Amazon Product Scraper API - ASIN, Search, Bestsellers, Price

The Amazon product data suite — 7 Actors, one engine

Amazon product data by ASIN, product URL, search phrase, category or Best Sellers list — on 19 marketplaces, in one run and one schema.

A run that finds nothing costs nothing. There is no run-start fee, error and coverage rows are free, and on a schedule a product is billed only when it has actually changed.

⭐ Why this one

One Actor covers what the category splits across a dozen: paste an ASIN, a product URL, a search phrase, a category or browse-node URL, or a Best Sellers / New Releases / Movers & Shakers list, and it returns one normalised schema for all of them.

Every product row carries the fields rivals drop: the full 5-to-1 star breakdown, the Buy Box seller with its id and whether Amazon fulfils it, Best Sellers Rank in every category the product sits in, variant ASINs with their named options, list price, delivery date and the whole product-details table.

The price is always in the marketplace's own currency — no export prices in the wrong currency, which is the single most common complaint about the alternatives. Put it on a schedule and it bills only the products that changed. Failures are free, and there is no captcha solver to buy.

🎯 Use cases

• Repricing — track a catalogue of ASINs and react the moment a competitor's price or Buy Box changes.
• Competitor watch — pull a rival's search rankings, Best Sellers and new arrivals on a schedule.
• Buy Box tracking — see who holds the Buy Box on each ASIN, whether Amazon fulfils it, and when it flips.
• Catalogue enrichment — turn a list of ASINs into full product records with price, BSR, star breakdown, variants and images.

🚀 Recipes

Look up a list of ASINs

Paste ASINs, one per line. Each comes back as one full product row.

{ "inputs": ["B0CHX3QBCH", "B07XLTPC5L", "0735211299"], "marketplace": "com" }

Scrape a search across several pages

{ "inputs": ["wireless earbuds"], "mode": "search", "maxPagesPerSearch": 5, "maxItemsPerInput": 200 }

Best Sellers for a category, enriched to full products

{
  "inputs": ["https://www.amazon.com/gp/bestsellers/electronics/"],
  "fetchProductDetails": true,
  "maxItemsPerInput": 50
}

A bad ASIN never costs you

An ASIN that does not exist comes back as a free row saying so — no charge, and the rest of your list still runs.

{ "inputs": ["B0CHX3QBCH", "B0ZZZZZZZZ"] }

📥 Input

Field	What it does
inputs	ASINs, product URLs, search phrases, category URLs, or Best Sellers / New Releases / Movers & Shakers list URLs. Mix them freely.
marketplace	Which Amazon a bare ASIN or search is read on — 19 marketplaces. URLs carry their own.
language	English on every marketplace, or each marketplace in its native language.
maxItemsPerInput / maxItemsTotal	How many products or list rows to return, per input and for the whole run.
maxPagesPerSearch	How deep to page a search or category.
fetchProductDetails	Read each list row from its product page and bill it as a product — never as both.
includeVariants / fetchVariantDetails	List variant ASINs on the product, and optionally fetch each as its own row.
onlyChangedSinceLastRun / stateName / changeFields	Monitoring: bill only products whose watched fields changed; name the feed; choose the fields.
maxCostUsd	A hard spend ceiling for the run.

No proxy settings. Unblocking is handled and priced in. You never configure infrastructure.

🌍 Marketplaces

Bare ASINs and searches are read on the marketplace you pick; any URL you paste carries its own. Supported: amazon.com (US), .co.uk (UK), .de (Germany), .fr (France), .it (Italy), .es (Spain), .ca (Canada), .com.au (Australia), .in (India), .co.jp (Japan), .com.mx (Mexico), .nl (Netherlands), .se (Sweden), .pl (Poland), .com.br (Brazil), .sg (Singapore), .ae (UAE), .sa (Saudi Arabia) and .com.tr (Turkey). amazon.com, .co.uk and .de are verified against captured pages; the other 16 share the same page layouts and are supported best-effort — open an issue if one misbehaves. The same ASIN on .com, .co.uk and .de comes back in one language and one schema, so a multi-market catalogue is filterable without a cleanup pass — and the price is always in that marketplace's own currency: a page served in another currency is refused as a free error row, never billed.

📊 Measured coverage

Rivals claim coverage; this publishes it, and every number below is a floor the test suite asserts over real captured pages.

Field	Coverage on in-stock product pages	Notes
Title, ASIN, price	100%	Price always in the marketplace's own currency.
Star rating + full 5-to-1 breakdown	100%	The per-star percentages, not just the average.
Ratings count	100%	The exact number behind the average.
Buy Box seller	100%	With the seller id when the offer is third-party; the fulfiller where the page prints it.
Main image	100%	Plus every gallery image the page embeds, hi-res where Amazon has one.
Variant ASINs + named options	present on variation products	Colour, size, style — the child ASINs and their labels.
Best Sellers Rank per category	present where Amazon prints it	Every category the product ranks in, not just the top one.

An out-of-stock product has no price on the page, so price comes back null — never a stale or guessed value. In general a field Amazon does not print comes back null, because a fabricated price or rank silently corrupts whatever you filter on.

