Amazon ASIN Scraper With Every Size, Color & Variant

Amazon ASIN Scraper — Variants, Spec Tables and Media Inventory

Amazon ASIN Scraper turns a list of ASINs into complete catalogue records: the parent product, every child variant with its size, colour, style and pattern values, the full specification and product-details tables, every gallery image with the resolution Amazon itself states for it, product video URLs, and the rendered sold-by, ships-from, delivery and stock sentences. Every row is typed, normalized JSON with a fixed key set — no HTML, no selectors, no parsing. Paste ASINs, press Start, and export the whole variation family as JSON, CSV or Excel.

⚠️ Variant rows carry no per-child price or stock

This is the one limitation worth reading before your first run, because it is the thing an "every size and colour" scraper is most often assumed to do.

The Actor fetches one page per input ASIN — the parent detail page — and reads Amazon's twister payload off it. That payload states each child ASIN and its dimension values. It does not state each child's price, stock level or delivery date. So every entry in variationMatrix, and every row emitted by variantsAsSeparateRows, carries only what the parent page actually said about that child: asin, its dimension values, a joined label, isSelected and a product URL. Nothing is copied down from the parent, because a copied price would be wrong for most children.

Two further things to know:

• Child URLs are always built on amazon.com. variationMatrix[].url and the variant row's url are composed as https://www.amazon.com/dp/<ASIN> regardless of the marketplaceDomain you set. On a non-US storefront, rewrite the host yourself.
• Books are the exception. bookVariations comes from the format-swatch widget, where Amazon prints the price on each swatch — so those entries do carry a priceText.

If you need a real price and stock figure per child, take the ASINs out of variationMatrix and feed them back in as a second run. Each child then becomes its own product row, with its own price, inStock, warehouseAvailability and deliveryDateText, and still returns the whole family it belongs to.

What is Amazon ASIN Scraper With Every Size, Color & Variant?

Amazon ASIN Scraper With Every Size, Color & Variant is an Apify Actor that takes ASINs and returns a full catalogue record for each one. It is record-driven, not search-driven: you supply the identifiers, and every input ASIN produces exactly one product row plus, optionally, one row per child variant.

No Amazon account, login or cookie is required. Every field it returns is visible to an anonymous visitor on the public detail page. The only credential you need is your Apify token.

• Scrape the product record for a parent or child ASIN — title, brand, price, ratings, description, features, breadcrumbs and bestseller ranks
• Scrape the variation matrix — parent ASIN, every dimension with its full value list, every child ASIN with its size, colour, style and pattern values, and which one Amazon is showing
• Scrape the specification and product-details tables as complete name/value lists, in both the table and bullet-list layouts Amazon uses
• Scrape the media inventory — every gallery image with its hi-res, large and thumbnail URLs and Amazon's own stated width x height, plus product video URLs, titles and durations
• Scrape the rendered fulfilment strings — sold by, ships from, delivery date, the "Only N left in stock" sentence and the "N bought in past month" badge, verbatim
• Export as JSON, CSV or Excel. No proxy management, no parsing

Nothing in the output is calculated. There is no fee schedule, no margin, no ROI and no sales estimate in this build — every value is a string or number Amazon rendered on the page or embedded in its own JSON, or a direct reformat of one. Missing data comes back as null or an empty list, never as an invented value.

What data does Amazon ASIN Scraper collect?

Four entity types come out of a single page fetch: the product itself, its child variants, its specification tables, and its media inventory — plus the fulfilment strings Amazon printed in the buy box.

Data Type	Key Fields	JSON Field Names
Product (parent ASIN)	Title, brand, price, list price, rating, review count, stock, breadcrumbs, bestseller ranks	asin, title, brand, price, listPrice, stars, reviewsCount, inStock, breadCrumbs, bestsellerRanks
Child variant (size / colour / style / pattern)	Child ASIN, dimension values, joined label, selected flag, parent identity	variationMatrix, variationDimensions, variationDimensionNames, variationCount, selectedVariation, parentAsin, hasVariations, variationSource
Specification table	Technical spec rows and the Product details block, as name/value pairs	productSpecification, productDetails, attributes, productOverview
Media inventory	Every gallery image with hi-res / large / thumbnail URLs and Amazon's stated resolution, plus product videos	imageInventory, imageCount, imageResolutions, videoInventory, videoCount
Fulfilment strings	Sold by, ships from, delivery text, stock sentence, bought-in-past-month badge	soldBy, fulfilledBy, deliveryMessage, deliveryDateText, fastestDeliveryText, warehouseAvailability, pastSales
Book / media editions	Format name, edition ASIN, printed price text, selected flag	bookVariations
Rich content	A+ modules, brand story cards, comparison table	aPlusContent, brandStory, productComparison

Need more Amazon data?

If you are working from category and search pages rather than known ASINs, Amazon Product Scraper takes category or product URLs instead of identifiers. For demand-side work, Amazon Today's Deals Scraper covers the deals surface and Amazon Trending Products Scraper covers movers by category. This Actor is the one to reach for when you already have the ASIN and need everything the detail page states about it.

How does Amazon ASIN Scraper differ from the official Amazon Product Advertising API?

Amazon's own programmatic surface for product data is the Product Advertising API (PA-API 5.0). It returns the item resources Amazon documents, to callers Amazon has approved. Amazon ASIN Scraper reads the public detail page as an anonymous visitor and returns what that page states.

