Amazon Product Scraper

What is Amazon Product Scraper and How Does It Work?

Amazon Product Scraper is a powerful web scraping tool that extracts product data from Amazon using product URLs. It collects product details like price, stock availability, ratings, reviews, seller info, and more — all exportable in JSON or CSV format.

Why Use Amazon Product Scraper?

Use this scraper to:

• Monitor pricing and stock for competitor products
• Track product ratings and reviews for market analysis
• Benchmark product performance within categories
• Automate data extraction for eCommerce analytics and advertising optimization

Features

• Scrapes product details from Amazon product URLs
• Supports both single URL and multiple URLs as input
• Extracts comprehensive product data:
  • Title
  • Price
  • Images (all product images)
  • Variants and their prices
  • Ratings (average rating)
  • Reviews (configurable number of reviews)
  • Product details and specifications
  • Product description
• Smart proxy rotation to avoid IP blocking
• User-agent randomization for better reliability
• Automatic retry mechanism for failed requests
• Support for multiple Amazon regional domains

How to Use Amazon Product Scraper on Apify

Using the Actor

To use this scraper on Apify, follow these simple steps:

1. Go to the Amazon Product Scraper on Apify platform

2. Input Configuration:

   • Enter one or more Amazon product URLs you want to scrape
   • Configure which data components to scrape:
     • Product reviews
     • Product variants/options
     • Detailed product specifications
     • Seller information
   • Select language settings if needed
   • Choose proxy country or use auto-select

   Input Configuration

The actor accepts the following input parameters:

{
  "productUrls": [
    "https://www.amazon.com/product-url-1",
    "https://www.amazon.com/product-url-2"
  ],
  "scrapeReviews": true,
  "scrapeProductVariants": true,
  "scrapeProductDetails": true,
  "scrapeSellerInfo": true,
  "language": "en",
  "proxyCountry": "AUTO_SELECT_PROXY_COUNTRY"
}

• productUrls (required): One or more Amazon product URLs to scrape (array of URLs).
• scrapeReviews (optional): Whether to scrape product reviews (default: false).
• scrapeProductVariants (optional): Whether to scrape product variants/options (default: false).
• scrapeProductDetails (optional): Whether to scrape detailed product specifications (default: false).
• scrapeSellerInfo (optional): Whether to scrape seller information (default: false).
• language (optional): Language to use on Amazon (default: "en", options include: en, de, es, fr, it, ja, zh_CN, pt, nl, pl, tr, ar, sv, ko, hi, cs, da, he, ru, th).
• proxyCountry (optional): Country for proxy (default: "AUTO_SELECT_PROXY_COUNTRY", options include: US, GB, DE, FR, JP, CA, IT).

3. Run the Actor:

   • Click the "Start" button to begin scraping
   • The actor will process each URL and extract the requested data

4. Access Your Results:

   • Once complete, view your results in the "Dataset" tab
   • Download the data in your preferred format (JSON, CSV, Excel, etc.)
   • Alternatively, access the data via the Apify API

5. Schedule Regular Runs (Optional):

   • Set up scheduled runs to monitor products over time
   • Configure webhooks to receive notifications when runs complete

Output

The actor outputs a dataset with detailed information about each product. Each item in the dataset contains:

• url: The URL of the product page
• title: Product title
• price: Current product price
• images: Array of product image URLs
• details: Object containing product specifications and details
• description: Product description
• average_rating: Average product rating (out of 5)
• review_count: Total number of reviews
• variants: Array of product variants with their names and prices
• reviews: Array of product reviews, each containing:
  • title: Review title
  • rating: Individual review rating
  • date: Review date
  • text: Review content
  • reviewer: Reviewer name

How the Scraper Works

The Amazon Product Scraper uses Playwright, a modern headless browser automation library, to navigate Amazon product pages and extract data. Here's how it works:

1. Browser Automation: The actor launches a headless browser instance that mimics real user behavior when visiting Amazon pages.

2. Data Extraction Process:

   • Navigates to each product URL
   • Waits for critical page elements to load
   • Uses both DOM parsing and JavaScript evaluation to extract product information
   • Handles different page layouts and element structures across various Amazon domains
   • Extracts structured data including product details, prices, images, and reviews

3. Variant Handling: The scraper can detect and extract information about product variants (sizes, colors, models) by:

   • Identifying variant selectors on the page
   • Simulating clicks on variant options
   • Capturing price and availability changes for each variant

4. Review Extraction: For reviews, the scraper:

   • Navigates to the review section
   • Paginates through reviews up to the configured maximum
   • Extracts review text, ratings, dates, and reviewer information

5. Error Handling: The actor implements robust error handling with automatic retries for failed requests, timeouts, and navigation errors.

Anti-blocking Measures

This actor uses several techniques to avoid being blocked by Amazon:

• Random user agent rotation: Cycles through different browser user-agent strings to avoid detection
• Headless browser with realistic behavior: Simulates human-like navigation patterns and timing
• Proper request throttling and timing: Adds random delays between actions to avoid triggering rate limits
• Smart proxy management: Can use different proxies based on the target Amazon domain
• Session management: Maintains consistent sessions to appear as a legitimate user

Performance Optimization

The actor is optimized for both speed and reliability:

• Parallel processing of multiple URLs (when multiple URLs are provided)
• Efficient memory management for processing large numbers of products
• Selective data extraction based on configuration options to improve performance

Limitations

• The actor may not work for all Amazon regional domains
• Some product details may vary depending on the product category
• Amazon's website structure changes frequently, which may require updates to the scraping logic

License

This project is licensed under the MIT License - see the LICENSE file for details.