📤 Output

Row	What it is
product	One product: ASIN, marketplace, title, brand, price and currency, list price and discount, availability, Buy Box seller and fulfiller, star rating with the full 5-to-1 breakdown, ratings count, bought-in-past-month, bullet points, description, images, variant ASINs, breadcrumb categories, Best Sellers Rank per category, date first available, dimensions, weight and the product-details table.
list-item	One row of a search, category or ranked list: position, rank, ASIN, title, price, currency, rating, ratings count, sponsored flag and image.
unchanged	Free. Monitoring: a product that has not changed since the last run.
coverage	Free. Per input: pages read, items found and billed, whether the walk reached the end, and if not, exactly what stopped it.
error	Free. Anything that could not be read, with a stable reason. Nothing you typed is dropped in silence.

💰 Pricing

Pay per product. Failures are free and there is no run-start fee.

Event	FREE	BRONZE	SILVER	GOLD	PLATINUM	DIAMOND
product	$16.00	$13.30	$10.10	$8.00	$7.50	$6.90
list item	$6.40	$5.30	$4.00	$3.20	$3.00	$2.80

Prices are per 1,000. Your tier is your Apify plan.

On the Gold plan, 1,000 products cost $8.00. A product row is priced above the category on purpose: every product page is read from inside the marketplace's own country so the price is in its own currency — the unblocking that makes that possible is most of what a row costs to serve — and each row carries the BSR, the star breakdown and the Buy Box seller that cheaper scrapers leave empty. There is no captcha solver to buy on top. With onlyChangedSinceLastRun on, a scheduled re-run over unchanged products costs $0.00 — a product whose watched fields did not move comes back as a free unchanged row instead of being re-billed. With it off, every run bills every row it delivers.

Free vs paid Apify plan: this Actor runs on the free plan. The free plan's $5 monthly credit covers about 312 products or 781 list rows at the FREE-tier price; beyond that you need a paid plan, as with every Store Actor.

⏰ Run it on a schedule

Set a cron — 0 * * * * (hourly) — and leave onlyChangedSinceLastRun on. Because already-delivered products are never charged again, a quiet run costs close to nothing and a busy one bills only what moved. That is what makes a daily watch rational instead of a bill you cancel after a week.

❓ FAQ

How do I get Amazon product data by ASIN?

Put the ASIN — or several, one per line — in inputs, pick a marketplace, and run it. Each comes back as one full product row: price, availability, Buy Box, star breakdown, sales rank, variants and images.

How do I track an Amazon price over time?

Put the ASINs on a schedule with onlyChangedSinceLastRun on. A product is billed only when a watched field changes since the last run — by default price, availability, sales rank or Buy Box seller; unchanged products come back free. Sales rank moves on most products every few hours, so drop salesRank from changeFields if you only want price events. Add maxCostUsd to cap a busy day.

Does Amazon have a product API?

Amazon's own Product Advertising API is gated behind an affiliate account with sales requirements and returns a limited set of fields. This Actor needs no account and returns the full page — Buy Box, the star breakdown, BSR per category and variants — for any public ASIN, search or list.

How many pages of Amazon search results can I scrape?

As many as the search actually returns — there is no fixed page cap here. Set maxPagesPerSearch to the depth you want. For a whole category, point it at the category or Best Sellers URL instead, which is the more complete route.

Do I need a proxy or a captcha solver?

No. Unblocking is handled and priced into the per-row fee. There is no proxy field to configure and no captcha solver to buy.

Why is the price in the wrong currency on some other scrapers?

Because they read the page from wherever their network happens to exit, and Amazon then shows an export price. This Actor always reads each marketplace from its own country, so the price and currency match the marketplace you asked for.

How do I scrape Amazon Best Sellers, New Releases or Movers & Shakers?

Paste the list URL into inputs. You get all 100 ranks across the two pages — the ones the page prints in full, plus the rest as rank and ASIN, clearly flagged rather than padded with guesses. Turn on fetchProductDetails to read every ranked row as a complete product.

Can I get the Buy Box seller and whether Amazon fulfils it?

Yes. Every product row carries the Buy Box seller, its seller id when the offer is third-party, and whether the item ships from Amazon — the fields you need to tell an FBA offer from a merchant-fulfilled one.

How do I export Amazon product data to CSV, Excel or JSON?

Every run's dataset exports to CSV, Excel, JSON, XML or HTML from the Storage tab, or through the Apify API. The dataset ships with ready-made views for products, list rows, changes and the coverage report.

What happens when an ASIN does not exist or a page is blocked?

You get a free error row naming the reason, and the rest of your input still runs. Nothing you typed is dropped in silence, and a run that returns nothing costs nothing.

🔗 Looking for Amazon reviews?

Reviews live in a separate suite, so this one stays focused on product data:

• amazon reviews scraper api — every review of a product.
• amazon review monitoring scraper api — new reviews on a schedule.
• amazon ratings scraper api — the star rating and count only.

⚖️ Unofficial

This is an independent tool and is not affiliated with, endorsed by or sponsored by Amazon. It reads only publicly available pages. "Amazon" is used here only to describe what the tool reads.

🛟 Support

Something not working, or a marketplace you want checked? Open an issue on the Issues tab — it is read and answered. If the Actor does what you needed, a rating on the listing helps more than you would think.