Feature	Product Advertising API 5.0	Amazon ASIN Scraper
Access requirement	Amazon Associates account in good standing	Apify account only
Ongoing eligibility	Tied to qualifying sales generated through your Associates links	❌ No sales requirement
Data scope	The item resources Amazon documents for the API	What the storefront page renders to an anonymous visitor
Rendered delivery, stock and sold-by sentences	Not part of the documented item resources	✅ deliveryDateText, warehouseAvailability, soldBy, pastSales, verbatim
On-page specification tables	Not returned as page tables	✅ productSpecification and productDetails as name/value lists
Stated image resolutions per gallery image	Not exposed	✅ imageInventory[].width, .height, .statedRenditions
Which variant is currently shown	Page state, not an item attribute	✅ isSelected and selectedVariation
Setup	Associates application and approval	Paste ASINs and press Start

Check the current PA-API 5.0 documentation for its exact resources, eligibility terms and request limits before choosing between them — those terms change and are not reproduced here.

Use the official API when you are an approved Associate, want Amazon's supported contract, and need documented item attributes. Use Amazon ASIN Scraper when you have no Associates account, or when you specifically need what the page says today: the live variation family with the selected child marked, the rendered delivery and stock text, and the specification tables as printed.

Why do developers and teams scrape Amazon ASINs?

For AI engineers and agent builders

A product-research agent needs the variation family before it can answer anything useful about sizing or colourways. Index title, variationDimensions, productSpecification and productDetails into a vector store keyed by asin, and an agent can answer "which colours does this come in, and what is the stated fabric" from live catalogue data rather than a stale export. Because every response is typed JSON with a fixed key set, there is no HTML-parsing step inside the agent loop — the row goes straight into the context window or into a tool result.

For catalogue and merchandising teams

Rebuilding a supplier's option grid means reproducing all of it, not a summary. Turn on variantsAsSeparateRows and one run gives you a spreadsheet-shaped grid: one line per child ASIN with size, color, style and pattern in their own columns, variantOfAsin for the join back to the parent, and isSelectedVariant marking the default. Pair it with imageInventory and you get the media assets and their stated resolutions in the same pass, so you know which shots are large enough to reuse before anyone downloads them.

For pricing and market analysts

The fulfilment block is where competitive signal hides. soldBy and fulfilledBy tell you whether Amazon or a third-party merchant holds the buy box, warehouseAvailability carries the "Only N left in stock" sentence, and pastSales carries the "N bought in past month" badge — all as rendered, with nothing derived from them. Set deliveryPostcode and re-run the same ASINs for a second location, and deliveryDateText plus warehouseAvailability change with it, which is how you compare regional availability on identical inputs.

For developers building data products

Every run writes to an Apify dataset with a stable key set, so a scheduled job can append to a warehouse table without a migration each week. Fields never disappear when you switch a toggle off — includeSpecificationTables: false yields productSpecification: [], not a missing key — so downstream code never needs a presence check. Schedule the run in Apify, attach a webhook to the run-succeeded event, and pipe the dataset into BigQuery, S3 or an internal API without writing any scraping code.

How to scrape Amazon ASINs (step by step)

Amazon ASIN Scraper runs on Apify. Start it from the Apify Console or call it through the Apify API.

1. Open Amazon ASIN Scraper on Apify and click Try for free
2. Paste your ASINs into ASINs to map (asins) — one per line, 10 alphanumeric characters each, taken from the product URL after /dp/. This is the only required input, and pointing it at a child ASIN still returns the whole family
3. Set the storefront: Storefront domain (marketplaceDomain, default amazon.com) and Proxy exit country (exitCountry). Match them — a US residential exit against a non-US Amazon domain is served a CAPTCHA shell
4. Choose your output shape: leave Map every size, colour and variant on for a nested matrix, or turn on Emit each variant as its own row for a flat grid, capped by Variant rows per product cap
5. Click Start, then download the dataset as JSON, CSV or Excel — or read it through the Apify API

The dataset ships with two views: the default Variations, specs & media per ASIN table, and a One row per size / colour variant view that shows the flat grid columns.

What to do when Amazon changes its structure

The Actor is maintained, and the variation reader already carries four independent strategies — Amazon's twister payload, asinVariationValues, a raw regex over the display-data blob, and the rendered twister widget — tried in order until one produces dimensions. The output key set stays the same whichever wins, so your integrations keep working while the extraction path underneath changes.

⬇️ Input

Every parameter below is read directly from the Actor's input schema. asins is the only required field; everything else has a default.

