Amazon Bestsellers Scraper

An Amazon bestsellers scraper that returns a ranked product list from any Amazon category, on any of 15 marketplaces, with the price in that marketplace's own currency. Give it a category and it returns Amazon's own rank badge alongside the ASIN, title, price, star rating and rating count.

No login, no cookie, no captcha solve. Amazon renders its ranking pages server-side for logged-out visitors, which is the surface this actor reads.

What you get

• Amazon's real rank, taken from the rank badge on the page, not renumbered by row order. When Amazon skips ranks, you see the skip instead of a tidy fiction
• Price in the correct currency, because the proxy exit is pinned to the marketplace's own country. This matters more than it sounds: an unpinned request to amazon.com came back priced in Egyptian pounds
• 15 marketplaces on one input: US, UK, Germany, France, Italy, Spain, Netherlands, Sweden, Poland, Canada, Australia, Mexico, Brazil, India and Japan
• Three rankings from the same actor: Best Sellers, New Releases, and Movers & Shakers
• Rating and rating count per product, so you can separate a bestseller with 280.000 ratings from one that climbed on 40
• A useful error when you point it at a hub page. Amazon's top-level category pages render in the browser and carry no products in the HTML. Instead of returning zero rows and letting you guess, the actor names the problem and tells you to use a leaf category
• No actor-start fee. A run that returns nothing costs nothing

Why scrape Amazon bestsellers

The bestseller rank is the closest thing Amazon publishes to a sales figure. Amazon will not tell you units sold, but it will tell you, hourly, which products outsell which in every category on every marketplace. For anyone deciding what to stock, what to launch, or what a competitor is winning with, that ranking is the primary signal and there is no API for it.

The second reason is drift. A ranking is only interesting as a series. Yesterday's number one that is now at 14 tells you something the snapshot never will, and the only way to build that series is to capture it on a schedule. This actor is built for that: small runs, cheap per row, no start fee eating the budget of a job you run every hour.

Currency is where most Amazon scrapers quietly break. Amazon prices off the requesting IP, so a scraper running through a rotating proxy without country pinning returns prices in whatever country the exit landed in that minute. The rows look fine. The numbers are unusable. This actor pins the exit to the marketplace before the request goes out.

Input

Field	Type	Required	Default	What it does
category	string	yes	electronics	Category slug, numeric node id, or a full Amazon ranking URL
marketplace	select	no	com	Which Amazon site, and therefore which currency and proxy exit country
list_type	select	no	bestsellers	bestsellers, new_releases or movers_and_shakers
limit	integer	no	30	Products to return, 1 to 60
retries	integer	no	4	Retry attempts per page, each with a rotated user agent and TLS fingerprint

Output

{
  "rank": 1,
  "asin": "B08JHCVHTY",
  "title": "Blink Plus Plan with monthly auto-renewal",
  "url": "https://www.amazon.com/Blink-Plus-Plan-monthly-auto-renewal/dp/B08JHCVHTY",
  "price": 11.99,
  "price_text": "$11.99",
  "currency": "$",
  "rating": 4.4,
  "rating_text": "4.4 out of 5 stars",
  "reviews_count": 279727,
  "image": "https://images-na.ssl-images-amazon.com/images/I/...",
  "marketplace": "com",
  "category": "electronics",
  "list_type": "bestsellers"
}

Use cases

Deciding what to stock. Pull the top of a category on the marketplaces you sell in, then join on ASIN against your own catalogue. What is ranked and absent from your range is the gap, and the rating count tells you whether it is a genuine seller or a recent spike. Run it across com, co.uk and de and you get the same question answered per market in one afternoon.

Tracking a launch. New Releases is the list that matters when a competitor ships something. Schedule it hourly on their category with list_type: new_releases, deduplicate on ASIN, and you know within the hour when something new enters the ranking, at what price and with how many early ratings.

