Taobao Tmall 淘宝 Product Scraper - Prices, Sales & Shops

Taobao Scraper is a Taobao API that searches Taobao (淘宝) products by keyword and works as a Tmall scraper for Tmall (天猫) products, reads the questions and answers buyers post under an item (问大家), reads the buyer reviews of an item, lists every product of a shop, and returns clean, flat JSON records. For every searched product you get the price and discount price, the sales count, the shop name, ID and ratings, the seller level, where the item ships from, stock and images. It is built for sellers, sourcing teams, analysts and developers who need China e-commerce product data without a Taobao account. No Taobao account, API keys, proxies or cookies are needed.

Pricing is pay-per-event: $0.008 per product or question and $0.0045 per review, plus a $0.03 start fee per run. shopProducts mode charges $0.06 per shop instead. 100 products from one keyword cost about $0.83; 100 reviews of one item cost about $0.48.

What is Taobao?

Taobao (淘宝) is a large Chinese online marketplace run by Alibaba, where many independent shops sell to consumers. Tmall (天猫) is its sister platform for brand and official stores, and Tmall products also appear in Taobao search results. This Actor works on the Taobao search results page, so it covers both regular Taobao shops and Tmall stores.

What data can you get?

The Actor has three modes, chosen with mode:

• search (default) returns products matching keywords, with the full field set below.
• questions returns the questions under items (type is question), each with the answer Taobao shows first.
• shopProducts lists the best-selling products of shops, 60 per request, from a seller ID plus the shop ID. In its output, shopId holds the seller ID (as in the other modes), which is not the shop ID you enter. It returns the same fields as a search result except the shop name, ratings, seller level, location, stock and extra images, which stay empty.
• shop lists the products of shops (type is product). A shop listing carries only itemId, url, title, price, salesText, salesCount, shopId and imageUrl; the other product fields are present but empty (null).

In search and shop modes each dataset item is one product (type is product). The most useful fields come first.

Taobao product price and sales

Field	Description
price	List price in Chinese yuan (CNY), as a number
discountPrice	Price in CNY after the shop's current promotion; equals price when there is none
discountType	Name of the current promotion, for example 特价秒杀; null when there is no promotion
salesText	Buyer count as Taobao shows it, for example 1万+
salesCount	Lower bound of salesText as a number (1万+ becomes 10000)
commentCount	Review count; usually empty in search results
stock	Stock count shown on the listing

Taobao shop data

Field	Description
shopName	Shop name
shopId	Shop ID
isTmall	true when the shop is a Tmall store
sellerLevel	Seller level shown by Taobao
sellerPositiveRate	Positive-review rate of the seller, in percent
descriptionRating	Shop rating for item as described, out of 5
serviceRating	Shop rating for service, out of 5
shippingRating	Shop rating for shipping, out of 5
sellerLocation	Seller's location; empty for many shops

Taobao product data

Field	Description
itemId	Taobao item ID
url	Link to the item page on taobao.com
title	Item title (Chinese)
titleEn	English title; usually empty
subtitle	Selling-point line under the title; empty for many items
spuId	Standard product unit ID that groups the same product across shops
imageUrl	Main image URL
imageUrls	Further image URLs
searchKeyword	The search keyword this item was found with
type	Record type: product

Taobao delivery and ship-from location

Field	Description
itemLocation	Where the item ships from (province and city)

Taobao buyer questions and answers (mode: questions)

Each record is one question under an item (type is question).

Field	Description
question	Question text
questionId	Question ID
answerCount	Number of answers the question has
answers	The answer Taobao shows first, as a list of {answerId, text, answeredAt}. It is not the full thread, and no answerer details are returned.
askedAt	When the question was asked (ISO 8601, UTC)
updatedAt	When the thread last changed (ISO 8601, UTC)
itemId	ID of the item the question belongs to
itemTitle	Title of that item
url	Link to the question page
type	Record type: question

Reviews (mode: reviews)

Each record is one buyer review of an item (type is review). Reviewer fields are what Taobao shows publicly next to the review; many buyers review anonymously, and then Taobao masks the nickname (for example t***2) and gives no user ID.