Parameter	Required	Type	Description	Example Value
asins	Yes	array	One ASIN per line — 10 alphanumeric characters, e.g. B07GBZ4Q68. Take it from the product URL after /dp/. Point it at any child ASIN and you still get the whole family back. ASINs are fetched concurrently.	["B0BSHF7WHW", "B08N5WRWNW"]
marketplaceDomain	No	string	Which Amazon storefront to read the listing from, e.g. amazon.com, amazon.co.uk, amazon.de. Variation labels come back in that storefront's language. Free text, not a dropdown. Default "amazon.com".	"amazon.co.uk"
pageLanguage	No	string	Language code sent as ?language=… and in Accept-Language. Dimension labels (Size, Color) and spec-table keys are rendered in this language, so English gives the most predictable column names. Default "en". One of "", en, en_US, en_GB, de, fr, es, it, nl, pl, pt, pt_BR, sv, tr, cs, da, ja, zh_CN, zh_TW, hi.	"en"
deliveryPostcode	No	string	Optional destination postcode (e.g. 10001). Sets Amazon's delivery cookie before scraping, so the delivery date string and stock sentence you get back are the ones for that location. Default "".	"10001"
includeVariationMatrix	No	boolean	Read Amazon's twister payload and return parentAsin, variationDimensions (each dimension with its full value list and the selected value), and variationMatrix — one entry per child ASIN with its dimension values and isSelected. Default true.	true
variantsAsSeparateRows	No	boolean	In addition to the product row, push one row per child ASIN (rowType: "variant") with size, color, style, pattern, variantOfAsin and the child's product URL. Default false.	true
maxVariantRowsPerProduct	No	integer	Upper bound on how many child rows a single product may emit when the option above is on. Apparel parents can carry hundreds of children. Minimum 1, maximum 200. Default 30.	50
onlyProductsWithVariations	No	boolean	Drop any ASIN whose page shows no twister at all. Use it when you are hunting variation parents and single-variant listings are noise. Default false.	false
includeSpecificationTables	No	boolean	Return productSpecification (technical spec tables) and productDetails (the Product details block, in both its table and bullet-list layouts) as complete name/value lists. Default true.	true
includeMediaInventory	No	boolean	Return imageInventory — every gallery image with its hi-res, large and thumbnail URLs plus the width x height Amazon itself states — and videoInventory with product video URLs, titles and durations as given. Default true.	true
includeFulfilmentFacts	No	boolean	Return soldBy, fulfilledBy, deliveryMessage, deliveryDateText, fastestDeliveryText, warehouseAvailability (the "Only N left in stock" sentence) and pastSales (the "N bought in past month" badge) exactly as rendered. Default true.	true
trimRichContent	No	boolean	A+ modules, brand story and comparison tables dominate the size of an Amazon record. Turn this on to blank them and keep the dataset small. The variation matrix, spec tables and imageInventory are not affected. Default false.	false
exitCountry	No	string	Route requests through this country. Match it to the storefront domain — a US residential exit against a non-US Amazon domain is served a CAPTCHA shell. Default "AUTO_SELECT_PROXY_COUNTRY". One of AUTO_SELECT_PROXY_COUNTRY, US, GB, DE, FR, ES, IT, CA, AU, JP, IN, BR, MX, NL, PL, TR, SE, AE.	"GB"
proxyConfiguration	No	object	Full proxy override — Apify groups or your own URLs. Overrides the exit country above. Most runs can leave this alone. Prefilled with Apify Residential.	{"useApifyProxy": true, "apifyProxyGroups": ["RESIDENTIAL"]}

Accepted alias keys. For backward compatibility with the base Amazon input shape, the Actor also reads amazonDomain, language, zipCode and proxyCountry, and each takes priority over its schema equivalent above. asinList and productAsins are accepted as alternatives to asins. Fee and profit inputs from an earlier build (unitCost, inboundShippingCost, extraCostPerUnit, referralFeePercentOverride, fulfilmentFeeOverride, feeCategoryOverride, fulfilmentChannel, includeFeeBreakdown, onlyProfitableAsins) and useCaptchaSolver are accepted and ignored, each with a log line explaining why — this build calculates nothing and ships no inline CAPTCHA solver.

pageLanguage: "" ("Marketplace default") behaves exactly like omitting the key: no ?language= is appended to the product URL, reviewsLink and unNormalizedProductUrl lose their language suffix, and the Accept-Language header falls back to English.

Example input

{
  "asins": ["B0BSHF7WHW", "B08N5WRWNW", "1250301696", "B00FLYWNYQ"],
  "marketplaceDomain": "amazon.com",
  "pageLanguage": "en",
  "deliveryPostcode": "10001",
  "includeVariationMatrix": true,
  "variantsAsSeparateRows": true,
  "maxVariantRowsPerProduct": 50,
  "onlyProductsWithVariations": false,
  "includeSpecificationTables": true,
  "includeMediaInventory": true,
  "includeFulfilmentFacts": true,
  "trimRichContent": false,
  "exitCountry": "US",
  "proxyConfiguration": {
    "useApifyProxy": true,
    "apifyProxyGroups": ["RESIDENTIAL"]
  }
}

Common pitfall: turning onlyProductsWithVariations on while includeVariationMatrix is off. With the matrix off, hasVariations is null on every record, so the filter drops every row and the dataset comes back empty. The reverse combination is handled for you: if variantsAsSeparateRows is on and includeVariationMatrix is off, the Actor logs a warning and turns the matrix back on, because no variant row could otherwise exist.

The second most common mistake is a storefront typo. marketplaceDomain is free text and nothing validates it — a mistyped domain is fetched as typed, fails, and lands as an uncharged error row rather than an obvious input error.

⬆️ Output

Typed, normalized JSON with a fixed key set. Every product row carries the same 81 keys on every run, whatever your toggles are set to — switching a feature off empties its fields rather than removing them, so downstream code never needs a presence check. Export as JSON, CSV or Excel, or read the dataset through the Apify API.

Alongside the dataset, the run writes an OUTPUT record to the default key-value store: an object keyed by ASIN holding every parsed record from the run, including ones the filter dropped and ones that errored.