Building a rank time series. The single most valuable thing here is the same run repeated. Store rank, asin and the timestamp, and after a fortnight you have a movement chart Amazon does not sell. Movers & Shakers gives you the same signal pre-computed for the last 24 hours when you want the answer without the history.

Pricing research across borders. The same ASIN often carries a very different price on amazon.de than on amazon.co.uk. Because the exit country is pinned per marketplace, running the same category across several marketplaces gives you comparable numbers rather than a currency soup.

How it compares

	this actor	junglee/amazon-bestsellers	crawlerbros/amazon-bestseller-scraper
Per 1.000 products	$3,00	$5,90	$5,00
Actor-start fee	none	none	$0,05 per run
Run failure rate	0% across validation runs	10,9% over 30 days	–
Marketplaces	15	fewer	fewer
New Releases and Movers & Shakers	yes, same input	separate actors	no
Currency pinned to marketplace	yes	not stated	not stated
Rating and rating count	yes	yes	yes

Honest about the other side: junglee/amazon-bestsellers has 331 monthly users and 11 reviews against our zero, and it has been running for years. The reason to switch is that it fails one run in nine and costs twice as much, not that it cannot do the job.

Pricing

One event. product costs $0,0030 per ranked product returned, which is $3,00 per 1.000. All pricing is pay-per-event, so you only pay for products you actually receive. No actor-start fee, no per-compute-unit charges, and a run that returns nothing costs nothing.

Limits and gotchas

• 60 products is the real ceiling per run, and the ranks have a gap. Amazon's pg parameter steps by 50 but only 30 of each page's ranks are in the HTML. You get ranks 1-30 and 51-80. Ranks 31-50 and 81-100 are loaded in the browser and are not available to any HTML scraper, so an actor claiming a clean top 100 is either renumbering rows or using a browser.
• Use a leaf category, not a hub. gp/bestsellers/ce on amazon.de is a hub page and returns no products. gp/bestsellers/ce/562066 returns 30. The actor detects this and tells you, rather than reporting an empty run.
• Category slugs are not portable between marketplaces. electronics works on com, co.uk and fr; on de the equivalent is ce. When in doubt, open the ranking in your browser and paste the URL into category.
• Books rankings carry fewer fields. Amazon does not render a star rating on book tiles in every marketplace, so rating and reviews_count can be empty there while price and rank are present.
• Prices reflect the pinned exit country, which is what you want, but it means a marketplace you cannot reach from that country will behave oddly. All 15 listed marketplaces were checked.
• Movers & Shakers is only meaningful hourly. It is a 24-hour delta list, so running it daily gives you a snapshot of a window you have already missed most of.

FAQ

Can I scrape Amazon bestsellers without an API key? Yes. Amazon publishes its ranking pages to logged-out visitors and search engines, and that is the surface this actor reads. There is no credential field in the input.

Why does my Amazon scraper return prices in the wrong currency? Because Amazon prices off the requesting IP and most rotating proxies pick a random exit country per request. This actor pins the exit to the marketplace's country before the request goes out, which is why amazon.de returns euros rather than whatever the proxy landed in.

Can I get the full Amazon top 100? Not from the HTML. Amazon renders 30 ranks per 50-rank page, so ranks 31-50 and 81-100 exist only after the browser loads them. This actor returns the 60 that are really there and tells you which ranks they are.

Which Amazon marketplaces are supported? Fifteen: com, co.uk, de, fr, it, es, nl, se, pl, ca, com.au, com.mx, com.br, in and co.jp. Each one is fetched through an exit in its own country.

How do I find the category id for a marketplace? Open the ranking you want in a browser and copy the URL. Both the slug form and the numeric node id work, and you can paste the whole URL into category if you would rather not pick it apart.

Related Actors

• Google Shopping Scraper — offers from every merchant for an EAN or SKU
• EAN to ASIN Converter — turn a barcode into an Amazon ASIN
• Google Shopping Reviews — product reviews aggregated across merchants