AliExpress Reviews Scraper | Feedback, Ratings and Photos avatar

AliExpress Reviews Scraper | Feedback, Ratings and Photos

Pricing

from $1.91 / 1,000 reviews

Go to Apify Store
AliExpress Reviews Scraper | Feedback, Ratings and Photos

AliExpress Reviews Scraper | Feedback, Ratings and Photos

AliExpress reviews scraper (速卖通, 全球速卖通): 100 reviews in 1 second, 36 fields each. Paste a product link, product ID or keyword and get stars, text, English translation, buyer photos, variant bought, country, helpful votes, seller reply and rating breakdown. No account. Export JSON, CSV or Excel.

Pricing

from $1.91 / 1,000 reviews

Rating

0.0

(0)

Developer

SilentFlow

SilentFlow

Maintained by Community

Actor stats

0

Bookmarked

2

Total users

1

Monthly active users

2 days ago

Last modified

Share

AliExpress Reviews Scraper (速卖通) | Feedback, Star Ratings, Buyer Photos and Translations (2026)

1 URL. 100 AliExpress reviews. 36 fields per review.

From SilentFlow. Creators of Pinterest Scraper (#1 for “pinterest pins scraper” in Apify Store search) and Leboncoin Scraper Ppe, with 240,000+ combined lifetime runs.

Turn the reviews under any AliExpress product into a table you can filter: stars, text, English translation, buyer photos, the variant each buyer ordered, their country, helpful votes, the seller's reply, and the product's rating breakdown on every row. 100 reviews in 3 seconds on the prefilled product link, no AliExpress account needed.

Why choose this actor?

Sourcing a product on a 4.8 average without reading the 1-star reviews behind it? Scrolling the review widget 20 reviews at a time, switching languages, to find out whether the case really fits the phone? Pasting product IDs one by one into a tool that cannot search?

  • 38 fields per review, and they are filled. On the prefilled run (one product link, 100 reviews, measured on 30 Sep 2026) 38 of the 52 fields were filled on at least half of the rows. Every review arrives joined to its product: title, main image and gallery, average rating, number of reviews and of ratings (3,055 and 15,439 on that product), the five star counts, photo reviews and the tags buyers use most. Next to the stars and text you get the variant ordered, the buyer's country as a code and as a name, the shipping method, the review's rank and flags for photos, additional reviews and seller replies. The most used AliExpress reviews scraper announces 34 fields and ships no product title.
  • Product links, product IDs and keywords in the same run. Paste a product page from aliexpress.com, aliexpress.us or a regional site, paste a search page you ran in your browser, list product IDs, or type a keyword and get the reviews of the products AliExpress ranks first. Three kinds of input with no mode to pick, where the most used AliExpress reviews scraper takes a list of products and cannot search. In our test the keyword wireless earbuds returned 60 reviews from 5 products in 3 seconds, each row with the product's price and units sold.
  • Nothing to fill in before you press Start. No field is required. The product link is prefilled, and pressing Start returns 100 reviews in 3 seconds. The most used AliExpress reviews scraper will not start until you give it a product list. Every other setting is optional: a cap, a per-product cap, a star or photo filter, the sort, and a switch that keeps written reviews only.

Try AliExpress Reviews Scraper

Try your first AliExpress reviews

Paste this input into the Input tab (JSON view) and press Start. It is the prefilled input of the form: the first 100 reviews of a MagSafe phone case, in the order AliExpress shows them to shoppers.

{
"startUrls": [
{ "url": "https://www.aliexpress.com/item/1005007011478793.html" }
],
"maxItems": 100,
"maxReviewsPerProduct": 0,
"reviewFilter": "all",
"sort": "default",
"withTextOnly": false
}

Runs of this size fit easily in the Apify Free plan, so you can try the actor on the products you are looking at before scheduling anything. Each review is one row of the dataset; export it as JSON, CSV, Excel or HTML from the Storage tab, or read it from the API. The Reviews view shows stars, text, variant bought, buyer, country, date, product and link; the Product ratings view lines up each product's image, title, rating, review and rating counts, positive share and photo reviews.