Toggle behaviour, exactly as implemented:

Toggle	What happens to the fields
includeVariationMatrix: false	hasVariations: null, variationSource: null, parentAsin: null, variationDimensions: [], variationDimensionNames: [], variationMatrix: [], variationCount: 0, selectedVariation: null, bookVariations: []
includeSpecificationTables: false	productSpecification: [], productDetails: []
includeMediaInventory: false	imageInventory: [], imageCount: 0, imageResolutions: [], videoInventory: [], videoCount: 0
includeFulfilmentFacts: false	soldBy, fulfilledBy, deliveryMessage, deliveryDateText, fastestDeliveryText, warehouseAvailability, pastSales all null
trimRichContent: true	aPlusContent: null, brandStory: null, productComparison: null, galleryThumbnails: [], highResolutionImages: []

Field-name quirks, preserved exactly. breadCrumbs carries a capital C. unNormalizedProductUrl is spelled with that internal capital N. videosCount (the legacy image-block video counter) and videoCount (the length of videoInventory) are two different fields. attributes, productOverview and variantAttributes use {"key", "value"} pairs while productSpecification and productDetails use {"name", "value"}. fulfilledBy uses British spelling while color on variant rows uses American. None of this is tidied up on the way out — it is reproduced as it exists, so your field mapping matches what actually arrives.

Product row fields

Every key on a rowType: "product" row.

Field	Type	Description
rowType	string	Always "product" on this row type
asin	string	The ASIN scraped, uppercased
originalAsin	string	The same ASIN as submitted
title	string | null	Product title
brand	string | null	Brand as printed in the byline or the overview table
author	string | null	Author, on book and media pages
url	string	https://www.<domain>/gp/product/<asin>?smid=
unNormalizedProductUrl	string	https://www.<domain>/dp/<asin>, plus ?language= when a language is set
input	string	https://www.<domain>/dp/<asin> — the join key back to your input
reviewsLink	string	https://www.<domain>/product-reviews/<asin>
visitStoreLink	string | null	Absolute brand-store URL
position	integer	1-based counter in the order records were built during the run
isPinnedOffer	boolean	true when position is 1
variantOfAsin	null	Always null on a product row; populated on variant rows
price	object | null	{"value": float, "currency": symbol}
listPrice	object | null	Struck-through list price, same shape
priceRange	string | null	Displayed price range, as printed
shippingPrice	object | null	Same shape; value is 0.0 when the page says free
inStock	boolean | null	Read from the availability sentence
inStockText	string | null	The availability sentence itself
condition	string | null	New / used condition label
stars	float | null	Average rating
starsBreakdown	object	Rating histogram as fractions, e.g. {"5star": 0.81, "4star": 0.1}
reviewsCount	integer | null	Review count
hasReviews	boolean	true when reviewsCount is above zero
answeredQuestions	integer | null	Answered-questions count
isAmazonChoice	boolean	Amazon's Choice badge present
amazonChoiceText	string | null	The badge text
monthlyPurchaseVolume	string | null	The bought-in-past-month faceout text
aiReviewsSummary	object | null	{"text", "keywords": [{"name", "sentiment", "text", "customersMentionedCount": {"total"}, "partialReviews": []}]}
description	string | null	Product description
bookDescription	string | null	Book blurb, read only when description is empty
features	array of strings	The bullet feature list
attributes	array	[{"key", "value"}] from the spec tables and detail bullets
productOverview	array	[{"key", "value"}] from the product-overview table
importantInformation	string | null	The Important information block
sustainabilityFeatures	object | null	{"summary", "certifications": []}
returnPolicy	string | null	Return-policy text
support	string | null	Product-support text
breadCrumbs	string | null	Category path joined with >
categoriesExtended	array	[{"name", "url", "node"}] — one entry per breadcrumb, URL built on your storefront domain
bestsellerRanks	array	[{"rank", "category", "url"}]
aPlusContent	object | null	{"title", "rawText", "rawImages": [{"name", "url"}], "rawVideos": [{"url", "previewImageUrl"}], "modules": [{"type", "title", "text", "video", "image", "items"}]}
brandStory	object | null	{"title", "image", "items": [{"title", "text", "url", "image", "items"}]}
productComparison	object | null	{"products": [{"asin", "title", "thumbnail", "price", "stars", "reviewsCount"}]}
galleryThumbnails	array of strings	Legacy thumbnail URL list
highResolutionImages	array of strings	Legacy hi-res URL list
thumbnailImage	string | null	Main image URL
videosCount	integer	Video count from the legacy image-block parse
hasVariations	boolean | null	true when a twister was found, false for a genuine single-variant product, null when the matrix is switched off
variationSource	string | null	Which strategy won: twisterData, asinVariationValues, dimensionValuesDisplayData_regex or twisterDom
parentAsin	string | null	Parent ASIN, when the payload states one
variationDimensionNames	array of strings	Raw dimension keys, e.g. ["size_name", "color_name"]
variationDimensions	array	[{"name", "label", "values": [], "valueCount", "selectedValue"}]
variationMatrix	array	[{"asin", "dimensions": {}, "label", "isSelected", "url"}] — one entry per child ASIN
variationCount	integer	Length of variationMatrix — the full family size, never truncated
selectedVariation	object | null	The matrix entry Amazon is currently showing
bookVariations	array	[{"format", "asin", "priceText", "url", "isSelected"}] — book and media format swatches
variantAsins	array of strings	Child ASINs harvested from the raw page payloads
variantDetails	array	[{"name", "thumbnail", "images": [], "asin", "price"}]
variantAttributes	array	[{"key", "value"}] — the selected twister options as labelled on the page
productSpecification	array	[{"name", "value"}] from every technical specification table
productDetails	array	[{"name", "value"}] from the Product details block, table and bullet layouts
imageInventory	array	[{"hiResUrl", "largeUrl", "thumbnailUrl", "variant", "width", "height", "resolution", "statedRenditions": [{"url", "width", "height"}]}]
imageCount	integer	Length of imageInventory
imageResolutions	array of strings	Every non-null resolution, e.g. "1500x1500"
videoInventory	array	[{"url", "title", "thumbnailUrl", "durationSeconds", "durationText", "group"}]
videoCount	integer	Length of videoInventory
soldBy	string | null	Sold-by party, trimmed to the name Amazon printed
fulfilledBy	string | null	Ships-from / fulfilled-by party
deliveryMessage	string | null	The whole delivery block text
deliveryDateText	string | null	Just the bold delivery date
fastestDeliveryText	string | null	The fastest-delivery date
warehouseAvailability	string | null	The stock sentence, e.g. the "Only N left in stock" line
pastSales	string | null	The bought-in-past-month badge, verbatim
seller	object | null	{"name", "id", "url", "reviewsCount", "averageRating"}; null when nothing was found
shipsFrom	string | null	Ships-from string from the buy box
delivery	string | null	Legacy delivery string
fastestDelivery	string | null	Legacy fastest-delivery string
locationText	string | null	The location line in the page header
loadedCountryCode	string | null	Delivery country the page rendered for