Field	Description
text	Review text
rating	good, neutral or bad, as the buyer rated the item
score	Average of the buyer's star scores (1 to 5), when given
reviewedAt	When the review was written (ISO 8601, UTC)
sku	The variant the buyer bought, for example colour and phone model
imageCount, imageUrls	Number of photos in the review and their links
likeCount, usefulCount	Likes and useful votes the review received
followUpText, followUpDaysLater	The buyer's later follow-up review and the days after delivery it was written; empty when there is none
reviewerNickname, reviewerId, reviewerAvatar, reviewerLevel, reviewerAnonymous	The reviewer's public nickname, user ID (empty for anonymous reviews), avatar link, membership level, and whether the review is anonymous
reviewId	Review ID
productId	ID of the item the review belongs to
url	Link to the item page
type	Record type: review

The numbers are ready to use: prices are numbers in CNY, the sales count comes both as text as Taobao shows it and as a number, and shop ratings are plain numbers out of 5.

How to use Taobao Scraper

1. Open the Actor on Apify and click Try for free (or Start).
2. Choose a Mode: search products by keyword (search), questions and answers of items (questions), products of shops (shop), products of shops by seller ID and shop ID (shopProducts) or buyer reviews of items (reviews).
3. For search, enter one or more keywords, in Chinese or English, and optionally choose a sort order, turn on Tmall only, and set a minimum and maximum price. For questions or reviews, enter item IDs or item links in Items. For shop, enter seller IDs or shop links in Shops.
4. Set Max results to the total number of records (products, questions or reviews) you want for the run.
5. Click Start and wait for the run to finish.
6. Open the Products view in the dataset and export the data as JSON, CSV or Excel, or fetch it through the API.

The Actor also runs with an empty input: {} searches 手机壳 (phone cases), which is a quick way to see the output format.

A common workflow is to search a keyword first, then use the itemId of interesting products in questions mode to see what buyers ask, and the shopId of their shops in shop mode to list the shops' catalogues.

Input

Field	Type	Required	What it does	Default
mode	string	Optional	search (products by keyword), questions (questions and answers of items), shop (products of shops) or reviews (buyer reviews of items)	prefilled with search; when left out, questions if items is given, shop if shops is given, otherwise search. Giving both items and shops without mode is an error.
keywords	array of strings	Optional	Used in search mode. Search keywords, Chinese or English (for example 手机壳 or phone case). They are searched one after the other, and results are collected keyword by keyword until maxResults is reached. A later keyword is only searched if the earlier ones did not already fill maxResults.	Falls back to 手机壳 when empty
items	array of strings	Optional (needed for questions and reviews modes)	Item IDs (the itemId field of search results) or item links containing id=<digits>, for example https://item.taobao.com/item.htm?id=672400567264. Short links such as e.tb.cn are not supported. Questions or reviews are collected item by item until maxResults is reached.	none
shops	array of strings	Optional (needed for shop and shopProducts modes)	Seller IDs (the shopId field of search results) or shop links containing user_id=<digits> or user_number_id=<digits>. Products are listed shop by shop, newest first, until maxResults is reached. In shopProducts mode each entry needs both IDs: sellerId:shopId (for example 1659916565:103710607) or a shop link containing both user_id=<digits> and shop_id=<digits>; products come best-selling first.	none
sort	string	Optional	Used in search mode. sales (best selling), priceHighToLow (price, high to low) or general (Taobao's general ranking). Price low to high is not offered.	sales
tmallOnly	boolean	Optional	Used in search mode. Only products from Tmall stores.	false
minPrice	number	Optional	Used in search mode. Only products whose price after discount (discountPrice) is at least this many yuan (inclusive); the list price (price) can be higher. Minimum 0.	none
maxPrice	number	Optional	Used in search mode. Only products whose price after discount (discountPrice) is at most this many yuan (inclusive); the list price (price) can be higher. Minimum 0.	none
maxResults	integer	Optional	Maximum number of records (products or questions) to return in total for this run, from 1 to 5000. A single keyword returns at most about 1000 products, so higher totals need several keywords. The run stops as soon as it is reached. You are charged per record returned.	20