How it works

SilentFlow · AliExpress suite
  AliExpress Scraper
Products, Prices and Sales
AliExpress Reviews Scraper
You are here

Copy to your AI assistant

Paste this block into Claude, ChatGPT or Cursor to give it full context about this actor:

You have access to the AliExpress Reviews Scraper on Apify: silentflow/aliexpress-reviews-scraper
It returns one row per AliExpress buyer review, read from public AliExpress product pages. No AliExpress account.
Input (all optional; the form is prefilled with one product link):
- startUrls (array of {"url": string}): AliExpress links.
Product pages https://www.aliexpress.com/item/<id>.html (also aliexpress.us and regional hosts): that product's reviews.
Search pages https://www.aliexpress.com/w/wholesale-<words>.html: the reviews of the products listed, in AliExpress order.
- productIds (array of strings): product IDs, e.g. ["1005007011478793"]. Long IDs starting with 3256 work too.
- keywords (array of strings): AliExpress searches, e.g. ["phone case"]. The reviews of the products it returns,
best match first. A product link or ID typed here is read as a product.
- maxItems (integer, default 100): reviews for the whole run. AliExpress lists 5,000 reviews of a product per order;
past that, with reviewFilter "all", the actor goes on star level by star level.
- maxReviewsPerProduct (integer, default 0 = no per-product cap): stop each product after this many reviews.
- reviewFilter (string, default "all"): "all", "5", "4", "3", "2", "1" (one star level), "photos" (with buyer photos),
"additional" (with an additional review the buyer posted after using the product).
- sort (string, default "default"): "default" (AliExpress's recommended order) or "latest" (most recent first).
- withTextOnly (boolean, default false): leave out the reviews that are only a star rating.
Output per row (52 fields, null when AliExpress shows no value):
- Identity: id (string, review id), url (product page anchored on the review id, unique per review), productId, productUrl, productTitle
- Content: rating (1-5), text (as written), translatedText (AliExpress's English translation), isTranslated,
variant (e.g. "Color:Navy Blue Material:For iPhone 11"), skuId, isFeatured, hasImages, imagesCount,
additionalReview, additionalReviewTranslated, additionalReviewAt (RFC 3339), hasAdditionalReview,
sellerReply, hasSellerReply
- Author: author (as AliExpress shows it, masked like "S***y" or "AliExpress Shopper"), authorCountryCode (ISO),
authorCountry (English name), authorAvatarUrl, authorGender ("male" or "female"), isAnonymous
- Measures: helpfulCount, notHelpfulCount, productRating (0-5), productReviewsCount, productRatingsCount,
productPositiveRate (percent), productFiveStarCount, productFourStarCount, productThreeStarCount,
productTwoStarCount, productOneStarCount, productPhotoReviewsCount, productAdditionalReviewsCount,
productTags (array of {name, count}), productPrice, currency, productSoldCount (price and units sold for products
found by keyword or search link)
- Place and time: shippingMethod, publishedAt (RFC 3339, day of the review)
- Media: images (review photos), additionalReviewImages, productImageUrl, productImages
- Meta: position (rank in the product's list under the chosen sort and filter; past the first 5,000 reviews of a product,
the order the run found the rest), keyword, scrapedAt (RFC 3339 UTC)
Use apify-client for Python or JavaScript: client.actor("silentflow/aliexpress-reviews-scraper").call(run_input=...)

How to scrape AliExpress reviews

Three runs, from the simplest to the most filtered. Each JSON is a complete input: paste it into the JSON view of the Input tab.

Get the 300 most useful reviews of a product

One product link, 300 reviews in AliExpress's recommended order, the order shoppers see first on the product page. A product with fewer reviews stops at its last one.

{
"startUrls": [
{ "url": "https://www.aliexpress.com/item/1005005566622938.html" }
],
"maxItems": 300
}

Read the 1-star reviews of a product before you order stock

One star level, most recent first. These are the defects, sizing problems and delivery failures that a 4.7 average hides, in the words of the buyers who hit them.