price.currency is the symbol the page used ("$", "£", "€", "zł", "kr", "AED"), not an ISO code — the longest symbol is matched first, so S$193.73 is not mislabelled as US dollars.

Scraped product record (nested variation matrix)

The default shape: one row per input ASIN, with the whole family nested inside it. Arrays are trimmed here for readability; a real record carries them in full.

{
  "rowType": "product",
  "asin": "B0BSHF7WHW",
  "originalAsin": "B0BSHF7WHW",
  "title": "Hanes Men's Essentials Short Sleeve T-Shirt, Cotton Crewneck Tee",
  "brand": "Hanes",
  "url": "https://www.amazon.com/gp/product/B0BSHF7WHW?smid=",
  "unNormalizedProductUrl": "https://www.amazon.com/dp/B0BSHF7WHW?language=en",
  "input": "https://www.amazon.com/dp/B0BSHF7WHW",
  "price": { "value": 12.0, "currency": "$" },
  "listPrice": { "value": 16.0, "currency": "$" },
  "inStock": true,
  "inStockText": "In Stock",
  "stars": 4.5,
  "starsBreakdown": { "5star": 0.71, "4star": 0.17, "3star": 0.07, "2star": 0.02, "1star": 0.03 },
  "reviewsCount": 48211,
  "hasReviews": true,
  "breadCrumbs": "Clothing, Shoes & Jewelry > Men > Clothing > Shirts > T-Shirts",
  "hasVariations": true,
  "variationSource": "twisterData",
  "parentAsin": "B0BSHDJ8YV",
  "variationCount": 84,
  "variationDimensionNames": ["size_name", "color_name"],
  "variationDimensions": [
    {
      "name": "size_name",
      "label": "Size",
      "values": ["Small", "Medium", "Large", "X-Large", "XX-Large"],
      "valueCount": 5,
      "selectedValue": "Large"
    },
    {
      "name": "color_name",
      "label": "Color",
      "values": ["Black", "White", "Navy", "Heather Grey"],
      "valueCount": 4,
      "selectedValue": "Navy"
    }
  ],
  "variationMatrix": [
    {
      "asin": "B0BSHF7WHW",
      "dimensions": { "size_name": "Large", "color_name": "Navy" },
      "label": "Large / Navy",
      "isSelected": true,
      "url": "https://www.amazon.com/dp/B0BSHF7WHW"
    },
    {
      "asin": "B0BSHG2K4M",
      "dimensions": { "size_name": "Medium", "color_name": "Navy" },
      "label": "Medium / Navy",
      "isSelected": false,
      "url": "https://www.amazon.com/dp/B0BSHG2K4M"
    }
  ],
  "selectedVariation": {
    "asin": "B0BSHF7WHW",
    "dimensions": { "size_name": "Large", "color_name": "Navy" },
    "label": "Large / Navy",
    "isSelected": true,
    "url": "https://www.amazon.com/dp/B0BSHF7WHW"
  },
  "bookVariations": [],
  "productSpecification": [
    { "name": "Fabric type", "value": "60% Cotton, 40% Polyester" },
    { "name": "Care instructions", "value": "Machine Wash" },
    { "name": "Country of Origin", "value": "Imported" }
  ],
  "productDetails": [
    { "name": "Item model number", "value": "5280" },
    { "name": "Date First Available", "value": "January 24, 2023" },
    { "name": "ASIN", "value": "B0BSHF7WHW" }
  ],
  "imageCount": 7,
  "imageInventory": [
    {
      "hiResUrl": "https://m.media-amazon.com/images/I/71aBcDeFgHL._AC_SL1500_.jpg",
      "largeUrl": "https://m.media-amazon.com/images/I/71aBcDeFgHL._AC_SX679_.jpg",
      "thumbnailUrl": "https://m.media-amazon.com/images/I/71aBcDeFgHL._AC_US40_.jpg",
      "variant": "MAIN",
      "width": 1500,
      "height": 1500,
      "resolution": "1500x1500",
      "statedRenditions": [
        { "url": "https://m.media-amazon.com/images/I/71aBcDeFgHL._AC_SX679_.jpg", "width": 679, "height": 679 },
        { "url": "https://m.media-amazon.com/images/I/71aBcDeFgHL._AC_SL1500_.jpg", "width": 1500, "height": 1500 }
      ]
    }
  ],
  "imageResolutions": ["1500x1500", "1200x1200"],
  "videoCount": 1,
  "videoInventory": [
    {
      "url": "https://m.media-amazon.com/images/S/vse-vms-transcoding-artifact-us-east-1/abc123/default.jobtemplate.mp4",
      "title": "How the Essentials tee fits",
      "thumbnailUrl": "https://m.media-amazon.com/images/I/51xYzAbCdEL.jpg",
      "durationSeconds": 31,
      "durationText": "0:31",
      "group": "brand"
    }
  ],
  "soldBy": "Amazon.com",
  "fulfilledBy": "Amazon.com",
  "deliveryMessage": "FREE delivery Thursday, July 30 on orders shipped by Amazon over $35",
  "deliveryDateText": "Thursday, July 30",
  "fastestDeliveryText": "Tomorrow, July 26",
  "warehouseAvailability": "Only 8 left in stock - order soon.",
  "pastSales": "10K+ bought in past month",
  "seller": {
    "name": "Amazon.com",
    "id": null,
    "url": "https://www.amazon.com",
    "reviewsCount": null,
    "averageRating": null
  },
  "categoriesExtended": [
    { "name": "Men", "url": "https://www.amazon.com/mens-clothing/b?node=7147441011", "node": "7147441011" }
  ],
  "bestsellerRanks": [
    { "rank": 3, "category": "Men's T-Shirts", "url": "https://www.amazon.com/gp/bestsellers/apparel/1045662" }
  ],
  "position": 1,
  "isPinnedOffer": true,
  "variantOfAsin": null,
  "loadedCountryCode": "US",
  "locationText": "Deliver to New York 10001"
}

