Shopee Scraper

Shopee Scraper collects products from keyword searches, category pages and individual product links across eight Shopee markets, for price monitoring, assortment research and marketplace analysis. Get one dataset row per product with its name, identifiers, link, price, currency, discount, rating and seller location when Shopee provides them. Optional details add product descriptions, images, brands and seller attributes.

You do not need a Shopee seller account, an API key or your own login. Enter search terms such as headphone or sepatu, paste category or product links, choose a market and download the result as JSON, CSV, Excel or through the Apify API. The actor is a practical way to scrape Shopee data when the official Shopee API only covers your own shop.

Shopee Scraper input

Choose a market with market: Indonesia (id), Singapore (sg), Malaysia (my), Thailand (th), Vietnam (vn), Philippines (ph), Taiwan (tw) or Brazil (br). The default remains Indonesia. Use terms in the market's own language for useful results. All existing input names and defaults remain available, so saved tasks and API integrations can keep their current configuration.

Provide searchTerms, categoryUrls, productUrls, or a combination. Searches run first, followed by categories and individual products. Each list accepts strings. API input also accepts comma-separated or newline-separated text. Category URLs can be full Shopee links or paths such as /Elektronik-cat.11044258. Individual product URLs accept both seller-title links ending in -i.shop.item and canonical /product/shop/item links.

pagesPerSearch controls how many pages to read per term, from one to twenty. Its default is one. maxProducts caps the entire run at between one and 5000 unique products; the default is 100. Products appearing in multiple searches or categories are delivered once, with the first source retained. Reaching the cap ends collection. A deeper search does not guarantee a fixed number of additional products because Shopee can repeat results between pages.

fetchDetails defaults to false. Enable it when descriptions, product images, brands or category identifiers matter to your project. concurrency sets parallel detail requests, from one to eight, with a default of three. attempts remains available from one to fifteen, with a default of four. Larger detail requests take longer than collecting listing cards alone.

Output and product fields

The default dataset contains one row for each delivered product. Existing fields retain their names and JSON types: shop_id, item_id, product_url, title, price, price_original, currency, discount_percent, rating, location, list_position, market, source, source_input and position.

Prices are numeric amounts in the row's stated currency. They are not converted into a different currency. Indonesian rupiah and Singapore dollars use different number separators on Shopee; the scraper normalises those separators while retaining the currency. A missing price is null, not zero. Original prices and discounts are populated only when the source publishes them.

With details enabled, rows may also include description, image, brand, brand_id, cat_id, condition, ctime, shop_location, item_status, images and attributes. Availability depends on the product page. Some pages expose a product preview with a name, description and image but omit brand or category data. Those limitations are reported in the run summary. sold_count and stock remain null in the anonymous detail route.

Additional fields help join the output with other product datasets: product_id, url, country, country_code, fetched_at and image_url. The original fields stay present. Use fetched_at to record when a row was collected and the uppercase country fields to identify its market. source continues to identify a search, category, individual URL or session API result.

Example Shopee Scraper output

A real row from a Singapore search for headphone (market: "sg", searchTerms: ["headphone"]):

{
  "shop_id": "1800354109",
  "item_id": "53609296010",
  "product_url": "https://shopee.sg/product/1800354109/53609296010",
  "list_position": 1,
  "market": "sg",
  "title": "Foldable Wireless Headphones Heavy Bass Bluetooth Gaming Headset Full Ear Gaming Card",
  "price": 9.0,
  "currency": "SGD",
  "discount_percent": 56,
  "rating": 4.7,
  "location": "SG",
  "price_original": null,
  "source": "search",
  "source_input": "headphone",
  "position": 1,
  "fetched_at": "2026-10-04T17:08:30.994238Z",
  "country_code": "SG",
  "country": "SG",
  "product_id": "53609296010",
  "url": "https://shopee.sg/product/1800354109/53609296010",
  "image_url": null
}

price is the current selling price in Singapore dollars, discount_percent is the discount Shopee shows on the card, and price_original is null because this card did not publish a struck-through price. image_url is filled when details are enabled.

