Amazon Best Sellers Scraper

Track Amazon Best Sellers rankings across any category. For each product the actor collects rank, ASIN, title, price, rating, review count, image URL, and the category it came from.

What it does

The actor opens any Amazon Best Sellers category page with a real Chromium browser (Playwright) behind a US residential proxy. Amazon serves bestseller data client-side via React — a plain HTTP request returns a bot-wall shell with no product data. The browser renders the full page, extracts all product cards in a single page.evaluate() call, and saves the results.

Each category shows up to 50 products per page across two pages (?pg=2). The actor scrapes both pages and stops as soon as maxItems is reached.

Output fields

Field	Type	Description
rank	integer	Bestseller rank within the category (1 = #1)
asin	string	Amazon Standard Identification Number
title	string	Product title
url	string	Canonical product page URL
imageUrl	string	Product image URL
price	string	Current price as displayed (e.g. $24.99)
rating	number	Average customer rating (0-5)
reviewCount	integer	Number of customer reviews
categoryName	string	Human-readable category name
categoryUrl	string	URL of the category page scraped
scrapedAt	string	ISO-8601 timestamp

Input

Parameter	Type	Default	Description
categoryUrl	string	https://www.amazon.com/gp/bestsellers/	Amazon Best Sellers category URL
maxItems	integer	—	Maximum number of products to collect

Examples

Top-level bestsellers (all categories):

{
    "categoryUrl": "https://www.amazon.com/gp/bestsellers/",
    "maxItems": 50
}

Electronics bestsellers:

{
    "categoryUrl": "https://www.amazon.com/gp/bestsellers/electronics/",
    "maxItems": 50
}

Books bestsellers:

{
    "categoryUrl": "https://www.amazon.com/gp/bestsellers/books/",
    "maxItems": 50
}

You can use any Amazon bestsellers URL — navigate to the category you want on Amazon and paste the URL.

Technical notes

• Browser: Playwright (Chromium) — required because Amazon renders bestseller content client-side
• Proxy: US residential (DataImpulse) — Amazon blocks datacenter IPs
• Memory: 2 GB recommended; minimum 1 GB
• Rate limiting: Concurrency is kept low (2 parallel pages) with 1-second delays to reduce detection risk
• Deduplication: Results are deduplicated by ASIN within a run

Limitations

• Scrapes one category URL per run (v0.1). Run the actor multiple times to collect data across categories.
• Amazon shows at most 100 products per category (50 per page x 2 pages). Set maxItems accordingly.
• Prices, ratings, and rankings change frequently — re-run regularly for fresh data.
• Amazon.com (US) only. International domains require a separate configuration.