Scraped variant row (exploded grid)

Set variantsAsSeparateRows: true and the product row above is followed by one row per child ASIN — a structurally different record with its own 17 keys, not a copy of the product row. This is the shape you paste straight into a sheet.

{
  "rowType": "variant",
  "asin": "B0BSHG2K4M",
  "variantOfAsin": "B0BSHF7WHW",
  "parentAsin": "B0BSHDJ8YV",
  "title": "Hanes Men's Essentials Short Sleeve T-Shirt, Cotton Crewneck Tee",
  "brand": "Hanes",
  "url": "https://www.amazon.com/dp/B0BSHG2K4M",
  "variantLabel": "Medium / Navy",
  "variantDimensions": { "size_name": "Medium", "color_name": "Navy" },
  "size": "Medium",
  "color": "Navy",
  "style": null,
  "pattern": null,
  "isSelectedVariant": false,
  "variationSource": "twisterData",
  "breadCrumbs": "Clothing, Shoes & Jewelry > Men > Clothing > Shirts > T-Shirts",
  "input": "https://www.amazon.com/dp/B0BSHF7WHW"
}

size, color, style and pattern are looked up by the literal dimension keys size_name, color_name, style_name and pattern_name. When a storefront uses different dimension keys, those four columns come back null and the values are still present in full inside variantDimensions — nothing is lost, it just moves.

Uncharged error rows

When an ASIN cannot be fetched or cannot be parsed, the Actor writes a minimal row so the failure is visible in your dataset rather than being an absence you have to notice. These rows are pushed without a charged event — you are not billed for them.

{ "asin": "B00FLYWNYQ", "error": "fetch_failed" }

{ "asin": "B08N5WRWNW", "error": "parse_error: <message>" }

Filter them out on the way in. In Python:

rows = [r for r in items if not r.get("error")]
products = [r for r in rows if r.get("rowType") == "product"]
variants = [r for r in rows if r.get("rowType") == "variant"]

Error rows carry no rowType, so a rowType check alone also separates real results from failures.

How does Amazon ASIN Scraper compare to other variant scrapers?

Checked on the Apify Store on 25 July 2026. Of the three closest listings, only one targets Amazon: junglee/amazon-asins-scraper. clearpath/shein-product-scraper covers Shein by keyword search, and s-r/free-google-shopping-variants covers Google Shopping variants by SKU. Claims below are as stated on those listings on that date; where a listing does not document something, it is marked not documented rather than assumed absent.

Feature	Amazon ASIN Scraper	Generic alternative
Full variation matrix per ASIN, with dimension value lists and a selected flag	✅ variationMatrix, variationDimensions, selectedVariation	junglee/amazon-asins-scraper: not documented in its README. s-r/free-google-shopping-variants documents variants for Google Shopping SKUs, not Amazon ASINs
Choice of nested matrix or one row per variant	✅ variantsAsSeparateRows, with a per-product cap	Not documented on any of the three
On-page specification and product-details tables as name/value lists	✅ productSpecification, productDetails	junglee/amazon-asins-scraper: not documented. clearpath/shein-product-scraper documents size charts and attributes for Shein
Image resolutions as stated by the source site	✅ imageInventory[].width, .height, .statedRenditions	Not documented on any of the three
Visible, structured error rows	✅ error field, pushed uncharged	junglee/amazon-asins-scraper documents error items with an error field and an error-code reference table
Documented input surface	14 schema parameters plus four accepted alias keys	junglee/amazon-asins-scraper documents ASINs plus store choice. clearpath/shein-product-scraper documents 8 fields. s-r/free-google-shopping-variants documents 6

If you are building an AI agent or a RAG pipeline, the output-format row is the decision-maker: parsing HTML inside an agent loop is a reliability failure mode, not a feature. And if what you actually need is a per-child price, read the constraint at the top of this page before choosing any variant scraper — it applies to anything that reads a single parent page.

