Amazon is the hardest major website to scrape. Not because the data is hidden — every product page shows price, title, reviews, and availability publicly. The problem is detection.
Here's what actually works in 2026.
Why traditional Amazon scrapers fail
Most scraping tools run from cloud servers. The moment you send a request from a datacenter IP to Amazon, you're flagged. Amazon's bot detection checks:
- IP reputation — datacenter ranges are blocklisted
-
TLS fingerprint — libraries like
requestsoraxiosproduce handshakes that don't match real Chrome - Behavioural signals — request rate, header patterns, missing cookies
The result: you get served fake prices, empty pages, or a CAPTCHA wall. Worse, you often can't tell the data is poisoned until you spot anomalies weeks later.
The browser-native fix
Running your scraper inside an actual Chrome browser eliminates all three detection vectors:
- Your IP — residential, not a datacenter
- Real TLS handshake — generated by Chrome itself, not a library
- Real session — your cookies, your browsing history, your fingerprint
There's no bot to detect because you're browsing normally.
What you can extract from Amazon
From product detail pages:
- Title, brand, ASIN
- Current price, list price, sale price
- Prime eligibility, shipping estimate
- Star rating, review count
- Stock status, seller info
- Bullet points, description, specs
From category and search pages:
- All products in a result set
- BSR (Best Seller Rank)
- Sponsored vs. organic placement
- Price ranges across variants
Step-by-step with a Chrome extension
- Open the Amazon product page or search result in Chrome
- Launch Clura from the Chrome toolbar
- Clura auto-detects product listings or detail fields using heuristic pattern matching
- Select the fields you want — title, price, rating, ASIN
- Export to CSV or Google Sheets
No selectors, no code, no proxy rotation. The extension handles pagination automatically for category pages.
Handling common edge cases
Dynamic pricing: Amazon shows different prices based on your location, login state, and Prime membership. Browser-native scraping captures your actual price — exactly what a customer in your session would see.
Variants: Product pages with size/colour variants require clicking each variant to get its price. Use Clura's multi-page extraction to loop through variants automatically.
Reviews: Review pages paginate in sets of 10. Set up a workflow that follows the "Next page" link and accumulates all reviews across pages.
Scale considerations
A single Chrome session can realistically scrape 50–200 Amazon pages per hour at human-like intervals. For larger catalogs, run multiple sessions across different time windows. Avoid hammering the same category page — spread requests across product URLs rather than hitting the same search result repeatedly.
For most businesses tracking 50–500 competitor ASINs daily, a single browser-native session is plenty.
The full step-by-step workflow, including how to handle Amazon's anti-bot measures and export formats, is in the Amazon scraper guide on Clura.