{
"productIds": ["1005005566622938"],
"reviewFilter": "1",
"sort": "latest",
"maxItems": 200
}

Collect written reviews of the top products for a keyword

A keyword, 20 reviews per product, rating-only reviews left out. In our test the same shape of input on wireless earbuds (latest first, 20 per product, 60 in total) returned 60 reviews from 5 products in 3 seconds, each row carrying the product's price and units sold.

{
"keywords": ["wireless earbuds"],
"maxReviewsPerProduct": 20,
"withTextOnly": true,
"maxItems": 100
}

Input parameters

FieldTypeDefaultDescription
startUrlsarray of linksprefilled with one product linkAliExpress product pages or search pages. See the accepted formats below.
productIdsarray of stringsnoneAliExpress product IDs, one per line.
keywordsarray of stringsnoneAliExpress searches. The reviews of the products it returns, best match first.
maxItemsinteger100Reviews for the whole run.
maxReviewsPerProductinteger0Stop each product after this many reviews. 0 (or -1) reads each product to its last review, within maxItems.
reviewFilterstringallall, one star level 5 to 1, photos (reviews with buyer photos) or additional (reviews the buyer updated after using the product).
sortstringdefaultdefault (AliExpress's recommended order) or latest (most recent first).
withTextOnlybooleanfalseLeave out the reviews that are only a star rating.

No field is required: a run needs at least one product link, product ID, search link or keyword, and the form comes with a product link prefilled. Products asked for by link or ID are read first, in the order you list them, then the products of each search in the order AliExpress ranks them, until maxItems is reached.

Accepted formats

  • Product pages: https://www.aliexpress.com/item/1005007011478793.html, with or without the tracking parameters AliExpress adds, and the same address on www.aliexpress.us, fr.aliexpress.com, m.aliexpress.com and the other regional hosts.
  • Search pages: https://www.aliexpress.com/w/wholesale-phone-case.html?SearchText=phone+case, with the sort and price filters you set on AliExpress kept as they are, and category pages https://www.aliexpress.com/category/<id>/<name>.html.
  • Product IDs: the number in a product link, such as 1005007011478793. The long IDs AliExpress prints on some pages (starting with 3256) and older short IDs are turned into the product's own ID, so the same product listed twice is read once.
  • Keywords: plain words, one per line (phone case, wireless earbuds, yoga mat). English keywords match best. A product link or a product ID typed among the keywords is read as a product.
  • Links to aliexpress.ru (a separate shop with its own catalogue) and to other sites are skipped and named in the log.

What data can you extract from AliExpress?

Every row describes one review, with the same 52 fields in the same order, null when AliExpress shows no value:

GroupFieldsCount
Identityid, url, productId, productUrl, productTitle5
Contentrating, text, translatedText, isTranslated, variant, skuId, isFeatured, hasImages, imagesCount, additionalReview, additionalReviewTranslated, additionalReviewAt, hasAdditionalReview, sellerReply, hasSellerReply15
Authorauthor, authorCountryCode, authorCountry, authorAvatarUrl, authorGender, isAnonymous6
MeasureshelpfulCount, notHelpfulCount, productRating, productReviewsCount, productRatingsCount, productPositiveRate, productFiveStarCount, productFourStarCount, productThreeStarCount, productTwoStarCount, productOneStarCount, productPhotoReviewsCount, productAdditionalReviewsCount, productTags, productPrice, currency, productSoldCount17
Place and timeshippingMethod, publishedAt2
Mediaimages, additionalReviewImages, productImageUrl, productImages4
Metaposition, keyword, scrapedAt3

A few facts that help when you use the data:

  • id is AliExpress's own review ID, stable across runs: use it to find the reviews you have not seen yet. It is a string, because it is longer than a spreadsheet keeps exactly as a number.
  • rating is the buyer's stars, 1 to 5. productRating is the product's average, the same on every row of that product.
  • productReviewsCount counts written reviews and productRatingsCount counts every star rating the product received. On the prefilled product that is 3,055 reviews against 15,439 ratings: most buyers rate without writing.
  • text is what the buyer wrote, in their language. translatedText is the English translation AliExpress shows under it, when it shows one; isTranslated tells you at a glance.
  • variant is the option the buyer ordered, as AliExpress prints it (Color:Navy Blue Material:For iPhone 11), the field to group by when one colour or size collects the complaints.
  • additionalReview is what a buyer added after using the product for a while, with its own date in additionalReviewAt and its own photos.
  • productTags are the impressions AliExpress sums up a product with (good quality, 414 buyers on the prefilled product).
  • position is the review's rank in the product's list under the sort and filter you chose, rating-only reviews included, so it stays comparable when you switch on withTextOnly. Past the first 5,000 reviews of a product, the rank goes on in the order the run found the rest.
  • publishedAt is the day of the review at midnight UTC in RFC 3339 (2026-02-25T00:00:00Z); scrapedAt is the moment the row was read.
  • productPrice, currency and productSoldCount come from the search card, so they are filled for products found by keyword or search link and null for products asked for by link or ID.

Output example

A real row from the keyword run of 30 Sep 2026 (wireless earbuds, latest first):

{
"id": "60097601874465190",
"url": "https://www.aliexpress.com/item/1005012524158014.html#review-60097601874465190",
"productId": "1005012524158014",
"productUrl": "https://www.aliexpress.com/item/1005012524158014.html",
"productTitle": "2026 New Air 3【 Pro 3 】 Bluetooth Wireless Earbuds with Heart Rate Monitoring, Active Noise Cancellation Headphones, Waterproof for Daily Use, For IPhone IOS Smartphone, Gaming /Sports /Fitness Earphones.",
"rating": 5,
"text": "Los compre para mi hija, a ella le gusto estan buenos llego antes de tiempo.",
"translatedText": "I bought them for my daughter, she liked them, they are good and arrived ahead of schedule.",
"isTranslated": true,
"variant": "Ships From:United States",
"skuId": null,
"isFeatured": true,
"hasImages": true,
"imagesCount": 1,
"additionalReview": "Llego bien, antes de tiempo, a mi hija le gusto",
"additionalReviewTranslated": "It arrived well, ahead of time, my daughter liked it.",
"additionalReviewAt": "2026-06-18T00:00:00Z",
"hasAdditionalReview": true,
"sellerReply": null,
"hasSellerReply": false,
"author": "AliExpress Shopper",
"authorCountryCode": "US",
"authorCountry": "United States",
"authorAvatarUrl": null,
"authorGender": null,
"isAnonymous": true,
"helpfulCount": 0,
"notHelpfulCount": 0,
"productRating": 4.9,
"productReviewsCount": 2,
"productRatingsCount": 10,
"productPositiveRate": 100,
"productFiveStarCount": 8,
"productFourStarCount": 2,
"productThreeStarCount": 0,
"productTwoStarCount": 0,
"productOneStarCount": 0,
"productPhotoReviewsCount": 2,
"productAdditionalReviewsCount": 1,
"productTags": null,
"productPrice": 21.09,
"currency": "USD",
"productSoldCount": 77,
"shippingMethod": "DGM Smartmail Expedited",
"publishedAt": "2026-06-18T00:00:00Z",
"images": [
"https://ae-pic-a1.aliexpress-media.com/kf/A4fda98e31c874fc198778792beda5952Y.jpg"
],
"additionalReviewImages": null,
"productImageUrl": "https://ae-pic-a1.aliexpress-media.com/kf/Sb5cfbb9b1c1b480a9f51b0cf05d8cc40e.jpg",
"productImages": [
"https://ae-pic-a1.aliexpress-media.com/kf/Sb5cfbb9b1c1b480a9f51b0cf05d8cc40e.jpg",
"https://ae-pic-a1.aliexpress-media.com/kf/Sd4e396cd1abe4e8abdd18af12559e993Y.jpg",
"https://ae-pic-a1.aliexpress-media.com/kf/S5aa455871a95472c965645aeb8b56df35.jpg",
"https://ae-pic-a1.aliexpress-media.com/kf/Sef791dacf016455691e6ffe4aaf7ddf7K.jpg",
"https://ae-pic-a1.aliexpress-media.com/kf/Sdac3379fc2ec4a1c8f6a5f99ac1b4794Y.png",
"https://ae-pic-a1.aliexpress-media.com/kf/S369d1fe2f3aa4cbebed651d49b188506P.jpg"
],
"position": 2,
"keyword": "wireless earbuds",
"scrapedAt": "2026-09-30T02:53:21Z"
}

Advanced usage

Watch the new reviews of your products every week

Most recent first, 50 reviews per product, on a schedule. Keep the id values you already have and every run tells you which reviews are new, before a bad one costs you a week of sales.

