TikTok Shop Trending Scraper

The Actor reads TikTok Shop's homepage for a chosen storefront and collects every product rail on it in one run, from flash sales and premium offers to whatever else TikTok is featuring that day, tagging each row with the rail it came from. There is no scraperType: the homepage is a single page carrying several rails at once, not a set of alternative surfaces to pick between.

{
  "region": "US",
  "maxItems": 100,
  "enrichProductDetails": false
}

Accepted input

region is required, defaults to US, and selects the storefront homepage read (US, GB, SG, MY, PH, TH, or VN); this changes the products, prices, and currency returned, not only the display language. maxItems defaults to 100 and bounds the number of rows saved across every rail combined; a value of 0 removes the bound. enrichProductDetails defaults to false and adds the full product page to every row: description, every variant, the shop profile, a page of reviews with the star breakdown, and the category path. It costs one extra request per row.

Result fields

Every row carries product_id and rail, naming the carousel the product was read from. rail values are derived from TikTok's own container names for each carousel; flash_sale_banner and premium_offers are two seen on the US storefront. component_id is TikTok's own identifier for the homepage rail a row came from, and it is present only on the rails the Actor collects through a separate hydration request; rails read from the homepage's plain server-rendered response carry no component_id. sale_end_time is set on rows carrying a time-boxed offer and is absent otherwise.

The rest of the row is the standard listing card: title, image, product_price_info, rate_info, sold_info, seller_info, seo_url, sku_info, brand_info, product_marketing_info, and product_card for the untouched original. brand_info is not filled on every card; TikTok attaches it to some listings and not others even within the same rail.

{
  "product_id": "1729668107411035024",
  "rail": "flash_sale_banner",
  "title": "Molitree Women's Sherpa Fleece Overalls 2025 Warm Winter Casual Loose Jumpsuits Sherpa Bib Overalls Fuzzy Ski Pants with Pockets Basic Womenswear Coveralls Strap",
  "product_price_info": { "currency_name": "USD", "sale_price_format": "60.16", "discount_format": "30%" },
  "rate_info": { "score": 4.6, "review_count": "571" },
  "sold_info": { "sold_count": 8536 },
  "seller_info": { "seller_id": "7495810325667482512", "shop_name": "Molitree" },
  "sale_end_time": "2145916800"
}

A row from premium_offers on the same US homepage carried a brand_info value the flash-sale row above did not:

{
  "product_id": "1729385034800730797",
  "rail": "premium_offers",
  "title": "Micro Ingredients Organic Spirulina and Chlorella Tablets",
  "brand_info": { "brand_name": "Micro Ingredients" },
  "rate_info": { "score": 4.7, "review_count": "1230" },
  "sold_info": { "sold_count": 18995 },
  "sale_end_time": "1785999599"
}

These are trimmed, live-verified rows; product_price_info, rate_info, sold_info, and seller_info each carry more fields than shown here. Numeric values arrive exactly as the source returns them, so sale_end_time and review_count are numeric strings while score and sold_count are numbers; new Date(Number(sale_end_time) * 1000) converts the offer deadline, which is a second-precision epoch rather than the millisecond one review timestamps use.

With enrichProductDetails set to true, every row also gains name, shop_info, review_info, categories, and promotion_tag, plus the full product_detail record.

A rail that is missing rather than empty

The homepage carries three groups of rails, read in order. Server-rendered rails are read first, and when a rail in this group is populated it is already complete and costs nothing extra to collect; for a signed-out request these rails commonly come back with no products at all, and an empty rail contributes no rows. deals rails are US-only and are simply absent from the response on every other storefront, rather than present and empty. The remaining rails come from a separate request that does answer a signed-out shopper and carry TikTok's own component_id; a personalized rail that requires a signed-in shopper is simply absent from this group rather than returned present-but-empty. None of this is a fetch failure; the run finishes normally either way, with only the rails TikTok actually populated contributing rows.

Whether maxItems drives further requests

It does not. The homepage is one response covering every rail; there is no further page to ask for. maxItems bounds the number of rows saved across all rails combined, in the order the Actor reads them (server-rendered rails first, then deals, then the hydrated rails), so a low maxItems favors whichever rails come earliest rather than sampling evenly across all of them.

Related Actors

For a single product's full record rather than a homepage snapshot, use the TikTok Shop Product Scraper. For discovering products by keyword instead of by what is currently featured, use the TikTok Shop Search Scraper. For the recommendation carousels attached to one product page rather than the homepage, use the TikTok Shop Recommendations Scraper. For what surrounds a keyword search (related terms, ranking shops, or search-page recommendations), use the TikTok Shop Search Insights Scraper.