Use cases for Shopee data

• Price monitoring. Schedule the same searches or product links daily and compare price, price_original and discount_percent over time to follow price changes and promotions on Shopee.
• Competitor and assortment research. Collect every product a category page shows and see which shops, price bands and seller locations dominate a niche in a given market.
• Market entry across Southeast Asia. Run the same term in Indonesia, Malaysia, Thailand, Vietnam, the Philippines and Singapore to compare price levels and supply before launching a product.
• Product research for resellers. Combine ratings, discounts and seller locations to shortlist products that sell at a healthy price.
• Catalogue enrichment. Enable fetchDetails to add brands, category identifiers, descriptions and images to a list of product links you already track.

Shopee Scraper compared with other options

The Shopee API (Shopee Open Platform) is meant for sellers and partners: it needs an approved app and authorisation from each shop, and it returns that shop's own orders and listings. It does not let you search the whole marketplace. Shopee Scraper works on the public catalogue, so you can look at any search term, category or product without a seller relationship.

Writing your own scraper is possible, but Shopee changes its pages often and each market uses its own domain, currency and number format. This actor handles the eight markets, normalises prices, removes duplicate products and keeps the output fields stable, so saved tasks and integrations continue to work.

Compared with single-purpose tools, one run can mix searches, category pages and product links under a single product cap, and you only pay for products that are actually delivered.

Optional session cookies

Prices on ordinary listing pages do not require your own session cookies. If you supply sessionCookies, the actor first tries Shopee's session API for each search page. Use only cookies from your own Shopee account, supplied as name/value pairs or a cookie string. A successful session response can include additional price ranges, sold counts and stock information.

Expired or rejected cookies produce session_refused, followed by the ordinary listing route. The run can still deliver the publicly available catalogue and prices. Cookies are not included in result rows, error messages or continuation output. They are optional; leave this field empty when you do not need the session route.

Shopee Scraper pricing

The existing pricing remains unchanged: $0.002 per run plus $0.001 per delivered product. The start fee applies once, including a run with no products. Product events apply only to rows actually delivered to the dataset. An empty result has no product charge. Set maxProducts to control the maximum product volume. Non-paying users have the existing ten-result run cap; rows beyond the cap are not charged.

Errors and run summary

The default key-value store retains summary and, when needed, errors. The summary includes product counts, counts with titles, prices, ratings and brands, the selected market, detail and session settings, elapsed collection time and error count. Additional partial and warnings fields explain incomplete collection or limited detail coverage.

A page that cannot be read produces blocked; a failed detail page produces detail_failed. Listing products retain their existing name and price when optional details fail. A product-only link that yields no usable product information is not delivered as an empty paid result. Missing search terms and URLs produce bad_input. These ordinary source errors complete as successful runs, retaining any real products already collected.

Frequently asked questions

Does every product have a price?

Most listing cards carry one, but Shopee can omit a price for a particular placement. The scraper retains the real product and returns null for unavailable amounts.

Why do some products have no brand or attributes?

Shopee does not expose the same detail coverage for every product. Enable fetchDetails, then check the summary warnings. Missing values are preserved as missing values rather than inferred from the title.

Can I combine searches and product links?

Yes. Searches, categories and product links share one run-wide product cap and deduplication rule. The first occurrence determines source and source_input.

What happens when a search has no useful matches?

Shopee sometimes returns popular recommendations even for an unusual term. The actor preserves the products that the source actually serves. It does not invent relevance scores or guarantee that every returned product matches the literal query.

Do I need a Shopee account or API key?

No. Searches, categories and product links work without an account. Session cookies are optional and only add fields that Shopee shows to signed-in users.

Which Shopee markets are supported?

Indonesia, Singapore, Malaysia, Thailand, Vietnam, the Philippines, Taiwan and Brazil. Each row carries market, country_code and the currency of that market.