Example: best-selling Taobao products for a keyword

{
  "keywords": ["手机壳"],
  "sort": "sales",
  "maxResults": 100
}

Example: Tmall products within a price range

{
  "keywords": ["蓝牙耳机"],
  "sort": "sales",
  "tmallOnly": true,
  "minPrice": 20,
  "maxPrice": 200,
  "maxResults": 100
}

Example: English keyword on Taobao, most expensive first

{
  "keywords": ["phone case"],
  "sort": "priceHighToLow",
  "maxResults": 50
}

Example: what Taobao buyers ask under an item

{
  "mode": "questions",
  "items": ["672400567264"],
  "maxResults": 50
}

Example: buyer reviews of an item

{
  "mode": "reviews",
  "items": ["672400567264"],
  "maxResults": 100
}

Example: the newest products of a Taobao shop

{
  "mode": "shop",
  "shops": ["1659916565"],
  "maxResults": 100
}

When you give several keywords, they are searched in order. The run only moves on to the next keyword when the earlier ones did not fill maxResults. With a small maxResults, the first keyword often fills it, and the later keywords are never searched.

Output example

One product record. Values are illustrative.

{
  "type": "product",
  "itemId": "123456789012",
  "url": "https://item.taobao.com/item.htm?id=123456789012",
  "searchKeyword": "手机壳",
  "title": "透明磁吸手机壳 适用iPhone 防摔全包超薄保护套",
  "titleEn": null,
  "subtitle": "镜头全包 不易发黄",
  "price": 59.0,
  "discountPrice": 49.0,
  "discountType": "特价秒杀",
  "salesText": "1万+",
  "salesCount": 10000,
  "commentCount": null,
  "isTmall": true,
  "shopId": "1000000001",
  "shopName": "示例数码旗舰店",
  "sellerLevel": 19,
  "sellerPositiveRate": 100.0,
  "descriptionRating": 4.84,
  "serviceRating": 4.88,
  "shippingRating": 4.9,
  "itemLocation": "广东 东莞",
  "sellerLocation": "广东 深圳",
  "stock": 25000,
  "imageUrl": "https://img.alicdn.com/imgextra/i1/1000000001/example-main.jpg",
  "imageUrls": ["https://img.alicdn.com/imgextra/i1/1000000001/example-2.jpg"],
  "spuId": "8000000001"
}

Question record (values are illustrative):

{
  "type": "question",
  "itemId": "123456789012",
  "itemTitle": "透明磁吸手机壳 适用iPhone 防摔全包超薄保护套",
  "questionId": "585847296732",
  "question": "质量怎么样？会刮花手机吗",
  "askedAt": "2026-10-13T03:19:20Z",
  "updatedAt": "2026-10-14T12:35:16Z",
  "answerCount": 4,
  "answers": [{"answerId": "585287107305", "text": "不会", "answeredAt": "2026-10-14T12:35:05Z"}],
  "url": "https://web.m.taobao.com/app/mtb/ask-everyone/detail?id=585847296732"
}

How much does it cost?

The Actor uses pay-per-event pricing. These are the charge events:

Event	Price (USD)	When it is charged
run-start	$0.03	Once per run, charged when the first request succeeds, even if that request returns nothing. It is not charged when the run fails from the start, or when your spending limit cannot cover it. Each further keyword, item or shop the run actually reaches pays one more start fee. That fee is charged before its first request and is kept even if it then fails or returns nothing. A later keyword, item or shop is only reached when the earlier ones did not fill maxResults.
result	$0.008	For each record returned: a product (search, shop and shopProducts modes) or a question (questions mode).
review	$0.0045	For each review returned (reviews mode).

