Amazon Best Sellers Scraper & API - Top 100 Products by Category, New Releases, Any Marketplace

Amazon Best Sellers Scraper is an Amazon bestseller list scraper and unofficial Amazon Best Sellers API: give it any Best Sellers or New Releases page from amazon.com, amazon.co.uk, amazon.de, amazon.co.jp, amazon.in or another Amazon marketplace, and get the full top 100 as clean JSON, CSV or Excel: rank, ASIN, title, price, currency, star rating, number of ratings, image and product URL, plus the list name and category path. Turn on includeSubcategories to pull every subcategory list in one run.

It reads the data Amazon already sends with the page over plain HTTP, with no headless browser, so a 100-product list takes a few seconds and costs very little to run. Amazon shows only the first 30 products of each page and loads the other 20 when you scroll; this scraper fetches those too, so you get all 50 per page and all 100 per list.

How to use

1. Paste one or more Amazon list URLs (Best Sellers or New Releases pages, any marketplace), e.g. https://www.amazon.com/Best-Sellers-Electronics/zgbs/electronics/. Or enter Category IDs like electronics or electronics/172541 with a domain and list type.
2. Set Max products per list (1-100). Turn on Also scrape subcategories to add each list's child categories (one level down).
3. Click Start. Download the results as JSON, CSV, Excel or HTML from the Output tab, or pull them through the Apify API, Make, Zapier, n8n or an MCP client.

What you get

• Rank and identity: rank (1-100), ASIN, product URL (https://www.amazon.<tld>/dp/ASIN), the highest-resolution image Amazon serves on the list.
• Price: numeric price and ISO currency (USD, GBP, EUR, JPY, INR, CAD, ...), plus the raw priceText exactly as shown ("$11.99", "20,46 €", "4 offers from $12.99").
• Reviews: star rating (0-5) and number of ratings.
• Context: list type (bestsellers / new-releases), list name ("Best Sellers in Electronics"), category id and node id, category path ("Any Department > Electronics"), list URL, marketplace, page number and scrapedAt timestamp.
• Clean and deduplicated: one row per product per list, ranks in order, no repeats across the two pages.

Use cases

• Product research and sourcing: what sells in a category right now, at which price points and rating levels, on each marketplace.
• Price and rank tracking: schedule a daily run per category and build rank histories by ASIN.
• New-product monitoring: new-releases lists show what just launched or is about to.
• Competitive intelligence: see where your products and your competitors' rank, across countries, in one dataset.
• Datasets for AI agents: an agent can ask for "top 50 headphones on amazon.de" and get structured rows back through MCP.

Input example

Field	Default	Description
categoryUrls	Electronics best sellers (.com)	Best Sellers or New Releases URLs, any marketplace. A plain category URL with node= is converted using listType.
categoryIds	-	Alternative to URLs: electronics, books, electronics/172541
domain	amazon.com	Marketplace for categoryIds
listType	bestsellers	bestsellers or new-releases, for categoryIds and plain category URLs
maxItemsPerList	100	1-100 (Amazon shows 100 per list, in two pages of 50)
includeSubcategories	false	Also scrape each list's child categories (one level)
maxSubcategories	50	Cap on added subcategory lists
proxyConfiguration	Residential	Apify residential proxies in the marketplace's country (recommended)

Example: Electronics and Books top 100 on amazon.com, plus New Releases in Electronics on amazon.de:

{
    "categoryUrls": [
        "https://www.amazon.com/Best-Sellers-Electronics/zgbs/electronics/",
        "https://www.amazon.com/Best-Sellers-Books/zgbs/books/",
        "https://www.amazon.de/gp/new-releases/ce-de/"
    ],
    "maxItemsPerList": 100
}

Example: every Electronics subcategory on amazon.co.uk, top 50 each:

{
    "categoryIds": ["electronics"],
    "domain": "amazon.co.uk",
    "listType": "bestsellers",
    "maxItemsPerList": 50,
    "includeSubcategories": true
}

Output example

One dataset item per ranked product. From a real run (Best Sellers in Electronics, amazon.com):

{
    "rank": 1,
    "asin": "B08JHCVHTY",
    "title": "blink plus plan with monthly auto-renewal",
    "price": 11.99,
    "priceText": "$11.99",
    "currency": "USD",
    "rating": 4.4,
    "ratingsCount": 280900,
    "imageUrl": "https://images-na.ssl-images-amazon.com/images/I/31YHGbJsldL._AC_UL900_SR900,600_.png",
    "url": "https://www.amazon.com/dp/B08JHCVHTY",
    "listType": "bestsellers",
    "listName": "Best Sellers in Electronics",
    "categoryId": "electronics",
    "categoryNodeId": "electronics",
    "categoryPath": "Any Department > Electronics",
    "listUrl": "https://www.amazon.com/gp/bestsellers/electronics",
    "marketplace": "amazon.com",
    "page": 1,
    "scrapedAt": "2026-10-08T00:36:48.481Z"
}

Fill rates on a 200-product test (amazon.com Best Sellers + amazon.de New Releases, Electronics): rank, ASIN, title, rating, ratings count, image, URL, list name and category path 100%; price 97% (the rest are items Amazon lists without a price, such as subscriptions or out-of-stock products).

