AliExpress Reviews Scraper

🚀 AliExpress Reviews Scraper

Extract customer reviews from any AliExpress product page — with the full review text, its English translation, star rating, date, buyer country, the exact variant purchased and the buyer's photos. Paste one or more AliExpress product URLs, choose how many reviews you want, and get clean structured data you can download as JSON, CSV or Excel, or pull straight from the Apify API.

Every record also carries the things most AliExpress scrapers leave behind: per-aspect ratings the buyer gave (e.g. "Quality of sound: Good", "Durability: Good"), follow-up reviews written weeks after delivery, and helpfulness votes. Turn on the optional product summary and you also get AliExpress's own buyer topic tags — "fast delivery", "excellent quality", "poor packaging" — with mention counts and sentiment, so you get aggregate sentiment without running any NLP yourself.

✨ Features

• 💬 Complete review text — the full review in the buyer's original language, never a truncated snippet.
• 🌍 English translations — AliExpress's translation of every non-English review, on the same record.
• ⭐ Ratings and dates — 1–5 star rating and review date on every record, ready for trend analysis.
• 📸 Buyer photos — direct full-resolution image URLs for the photos attached to a review.
• 🎨 Variant purchased — the exact SKU the review refers to (Color:black, Size:XL), so you can tell which option disappoints.
• 📐 Per-aspect ratings — the structured quality labels buyers pick, e.g. "Quality of sound", "Durability", "User Friendly".
• 🔁 Follow-up reviews — the second opinion buyers add after weeks of use, with its own date and day count.
• 🏷️ Buyer sentiment tags (optional) — AliExpress's mined topic tags with mention counts and positive/negative sentiment, plus average rating and full star breakdown, in one record per product.
• 🖼️ Photo-only and follow-up-only modes — collect just the reviews with buyer photos, or just the ones with follow-up feedback.
• 🚚 Shipping and buyer country — which shipping method was used and where the buyer is, on every review.
• ⚡ Fast and reliable — around 100 reviews per request with built-in retries, so a 1,000-review pull takes seconds.

💡 Why use AliExpress Reviews Scraper?

• Product research for dropshipping — read what buyers actually complain about before you add a product to your store.
• Supplier vetting — compare the review quality of the same product across competing AliExpress listings.
• Variant-level quality control — the variant field tells you whether the 1-star reviews cluster on one colour or size.
• Competitor and market research — mine thousands of real customer opinions about a category before entering it.
• AI and sentiment analysis — feed full review text into an LLM instead of scraping it yourself, or skip straight to the pre-mined sentiment tags.
• Product photo mining — collect real buyer photos for listings, ads or quality checks.

🛠️ How to use AliExpress Reviews Scraper

1. Add AliExpress product URLs — paste one or more links like https://www.aliexpress.com/item/3256812550576374.html. A bare product ID works too, and you can add as many as you want.
2. Set the review limit — choose how many reviews to collect per product (50 by default).
3. Choose which reviews to collect — all of them, only reviews with buyer photos, or only reviews with follow-up feedback.
4. Optionally include the product summary — turn this on for the average rating, star breakdown and buyer topic tags.
5. Run the Actor — start it from the Console, on a schedule, or via the API.
6. Download your data — export the dataset as JSON, CSV, Excel or XML, or send it straight to Google Sheets, Slack, Zapier, Make or your own webhook.

📥 Input

Field	Type	Description
productUrls	array	Required. AliExpress product page URLs (or bare product IDs) to scrape reviews from.
maxReviewsPerProduct	integer	Maximum reviews to collect per product. Default 50.
filter	string	all (default), image for reviews with photos only, additional for follow-up reviews only.
includeProductSummary	boolean	Also emit one record per product with ratings and buyer topic tags. Default false.
proxyConfiguration	object	Optional proxy settings. Not needed for normal runs.

Example input:

{
    "productUrls": [{ "url": "https://www.aliexpress.com/item/3256812550576374.html" }],
    "maxReviewsPerProduct": 50,
    "filter": "all",
    "includeProductSummary": false
}

📊 Sample output data

Each review is one dataset item. You can download the dataset in various formats such as JSON, HTML, CSV, or Excel.

[
    {
        "productId": "3256812550576374",
        "productUrl": "https://www.aliexpress.com/item/3256812550576374.html",
        "reviewId": "60096426953285053",
        "rating": 5,
        "date": "22 Apr 2026",
        "text": "These are actually awesome quality. The sound is great, pairing was simple. The value is insane.",
        "translatedText": "These are actually awesome quality. The sound is great, pairing was simple. The value is insane.",
        "reviewerName": "AliExpress Shopper",
        "reviewerCountry": "US",
        "isAnonymous": true,
        "variant": "Color:black",
        "images": ["https://ae01.alicdn.com/kf/Sa1b2c3d4e5f6.jpg"],
        "shipping": "AliExpress Selection Standard Shipping",
        "aspectRatings": [
            { "name": "Quality of sound", "value": "Good" },
            { "name": "Durability", "value": "Good" }
        ],
        "followUpText": "Still working perfectly after two months of daily use.",
        "followUpTranslatedText": "Still working perfectly after two months of daily use.",
        "followUpDate": "18 Jun 2026",
        "followUpAfterDays": 57,
        "helpfulVotes": 2,
        "unhelpfulVotes": 0,
        "isAiGenerated": false,
        "wouldBuyAgain": "Buy the same"
    }
]

With Include product review summary enabled, each product also produces one record like this:

[
    {
        "type": "productSummary",
        "productId": "3256812550576374",
        "productUrl": "https://www.aliexpress.com/item/3256812550576374.html",
        "averageRating": 4.9,
        "totalReviews": 1771,
        "totalRatings": 9944,
        "positiveRate": 92.5,
        "ratingBreakdown": {
            "fiveStar": 7892,
            "fourStar": 1311,
            "threeStar": 741,
            "twoStar": 0,
            "oneStar": 0
        },
        "sentimentTags": [
            { "tag": "well satisfied", "sentiment": "positive", "mentions": 430 },
            { "tag": "good sound", "sentiment": "positive", "mentions": 341 },
            { "tag": "fast delivery", "sentiment": "positive", "mentions": 218 }
        ]
    }
]

📋 Data fields

Review records

Field	Type	Description
productId	string	AliExpress product ID the review belongs to
productUrl	string	Product page URL
reviewId	string	Stable identifier for the review
rating	number	Star rating, 1–5
date	string	Date the review was posted
text	string	Full review text in the buyer's original language
translatedText	string	English translation of the review
reviewerName	string	Buyer's display name
reviewerCountry	string	Two-letter country code of the buyer
isAnonymous	boolean	Whether the buyer posted anonymously
variant	string	The SKU that was reviewed, e.g. Color:black
images	array	Full-resolution URLs of the buyer's photos
shipping	string	Shipping method used for the order
aspectRatings	array	Per-aspect labels and values chosen by the buyer
followUpText	string	Follow-up review added after further use
followUpTranslatedText	string	English translation of the follow-up review
followUpDate	string	Date of the follow-up review
followUpAfterDays	number	Days between the original review and the follow-up
helpfulVotes	number	How many shoppers marked the review helpful
unhelpfulVotes	number	How many shoppers marked the review unhelpful
isAiGenerated	boolean	AliExpress's own AI-generated-content flag
wouldBuyAgain	string	Buyer's repurchase intent, e.g. Buy the same

Product summary records (optional)

Field	Type	Description
averageRating	number	Overall star rating of the product
totalReviews	number	Number of written reviews
totalRatings	number	Number of star ratings, including those left without text
positiveRate	number	Share of positive ratings, in percent
ratingBreakdown	object	Count of 5, 4, 3, 2 and 1 star ratings
sentimentTags	array	Buyer topic tags with tag, sentiment and mentions count

💰 How much does it cost to scrape AliExpress reviews?

This Actor uses pay per event pricing, so you only pay for the data you actually receive — there is no charge for starting a run:

Event	Price
Review scraped	$0.002
Product summary scraped	$0.005

That works out to about $0.20 per 100 reviews and $2 per 1,000 reviews. Checking 20 products with 50 reviews each costs roughly $2 per run.

New Apify accounts include $5 in free usage every month on the free plan, which is enough for around 2,500 reviews before you pay anything.

⚙️ Tips and advanced options

• Keep the limit tight for a quick read. Reviews come back in AliExpress's own relevance order, which is weighted towards recent ones, so a limit of 20–30 already gives you a representative sample of a product for a few cents.
• Use the photo filter for image mining. With filter set to image, only reviews carrying buyer photos are collected and charged, so you don't pay for the 70% that have none.
• Follow-up reviews are the honest ones. filter: "additional" returns only the reviews buyers came back to update after weeks of use — the best signal for durability problems.
• Batch your products. Put every product URL into one run rather than starting a run per URL; products are fetched in parallel, so a batch finishes far faster than a queue of separate runs.
• Turn the summary on for a cheap overview. One product summary costs the same as 2.5 reviews and gives you the star breakdown plus pre-mined sentiment tags — a fast way to triage many products before pulling their full review text.
• Schedule it. Use Apify Schedules to run the Actor every morning and route new reviews to Slack, a Google Sheet or your own webhook via integrations.

❓ FAQ

Is scraping AliExpress reviews legal? This Actor collects only publicly available information — reviews that anyone can read on AliExpress without logging in. Web scraping of public data is generally legal, but how you use the data matters. Do not use it in ways that breach applicable privacy laws such as GDPR or CCPA, and consult a lawyer if you plan to process personal data. See Is web scraping legal? for a longer discussion.

How many reviews can I get from one product? Products with up to about 5,000 written reviews can be collected in full. Beyond that, AliExpress itself stops serving new records and starts repeating the ones it already returned, so a product advertising 13,000 reviews yields roughly 5,000 unique ones. The Actor detects the repetition, stops there, and never charges you twice for the same review.

In what order do reviews come back? In AliExpress's own default order, which is weighted towards recent and helpful reviews rather than strictly newest first — the same order you see on the product page. Every record carries a date field, so sort the dataset afterwards if you need exact chronology.

Do I get reviews in English? Yes. Reviews are returned in the buyer's original language in text, with AliExpress's English translation in translatedText. Reviews written in English have the same content in both fields.

Can I collect only reviews with photos? Yes — set filter to image and only reviews with buyer photos are collected and charged. additional does the same for reviews with follow-up feedback.

Can I filter reviews by star rating or keyword? Not at the source — AliExpress does not honour a rating filter on its review feed. Collect the reviews and filter the dataset afterwards; every record carries rating and the full text. If you only want an overview of the rating distribution, the product summary gives you the full star breakdown for the price of 2.5 reviews.

Do I need a proxy? No. Normal runs work without one, and proxy settings are there only for unusually large workloads.

Can I get reviews for many products at once? Yes. Add as many product URLs as you like to productUrls; they are all processed in the same run and written to the same dataset.

Something is not working or I need a different field. Report it on the Issues tab of this Actor and it will be looked at. If you need a tailored version or a different AliExpress dataset, custom solutions are available.

Start mining real buyer feedback from AliExpress today! 🚀