How to scrape Amazon

A practical guide to extracting Amazon product data: search results, product pages, reviews, variants, prices, seller data, and anti-bot considerations.

Amazon is one of the most valuable and difficult e-commerce sources to scrape. Product pages contain price, seller, reviews, variants, rankings, images, and specifications, but the site is dynamic, localized, heavily monitored, and different Amazon marketplaces can expose different layouts.

Start by deciding which Amazon page type you need. Search results, product details, and reviews are separate workflows.

Page types

Page type	Best for	Typical fields
Search results	Product discovery and ranking	Title, URL, image, price, rating, review count, sponsored flag, ASIN
Category pages	Category research and assortment	Product cards, rank position, category URL, price, rating
Product detail pages	Deep product intelligence	Title, brand, ASIN, price, seller, bullets, description, specs, variants, images, BSR
Review pages	Sentiment and quality analysis	Review title, rating, text, date, reviewer, verified purchase, helpful count
Seller pages	Marketplace monitoring	Seller name, storefront URL, seller rating, offer coverage

Octoparse’s Amazon template page follows the same split: listing pages for broad product collection, detail pages for richer product facts, Prime-specific collection, and review scraping for feedback analysis. Apify and Bright Data use similar product/detail/review divisions in their Amazon scrapers and datasets.

Step 1: Pick the input

Use keywords when you need discovery:

wireless keyboard
protein powder
kids winter boots

Use product URLs or ASINs when you already know the catalog:

https://www.amazon.com/dp/B08...
B08...

For price monitoring, ASINs or product URLs are usually better than repeating keyword searches. For market research, keyword and category searches reveal competitors and category movement.

Step 2: Collect listing data

Search and category pages are useful for breadth. Capture:

• Product title
• Product URL
• ASIN when available
• Price
• Rating
• Review count
• Image URL
• Sponsored/organic signal when visible
• Search keyword or category
• Result position
• Timestamp

The result position matters. A product’s rank for a keyword can be as important as its price.

Step 3: Visit product detail pages

Detail pages provide the fields needed for serious analysis:

• Brand
• Feature bullets
• Product description
• Specifications table
• Variants
• Seller and fulfillment signals
• Best Sellers Rank
• Stock or delivery hints
• Full image set
• Coupon or promotion signals

Amazon pages often change based on location, account state, and marketplace. Preserve the marketplace domain, delivery region, and scrape timestamp so you can interpret differences later.

Step 4: Scrape reviews separately

Review extraction is its own workflow. Sort order, pagination, language, star filters, and region can change the output.

Useful review fields:

• Rating
• Review title
• Review text
• Review date
• Reviewer display name
• Verified purchase flag
• Helpful vote count
• Product variant
• Review URL

For sentiment analysis, collect enough context to avoid mixing variants. Reviews for one size, color, or bundle can be misleading when analyzed as if they describe the whole product.

Technical challenges

Amazon scraping commonly runs into:

• Dynamic layouts and A/B tests
• Regional price and availability differences
• CAPTCHA and bot-detection challenges
• Variant-specific fields
• Missing or hidden seller data
• Review pagination limits
• Sponsored results mixed with organic results

Use conservative pacing, real browser rendering when needed, stable proxy and fingerprint configuration, and clear retry logic. For recurring workflows, cloud execution and subtask splitting help separate discovery, detail refresh, and review collection.

When to use templates

Use a template when you need common Amazon fields quickly: listing data by keyword, product details by ASIN or URL, Prime-filtered listings, or review data. Templates from Octoparse, Apify, and Bright Data reduce the setup work around page navigation, field mapping, pagination, and anti-blocking.

Build a custom workflow when you need unusual marketplace logic, custom SKU matching, integration with internal catalogs, or highly specific alerting.

Legal and policy notes

Amazon data can involve sellers, reviewers, and marketplace rules. Scrape only data you are allowed to use, respect robots.txt and applicable terms, avoid personal or sensitive data where possible, and prefer official APIs or partner feeds when they meet the use case.

Amazon scraping works best when the objective is narrow: monitor a catalog, analyze reviews, track competitors, or study category trends. A narrow goal keeps the field list smaller, the scrape lighter, and the output easier to trust.