewerFirstName": "Meme Hamze",
"rating": 5,
"reviewText": "perfect 👌",
"reviewDate": "16 October 2026",
"reviewDateIso": "2026-10-16",
"page": 1,
"scrapedAt": "2026-10-29T12:00:00.000Z",
"input": null,
"error": null
}

If an input cannot be scraped (wrong country, restaurant page not found, a blocked request) the run pushes one row with input and error filled in, the remaining fields null, and does not charge for it.

How to find a restaurant URL

  1. Open talabat.com and pick your country (UAE, Kuwait, Saudi Arabia, Bahrain, Oman, Qatar, Jordan, Egypt or Iraq).
  2. Search for the restaurant and open its page.
  3. Copy the URL from the address bar. It looks like https://www.talabat.com/uae/burger-hut: the first path segment is the country and the second is the restaurant slug.
  4. Paste one URL per line into Restaurant URLs, or paste several restaurants to collect their reviews in one run.

Numeric Vendor IDs (for example 40701) are also accepted when you already have them. They skip the restaurant page, so the restaurant name, rating and cuisine fields come back null.

Use cases

  • Reputation monitoring for restaurant owners and multi-brand groups across MENA markets.
  • Sentiment and quality analysis of delivery food, packaging and rider complaints.
  • Competitor benchmarking on the same marketplace.
  • AI/RAG datasets: natural language reviews with a star rating and a date.

Pricing

Pay per event: $0.50 per 1,000 reviews on the Free plan ($0.0005 per review). Prices are lower on paid Apify plans. You only pay for review rows that are actually delivered to the dataset, and platform usage is included in the event price. Error rows and rows removed by the rating filter are free.

How this compares

Prices are Apify Store list prices for the Free plan, checked on 29 October 2026. They can change.

ActorPrice per 1,000 reviewsNotes
subimpact/talabat-reviews-api (this Actor)$0.50Reviews only, plain HTTP, no proxy by default
scrapesage/talabat-scraper$1.00 (review event)Restaurants, menus and reviews
automation_craft/talabat-scraper$0.20 (review event)Restaurants, menus and reviews

The other two Actors are broader scrapers that also return menus and restaurant listings. This Actor is reviews-first and returns the review text itself, with a stable per-review id, so it is a good fit when reviews are the whole job.

Limits

  • Talabat returns written reviews newest first. This Actor keeps that order and pages through them until maxReviewsPerRestaurant or maxItems is reached, or until the restaurant has no more reviews.
  • The review list contains written reviews only. Ratings without text do not appear in it (the Burger Hut sample shows 1,000 ratings on the page and about 300 written reviews).
  • When Talabat returns the same review twice (identical name, date, rating and text), it is delivered and charged once.
  • Many restaurants have few or no written reviews. In a test of 13 UAE and Egypt restaurant pages on 29 October 2026, 6 returned reviews (12 to 516 each) and 7 returned none. A restaurant with no written reviews produces no rows and no charge; the run log says so.
  • Written reviews returned by Talabat went back to January 2022 in that test; older reviews were not returned.
  • A restaurant URL resolves to the branch Talabat shows for that URL. Chains with many branches may need one vendor ID per branch.
  • Talabat publishes the reviewer's first name only, and sometimes an initial. There is no reviewer profile or user id.
  • Talabat's own review id is always 0, so reviewId is generated by this Actor as a SHA-1 of the vendor id, date, first name, rating and review text. It is stable across runs for the same review, but it is not a Talabat id.
  • Aggregate restaurant rating and rating count are what Talabat shows on the page at scrape time.
  • Apify DATACENTER proxies are blocked by Cloudflare. The default (no proxy) works, and Apify RESIDENTIAL proxies work. If a direct request is blocked, the Actor retries it through a residential session and logs a warning.
  • If the Talabat API changes shape, the Actor writes one error row for the affected input instead of delivering empty or malformed rows.

Responsible use

This Actor collects publicly visible review data: review text, star rating, first name and date, exactly as Talabat shows them to any visitor. Use it in line with Talabat's terms of service and your local regulations, and do not use the output to profile or contact individual reviewers. Keep request volumes reasonable: the Actor paces itself, and the caps are there to keep runs small and predictable.

Run monitoring and privacy

If a run fails or times out, this Actor tells its developer the run ID and status, so we can fix problems fast. It never sends your input, your results or your account data.