{
"productIds": ["1005007011478793", "1005005566622938"],
"sort": "latest",
"maxReviewsPerProduct": 50,
"maxItems": 100
}

Save it as a task, then add a schedule (Schedules tab, weekly) and a webhook or an integration to push the rows to Google Sheets, Slack or your database.

Pull every buyer photo of a product

Only the reviews with photos. Each row lists the full-size images in images, next to the variant the buyer ordered and the stars they gave: real pictures of the product in use, grouped by colour or size. The prefilled product has 537 photo reviews.

{
"startUrls": [
{ "url": "https://www.aliexpress.com/item/1005007011478793.html" }
],
"reviewFilter": "photos",
"maxItems": 600
}

Compare competing products review for review

The same number of reviews from each product, so averages and complaint shares compare fairly. Group the rows by productId, then by rating or variant.

{
"productIds": ["1005007011478793", "1005005566622938", "1005009267907429"],
"maxReviewsPerProduct": 200,
"withTextOnly": true,
"maxItems": 600
}

In our test three product IDs with the photo filter, latest first and written reviews only returned 80 reviews in 6 seconds; 19 rating-only reviews were left out of the dataset.

Read the reviews behind the search you ran in your browser

Sort a search by orders on AliExpress, copy the address and paste it as a link. The products are read in the order of that page, 10 reviews at most from each, so 100 reviews spread over the bestsellers of the niche instead of stopping at the first one.

{
"startUrls": [
{ "url": "https://www.aliexpress.com/w/wholesale-phone-case.html?SearchText=phone+case&SortType=total_tranpro_desc" }
],
"maxReviewsPerProduct": 10,
"maxItems": 100
}

Build a sentiment dataset in Python

A full product history in one run: in our test one product link with maxItems at 1,500 returned 1,500 unique reviews in 28 seconds. Keep the English text for every row that has one.

{
"startUrls": [
{ "url": "https://www.aliexpress.com/item/1005007592605567.html" }
],
"withTextOnly": true,
"maxItems": 1500
}
from apify_client import ApifyClient
client = ApifyClient("YOUR_APIFY_TOKEN")
run = client.actor("silentflow/aliexpress-reviews-scraper").call(run_input={
"startUrls": [{"url": "https://www.aliexpress.com/item/1005007592605567.html"}],
"withTextOnly": True,
"maxItems": 1500,
})
rows = []
for r in client.dataset(run["defaultDatasetId"]).iterate_items():
rows.append({"stars": r["rating"], "text": r["translatedText"] or r["text"], "variant": r["variant"], "country": r["authorCountry"]})
low = [row for row in rows if row["stars"] <= 2]
print(f"{len(rows)} written reviews, {len(low)} with 1 or 2 stars")

FAQ

What does this actor return? One row per AliExpress buyer review, with 52 fields: the review (id, stars, text, English translation, variant ordered, photos, additional review, seller reply, helpful votes, date, rank), the buyer as AliExpress shows them (masked name, country, avatar, gender) and the product it belongs to (title, images, rating, review and rating counts, star breakdown, photo reviews, tags, and from a search its price and units sold).

Do I need an AliExpress account? No. The actor reads the reviews any visitor sees on an AliExpress product page. There is no login, no API key and nothing to install.

Which inputs can I give it? Product links, search links, product IDs and keywords, all in the same run. Products asked for by link or ID come first; then each search is read product by product in the order AliExpress ranks them, until maxItems.

Why do some reviews have no text? Many AliExpress buyers leave stars without writing anything. Those reviews ship with text set to null. Switch on withTextOnly and they are left out of the dataset; the run reads further into the product to reach maxItems, and position still gives each review's rank among all of them.