In shopProducts mode every shop pays two start fees ($0.06), because a shop request costs more. A shop's next page of 60 is only requested when the shop's fees and products already cover it. A shop request that was sent but got no answer is not repeated and the shop ends there; its two start fees are still charged, because the request may have been processed.

Worked examples:

• 100 products from one keyword cost $0.03 + 100 × $0.008 = $0.83.

• 100 products from two keywords, when the first keyword returns fewer than 100 and the run moves on to the second, cost 2 × $0.03 + 100 × $0.008 = $0.86. If the first keyword already returns 100 products, the second keyword is not searched and the run costs $0.83.

• 100 questions from one item cost $0.03 + 100 × $0.008 = $0.83. An item with fewer than 100 questions costs less, because only the questions returned are charged.

• 100 reviews of one item cost $0.03 + 100 × $0.0045 = $0.48. Reviews come 20 to a page, newest first; the run stops at the end of the item's reviews or when maxResults is reached. An item that returns no reviews at all still costs its start fee, and the next item still runs.

• 100 products from one shop cost $0.03 + 100 × $0.008 = $0.83. Two shops that together fill 100 products cost $0.06 + 100 × $0.008 = $0.86.

• 100 products from one shop in shopProducts mode cost 2 × $0.03 + 100 × $0.008 = $0.86; each further shop adds $0.06.

You never pay for more than maxResults items, and you are never charged for items that were not delivered. The run stops by itself if your spending limit is reached.

On the free Apify plan you get 2 free trial runs of this Actor with your own input, up to 10 results each, and nothing is charged; after that, runs return a stored sample dataset (a real result of the prefilled input). Use any paid Apify plan for full runs.

Use cases

• Price monitoring and competitor research: see how a product category is priced, which promotions run (discountType, discountPrice) and who sells the most (salesCount).
• Sourcing and product selection: find best-selling items per keyword, with shop quality at a glance through descriptionRating, serviceRating, shippingRating and sellerPositiveRate.
• Market research for cross-border sellers: see what sells on Taobao and Tmall before you list or source it.
• Brand and Tmall store tracking: use tmallOnly and isTmall to compare official Tmall stores with regular Taobao shops.
• Regional supply analysis: group products by itemLocation to see where items in a category ship from.
• AI and data projects: clean product records for catalogs, price models and recommendations.

Integrations and API

You can start runs from your own code through the Apify API, schedule them on the Apify platform (for example a daily price check), and send results to other tools with Apify integrations and webhooks. Results are stored in a dataset you can download as JSON, CSV or Excel.

Python example using the apify-client package:

from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")
run = client.actor("clover_folktale/taobao-scraper").call(run_input={
    "keywords": ["蓝牙耳机"],
    "sort": "sales",
    "tmallOnly": True,
    "minPrice": 20,
    "maxPrice": 200,
    "maxResults": 100,
})
for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item)

Tips and limits

• In shop mode there is no shop name, so shopName is empty there; shopId is filled.

• Reliability. The Actor retries transient errors on its own and stops cleanly instead of failing half-way. It never charges for items it did not deliver.

• Speed. A search page holds 10 products, a questions page 10 questions and a shop page about 30 products, so large runs make many requests; expect roughly 2-5 seconds per page. Different pages of one search do not overlap.

• Run size. maxResults is the total for the whole run (1 to 5000, default 20). A single keyword returns at most about 1000 products (often fewer, when the search runs out of pages), so to get more you need several keywords. With several keywords, results are collected keyword by keyword until the total is reached, and later keywords are not searched once the total is filled.

• Sales counts. salesCount is the lower bound of the sales text: 1万+ becomes 10000, 1000+ becomes 1000. Taobao only gives rounded buyer counts in search results.

• Price filters. minPrice and maxPrice are in yuan and both inclusive. A narrow price range can return no products at all; widen the range if that happens.

