Amazon Product Scraper

🛒 Amazon Product Scraper

Extract public Amazon product data without an account: title, ASIN, price, brand, rating, review count, images, variants and the featured seller. Paste product, search, category or bestseller URLs and get one clean row per product, ready for JSON, CSV or Excel.

Results cost $0.70 per 1,000 products, and the extras (offers, seller ratings, delivery pricing) are optional and charged only when you switch them on.

✨ Why use Amazon Product Scraper?

• 🔓 No login needed — it reads public catalog pages only.
• 🧩 Rich product rows — price, brand, stars, review count, breadcrumbs, features, high-resolution images and variants.
• 🔗 Flexible inputs — product URLs (/dp/, /gp/product/), search and category URLs, bestsellers, and a.co / amzn.to short links.
• 🎚️ Pay for what you use — offers, seller ratings and delivery locations are separate, opt-in events.
• 🌍 Storefront control — set language, proxy country, and an optional delivery country or ZIP.
• 🛡️ Built-in retries — blocked pages are retried through a residential exit, and optionally through a challenge solver.

🎯 Use cases

• 💲 Price and stock tracking — monitor a list of ASINs on a schedule.
• 📚 Catalog building — pull the first page of a search or category into a spreadsheet.
• 🕵️ Competitor research — compare featured sellers, ratings and delivery promises for a keyword.
• 🖼️ Image and listing collection — gather titles, features and images for comparison.
• 📈 Market research — follow bestseller pages in a category.

🚀 How to use

1. Add one or more Amazon URLs under Amazon URLs (a product page, a search such as https://www.amazon.com/s?k=keyboard, a category or a bestsellers page).
2. Set Max results per URL and, for searches, Max search pages per URL.
3. Optionally turn on offers, seller ratings or a delivery ZIP.
4. Click Start and watch rows arrive in the dataset.
5. Export to JSON, CSV or Excel, or read them through the API.

📥 Input

{
  "categoryOrProductUrls": [
    { "url": "https://www.amazon.com/dp/B09X7MPX8L" }
  ],
  "maxItemsPerStartUrl": 10,
  "scrapeProductDetails": true,
  "proxyConfiguration": {
    "useScrapelyProxy": true,
    "scrapelyProxyGroups": ["DATACENTER"]
  }
}

📤 Output

Each product is one dataset row. A URL that cannot be read saves no row: it is listed with the reason in the run's status message and in the OUTPUT record under failedInputs.

Example item (trimmed):

{
  "title": "SANDISK 1TB Extreme microSD Card + Adapter, Up to 190MB/s Read Speeds",
  "url": "https://www.amazon.com/dp/B09X7MPX8L",
  "asin": "B09X7MPX8L",
  "price": { "value": 199.75, "currency": "$" },
  "inStock": true,
  "inStockText": "In Stock",
  "brand": "SANDISK",
  "stars": 4.8,
  "reviewsCount": 149718,
  "breadCrumbs": "Electronics > Computers & Accessories > Computer Accessories & Peripherals > Memory Cards > Micro SD Cards",
  "thumbnailImage": "https://m.media-amazon.com/images/I/716kSUlHouL._AC_SY300_SX300_QL70_FMwebp_.jpg",
  "highResolutionImages": [
    "https://m.media-amazon.com/images/I/716kSUlHouL._AC_SL1500_.jpg"
  ],
  "features": [
    "Compatible with Nintendo-Switch (NOT Nintendo-Switch 2)"
  ],
  "variantAsins": ["B0G8M2FH2G", "B09X7C2GBC"]
}

💰 Pricing

You pay per event, plus platform usage (compute and proxy traffic).

Worked example: 1,000 products with default settings cost $0.70 in result events. Failed URLs are not charged. Platform usage is extra: compute at $0.30 per GB-hour, datacenter proxy at $1 per GB, and residential proxy at $2 per GB (used only when a datacenter exit is blocked).

❓ FAQ

Is it legal to scrape Amazon? The actor reads publicly visible pages only. You are responsible for using the data in line with applicable law and Amazon's terms.

Do I need a login or a proxy? No login. Proxies are included and selected for you.

Which URLs work? Search and category URLs containing /s or /b/, bestsellers URLs, product URLs containing /dp/ or /gp/product/, and short links on a.co or amzn.to. Other hosts return an invalid_url row.

Why is a price missing? Amazon hides the featured price when a product does not ship to the proxy's location. Set a matching proxy country or a delivery ZIP.

What if my delivery ZIP cannot be applied? Each product page is checked after the location is set. If Amazon rejects the ZIP (for example an invalid one) or a page still shows another place, the product is saved with deliveryLocationApplied: false and the location it showed in deliveryLocationShown, and the delivery-location event is not charged for it. The status message and the OUTPUT record say how many results got your location.

How many products can one URL return? Up to the limit you set. Search pages often stop after a few pages on Amazon's side, so narrow the category if you need more.

Can I schedule it or use the API? Yes. Use schedules for price monitoring and the API to start runs and read results.

⚠️ Limits

• Public catalog pages only; no Amazon account login and no separate review-text rows.
• Search and category pages are the hardest to read and can be blocked; they are retried through a browser and may still return an invalid_url row with a "blocked" description.
• Single-product runs may leave out some fields: attributes, bestseller ranks, star breakdown, description and A+ content are not always available.
• Amazon shows about 20 result pages per search or category (roughly 200 to 250 products). The run stops there and the status message says how many were found; split a broad search into narrower ones to get more.
• Offers and seller ratings need the matching input options.
• The offers lookup can return an empty list for a product that otherwise parses.