How many ASINs and variant rows can you scrape?

There is no hard cap in the Actor itself. asins has no maximum length in the schema, and there is no maxItems parameter. The row count is the number of valid, unique ASINs you supply, plus up to maxVariantRowsPerProduct extra rows per product when variantsAsSeparateRows is on.

The one ceiling is the variant-row cap, and it is worth being precise about what it limits. maxVariantRowsPerProduct (default 30, minimum 1, maximum 200) caps emitted variant rows only. It is the Actor's limit, not Amazon's, and it does not touch the matrix read: variationMatrix and variationCount still carry the complete family. So a parent with 300 children returns all 300 inside variationMatrix and, at the default, 30 variant rows — the first 30 in the order Amazon's payload listed them. The remainder is dropped from the row stream and written to the run log, but nothing in the row itself flags the truncation, so compare variationCount against your row count if you need to know whether you got the full grid. Values above 200 are clamped to 200 even if the schema check is bypassed.

Input ASINs are uppercased, deduplicated and validated against a 10-character alphanumeric pattern. Anything that fails is skipped with a log warning before fetching starts and never reaches the dataset; if nothing valid remains, the run fails immediately with a message naming the asins input. Fetching runs at a fixed concurrency of 5 ASINs at a time with a 30-second request timeout — there is no concurrency input to raise.

Integrate Amazon ASIN Scraper and automate your workflow

Amazon ASIN Scraper works with any language or tool that can send an HTTP request.

REST API integration

from apify_client import ApifyClient

client = ApifyClient("<YOUR_APIFY_TOKEN>")

run = client.actor("<YOUR_USERNAME>/amazon-asin-scraper-sizes-colors-variants").call(run_input={
    "asins": ["B0BSHF7WHW", "B08N5WRWNW"],
    "marketplaceDomain": "amazon.com",
    "variantsAsSeparateRows": True,
    "maxVariantRowsPerProduct": 50,
    "exitCountry": "US",
})

for row in client.dataset(run["defaultDatasetId"]).iterate_items():
    if row.get("error"):
        continue
    if row["rowType"] == "product":
        print(row["asin"], row["title"], row["variationCount"], row["warehouseAvailability"])
    else:
        print("  ", row["asin"], row["size"], row["color"])

Works in Python, Node.js, Go, Ruby, cURL. Branch on rowType to route product rows and variant rows to different handlers, and skip anything carrying error.

Automation platforms (n8n, Make, LangChain)

In n8n, use the Apify node — or an HTTP Request node pointed at the Apify run endpoint with your token — and pass the same JSON input shown above; a Switch node on rowType then splits products from variants into separate branches. In Make, use the Apify module with run-and-wait and map the dataset output straight into a Google Sheets or Airtable step, which is the fastest way to get the exploded size/colour grid into a spreadsheet. In LangChain, wrap the run call as a tool that returns the dataset items; because every row is already typed JSON, the tool result needs no parsing before it reaches the model. Apify schedules and webhooks handle recurrence, so a weekly catalogue sweep needs no code at all.

Is it legal to scrape Amazon product data?

Scraping publicly listed product data is broadly treated as permissible where the data is accessible without authentication, and Amazon ASIN Scraper collects only what an anonymous visitor already sees on the public detail page. No account, no cookie, no session, no login bypass.

What comes back is product-catalogue data: titles, prices, specification tables, image URLs, variation grids and the merchant names Amazon prints in the buy box. These are commercial business records, not personal data, so GDPR and equivalent personal-data regimes do not attach to this output the way they would to a profile scraper.

The obligations that do apply are contractual and database-related: Amazon's terms of service, and any local database-right or unfair-competition rules covering systematic reuse of a compiled catalogue. Consult legal counsel for commercial use cases — in particular before republishing extracted listings as a competing catalogue or reselling the data as a standalone product.

❓ Frequently asked questions

Does Amazon ASIN Scraper work without an Amazon account?

Yes. No Amazon account, login, cookie or session is used. Requests are anonymous, and every field returned is visible to any visitor on the public detail page. The only credential you need is your Apify token.

How often is the scraped data updated?

Every run fetches the live page. There is no cache and no stored snapshot — price, warehouseAvailability, deliveryDateText and the variation matrix are read from the HTML Amazon served during that run. To track change over time, schedule the run in Apify and diff consecutive datasets on asin.

What happens when an ASIN is invalid, delisted, or not available on the chosen marketplaceDomain?

It fails as a record, not as an empty search, and there are three distinct outcomes. An ASIN that does not match the 10-character alphanumeric pattern is dropped before fetching starts, with a log warning; if no valid ASIN remains, the run fails with an error naming the asins input. An ASIN that is valid in shape but returns 404 or 410 on your chosen storefront — the usual outcome for a delisted product, or one that simply does not exist on that marketplace — produces an uncharged {"asin": "...", "error": "fetch_failed"} row once the fetch gives up. A page that loads but breaks the parser produces {"asin": "...", "error": "parse_error: ..."}, also uncharged. Every other input in the same run continues normally.

A live ASIN with no twister on its page is not a failure: it returns a full product row with hasVariations: false, an empty variationMatrix and variationCount: 0. A single-variant product is reported as single-variant rather than dressed up as a one-row family.

Can I scrape login-gated Amazon content or review text?