Pricing

Pay per result: $1.50 per 1,000 results (one result = one ranked product saved to the dataset).

• One full list (100 products) = $0.15
• 100 lists (10,000 products) = $15

You're never charged for failed requests or duplicates. If you set a maximum cost per run, the scraper stops cleanly when it reaches it. Apify's free plan includes $5 of monthly platform credit, enough for thousands of products. Residential proxy traffic is billed separately by Apify (about 0.5 MB per list page).

Integrations

• Make, Zapier and n8n: start runs and send new rankings to Sheets, Slack, Airtable or your database.
• Google Sheets: export the dataset straight into a spreadsheet, or refresh it on a schedule.
• Apify API: run the actor and fetch results over REST, or with the official JavaScript and Python clients.
• Webhooks: get notified when a run finishes and process the data right away.
• Schedules: run the same lists daily or hourly to track rank and price over time.
• MCP for AI agents: through the Apify MCP server (https://mcp.apify.com), Claude, ChatGPT, Cursor and other agents can call this Amazon Best Sellers API directly and read the results.

Limits (read before large runs)

• 100 products per list. That's all Amazon publishes (two pages of 50). To get more, use subcategories: Electronics alone has about 18 direct subcategories, each with its own top 100.
• Best Sellers and New Releases only. Amazon's Movers & Shakers, Most Wished For and Most Gifted pages render without a product grid at the moment; URLs of that kind are accepted but usually return no products, and the run summary says so.
• List prices are what Amazon shows on the list page: the default offer, in the marketplace's currency. Some items (subscriptions, out-of-stock products, some media) show no price and get price: null.
• Rankings change hourly. Two runs minutes apart can differ. Use scrapedAt and rank to build time series.
• Residential proxies are the default because Amazon blocks most datacenter IP ranges. Switching to datacenter proxies works but means more retries and occasional failed lists.
• No product details (bullet points, descriptions, variations, seller, BSR in other categories). This actor covers the ranked lists; pair it with a product-detail scraper when you need the full page.

FAQ

Is it legal to scrape Amazon Best Sellers? This actor collects only publicly available product listing data, not personal data. You're responsible for using it in line with Amazon's terms of service and applicable law. This is not legal advice: if you're unsure, check with a lawyer.

How does it avoid blocks? Requests go through Apify Proxy (residential by default, in the marketplace's country) with realistic Chrome headers and a session pool. When Amazon returns a captcha, a 503 or an error page, that session is retired and the page is retried on a fresh IP, up to 8 times by default. Other errors are retried with exponential backoff.

Why do I get 50 products when Amazon's page shows 30? Amazon renders 30 cards and loads the remaining 20 when you scroll. The scraper makes that same background call, so each page yields all 50 products.

Which marketplaces work? Any amazon.* domain: .com, .co.uk, .de, .fr, .it, .es, .nl, .se, .pl, .be, .ca, .com.mx, .com.br, .com.au, .co.jp, .in, .sg, .ae, .sa, .eg, .com.tr. Prices are parsed in each locale's format (20,46 € and $1,234.56 both work) and currency is set per marketplace.

Can I get rank history? Run on a schedule and keep the datasets; each row has asin, rank, listUrl and scrapedAt.

Does it need a browser, cookies or a login? No. Plain HTTPS requests only, which is why it's fast and cheap.

How it works (for developers)

An Amazon list page embeds all 50 ASINs of the page with their ranks in a data-client-recs-list attribute and renders the first 30 product cards server-side. The remaining 20 cards come from a nextPage call to the same component endpoint the page's own script uses, sent with the page's x-amz-acp-params token on the same session. The scraper parses the cards with Cheerio, merges both halves, sorts by rank and reads the left-hand navigation for parent and child categories. Prices, ratings and counts are parsed per locale. Every field is read defensively: a malformed card just has fewer fields.

Run it locally:

npm install
npm test                                  # parser tests on saved fixtures
APIFY_LOCAL_STORAGE_DIR=./storage node src/main.js   # input in storage/key_value_stores/default/INPUT.json