• Nothing found. A search that finds nothing is skipped.

• Currency. Prices are in Chinese yuan (CNY) and are what Taobao shows at the time of the run.

• Sorting. Price low to high is not offered; use priceHighToLow, sales or general.

• Empty fields. commentCount and titleEn (the English title) are empty in search results. subtitle and sellerLocation are empty for many products. discountType is null when the product has no promotion.

• Questions. Each question carries the answer Taobao shows first, not the whole thread; answerCount says how many answers exist. An item without questions is skipped and counts as nothing found.

• Shops. Use the seller ID (the shopId field of search results). Products come newest first. A shop that Taobao does not know, or that has no products, is skipped as nothing found.

• Unreadable entries. If one entry of an ID or link list cannot be read (a typo, a stray cell, a link to the wrong kind of page), it is skipped and named in the log, and the status message says how many were skipped; the other entries still run and nothing is charged for a skipped one. Only a list with no readable entry stops the run with an input error.

• Free plan. On the free Apify plan you get 2 free trial runs of this Actor with your own input, up to 10 results each, and nothing is charged; after that, runs return a stored sample dataset (a real result of the prefilled input). Use any paid Apify plan for full runs.

• Privacy. Only public listing data is returned: what anyone sees on a Taobao search page, such as the shop name, its ratings and the item's ship-from location. No buyer data, no private shop data.

FAQ

Do I need a Taobao account, cookies or an API key?

No. No Taobao account, API keys, proxies or cookies are needed. You only need an Apify account to run the Actor.

Is this the official Taobao API?

No. This Actor is not affiliated with Taobao or Tmall. It is an independent tool that collects public search-result data and returns it as structured JSON, so you can use it as an alternative to building your own Taobao product search integration.

Is it legal to scrape Taobao?

The Actor only returns public listing data that anyone can see on a Taobao search page. It does not return buyer data, private shop data, phone numbers or messages. You are responsible for how you use the data and for following the laws and rules that apply to you.

How fresh is the data?

The data is collected live during each run. Prices, sales counts and stock are what Taobao shows at the time of the run.

Why are some fields empty?

Taobao search results do not include every field for every product, and a shop listing carries fewer fields than a search result. commentCount and titleEn are empty in search results, subtitle and sellerLocation are empty for many products, and discountType is null when there is no promotion.

Can I get SKUs or full product details?

Not yet. The Actor returns keyword search results, the questions and buyer reviews under items and the products of shops, with the fields listed above. The full SKU list and full product details are not included; each review does name the variant that buyer bought (sku).

How many results can I get?

You can set maxResults up to 5000 per run, but a single keyword returns at most about 1000 products, often fewer. To get more, add more keywords.

Can I schedule runs and export the data?

Yes. You can schedule runs on Apify, start them through the API, and export the dataset as JSON, CSV or Excel.

Something is missing. How do I ask for it?

Missing a field or a filter? Open an issue from the Actor page and it will be looked at.

Other tools by Silkline

Every tool below charges per result and needs no login or cookies.

E-commerce

• 1688.com 阿里巴巴 Wholesale Scraper - MOQ Prices & Suppliers
• JD.com 京东 Scraper - Shop Products & Buyer Reviews
• Douyin Shop 抖音商城 Product Scraper - Sales, Commission & Shops
• TikTok Shop Scraper - Products, Sales, Reviews & Hot Sellers
• AliExpress 速卖通 Scraper - Products, Prices & Sold Counts
• Amazon Product Scraper - Best Sellers, ASIN, Prices, Ratings

More from Silkline

• Xiaohongshu 小红书 RedNote Scraper - Notes, Comments & Replies (Chinese social media and content)
• TikTok Scraper - Video Search, Hashtags, Profiles & Sounds (Global platforms)
• Douyin Xingtu 巨量星图 KOL Scraper - Rates, CPM & Rankings (Marketing and creator data)