No. Only publicly accessible page content is returned. Review text in particular is not part of the output: Amazon serves no review nodes on the logged-out detail page and the alternate review surfaces return a sign-in shell, so on-page review extraction was removed rather than left returning empty fields. What you do get is the review metadata the public page states — stars, starsBreakdown, reviewsCount, answeredQuestions and a reviewsLink you can follow yourself.

What am I charged for?

The Actor bills per pushed row through the row_result pay-per-event. One product row is one event; each variant row emitted by variantsAsSeparateRows is another, which is why maxVariantRowsPerProduct doubles as a cost control. Error rows are pushed with no charged event at all. Products dropped by onlyProductsWithVariations are fetched and parsed but never pushed, so they are never charged either — though they do still appear in the OUTPUT key-value store record, which is the easiest way to see what the filter removed.

Does Amazon ASIN Scraper work for AI agent workflows and LLM pipelines?

Yes. It is callable as a standard HTTP endpoint through the Apify API, so LangChain, CrewAI, n8n or a hand-written tool definition can invoke it and receive typed JSON directly. Every response is normalized JSON with a fixed key set — no parsing step before the data reaches a model, and no HTML in the payload.

How does Amazon ASIN Scraper handle Amazon's anti-bot system?

Through proxying, session handling and a retry ladder — there is no inline CAPTCHA solver in this build. Requests go out through Apify Proxy, defaulting to the residential group; picking a country in exitCountry pins the exit, and proxyConfiguration overrides it entirely if you need your own URLs. Each ASIN gets its own sticky Apify session on the first attempt, so a primed delivery cookie stays on one IP; every retry rotates to a fresh IP instead.

A response is treated as blocked when the body carries a CAPTCHA or robot-check marker, not only when the status is non-200 — so a soft block returning HTTP 200 is caught rather than parsed into an empty record. Failed attempts back off with jitter before retrying: three attempts by default, four when you pin a specific exit country. Choosing US adds a geo-verification pass — the Actor checks the page's own delivery country, location line, currency and marketplace ID, retries if the page did not render as US, and falls back to a best-effort response if it never does. A 404, 410 or proxy-auth failure stops the retries immediately, since none of those is worth another attempt. Requests carry a rotating desktop User-Agent and a full browser header set rather than a default HTTP client signature.

Can I use Amazon ASIN Scraper without managing proxies?

Yes. The proxy input is prefilled with Apify Residential and exitCountry defaults to auto-select, so a run works with no proxy setup at all — you never create a proxy account or rotate an IP. If Apify cannot create the proxy configuration for any reason, the run logs a warning and continues on a direct connection rather than failing, which is worth knowing because a direct connection to Amazon is far more likely to be blocked.

What does setting a delivery postcode actually change?

deliveryPostcode runs a one-off priming sequence before any product is scraped: the Actor loads the storefront, reads Amazon's anti-CSRF token from the address-selection endpoint, and posts your postcode to the glow address-change endpoint. The resulting cookie is attached to every product request in the run, so deliveryMessage, deliveryDateText, fastestDeliveryText and warehouseAvailability are the strings Amazon renders for that location. Leave it empty and no priming happens — requests still carry a fixed US-English locale cookie, so what you get is Amazon's default location for the proxy exit IP. If priming fails, the run logs a warning and carries on without it rather than aborting.

How does Amazon ASIN Scraper compare to other Amazon ASIN scrapers?

Checked on the Apify Store on 25 July 2026, junglee/amazon-asins-scraper is the most established Amazon ASIN Actor and the closest comparison — its README documents ASIN and offer data, an error-item structure with an error-code reference table, multiple export formats and platform integrations, and its JSON example shows a ten-field record built around title, brand, price, stars and breadcrumbs. It does not document a variation matrix, specification tables or per-image resolutions. The other two closest listings are on different platforms: clearpath/shein-product-scraper documents Shein size charts, size conversions, model measurements and fit from keyword searches, and s-r/free-google-shopping-variants documents variant data by Google Shopping SKU. What this Actor documents that they do not is the Amazon variation family itself — every child ASIN with its dimension values, which one is selected, a choice between nesting it or exploding it into rows, and the specification tables and stated image resolutions from the same single fetch.

Does Amazon ASIN Scraper return data in a format LLMs can use directly?

Yes. Typed, normalized JSON with stable field names on every run. No HTML, no selectors, no parsing. Pass a row directly into an LLM context window, index it into a vector store, or route it through an agent tool. For RAG indexing, title, productSpecification, productDetails and variationDimensions carry the most descriptive text per record and chunk cleanly by asin.

What happens when Amazon changes its structure or blocks the scraper?

The scraper is maintained and the output schema stays stable — field names and types do not change on your end. The variation reader carries four independent strategies and falls through them in order, and the specification, media and fulfilment extractors each try several layouts, including the newer detail-page layouts where the classic selectors return nothing. A structural change on Amazon's side typically empties one field rather than failing a whole row, and a blocked request is retried through a fresh proxy session before it ever becomes an error row.

🔗 Related scrapers

Scraper Name	What it extracts
Amazon Product Scraper	Product records from Amazon category and product URLs
Amazon Today's Deals Scraper	Deal listings from Amazon's Today's Deals surface
Amazon Trending Products Scraper	Trending and fast-moving products by category
Alibaba Scraper With Supplier Contact Leads	Alibaba product rows plus outreach-ready supplier records

💬 Your feedback

Found a bug, or need a field that is on the Amazon detail page but not yet in the output? Open an issue on the Actor's Issues tab. Reports that include the exact ASIN, the marketplaceDomain you used and the input JSON are the fastest to reproduce and fix.