Are the reviews translated? The original text is always in text. When AliExpress shows an English translation under a review, it is in translatedText and isTranslated is true. AliExpress adds those translations on its own schedule: on 30 Sep 2026 the first 100 reviews of the prefilled product came without one on a first run and all 100 carried one two hours later. Keep text as the source of truth and use translatedText when it is there.

Are the buyer names real? They are what AliExpress shows: masked names such as S***y, or AliExpress Shopper for buyers who chose to stay anonymous (isAnonymous is true). AliExpress never shows full names, order numbers or contact details, and neither does this actor.

How fresh is the data? Live. Every run reads AliExpress at the moment it runs; nothing comes from a cache. With sort set to latest, the first rows are the reviews posted since your last run.

How many reviews can I get from one product? Every review of a product that has up to 5,000 of them. AliExpress lists at most 5,000 reviews of a product in one order; when you ask for more with reviewFilter on all, the actor goes on star level by star level (each star level is its own list of up to 5,000) and reads a star level that is still longer in the other order too, skipping the reviews it already has. On the largest products some reviews stay out of every list AliExpress serves. In our test a product with 10,483 reviews returned 1,500 unique reviews in 28 seconds, and 5,600 unique reviews in 2 minutes 17 seconds: past the first 5,000, most pages repeat reviews already read, so each new review takes longer. productReviewsCount tells you how many written reviews a product has before you raise maxItems.

Do the filters combine? Yes. reviewFilter 1 with sort latest and withTextOnly gives the most recent written 1-star reviews. reviewFilter photos with maxReviewsPerProduct 20 and a keyword gives 20 photo reviews from each product of the search. AliExpress applies one review filter at a time, so a star level and the photo filter go in two runs.

Why do productReviewsCount and productRatingsCount differ? AliExpress counts every star rating a product received, and shows a written review only for part of them. On the prefilled product: 15,439 ratings, 3,055 reviews. The five star counts add up to productRatingsCount.

Why are price and units sold empty for my product links? Those two come from the AliExpress search card. A product found by keyword or search link carries its price, and its units sold whenever the card shows them (AliExpress sometimes serves a lighter card without them, and the field is then null); a product asked for by link or ID ships with its title, images and rating summary, and productPrice, currency and productSoldCount set to null. To get both, pair this actor with the AliExpress Scraper on the same products.

What is AliExpress? AliExpress (全球速卖通, 速卖通 for short) is Alibaba Group's international retail marketplace, where sellers, most of them in China, ship single items to buyers worldwide. Its reviews come from buyers who ordered the product, and they are the main quality signal dropshippers and resellers check before sourcing.

Are the links and images permanent? productUrl is the permanent product page, built from the product ID without tracking parameters. AliExpress gives a review no page of its own, so url is that product page anchored on the review ID: it opens the product and stays unique per review, handy as a key in a spreadsheet or a database. Review photos and product images point to the full-size files on AliExpress's own image servers and keep working as long as the buyer or seller keeps them online.

More SilentFlow scrapers

🛒 E-commerce

💼 Jobs

🏠 Real estate

📣 Social media

🏢 Business data

✈️ Travel

🎯 Lead generation

🔎 SEO

Support

Need something this scraper doesn't do yet? We ship features fast.

  • Feature requests go straight to our backlog
  • Enterprise needs? We do custom integrations and high-volume plans
  • Pricing questions? Check the Monetization tab on the actor page

Response time: usually under 24 hours.

This actor extracts publicly available data from AliExpress. It does not bypass any login, paywall or CAPTCHA. Users are responsible for complying with AliExpress's terms of service and applicable data protection laws (GDPR, CCPA, and PIPL for data about companies and people in China). Reviews carry personal data in a masked form (display name, country, avatar, gender): handle it accordingly, keep only what your purpose needs, and do not try to identify the buyers behind masked names. The data returned is informational; verify it on AliExpress before you rely on it for a sourcing or legal decision.

Built by SilentFlow for sellers, sourcing teams and analysts who would rather read every review than guess from a star average.