7 Best Amazon Scrapers Ranked by Performance
Amazon’s anti-scraping technology keeps getting harder to bypass. To find out which tools actually hold up, we benchmarked the leading 5 Amazon scraper APIs over 2,750 requests across 11 Amazon domains, scoring every provider on success rate and end-to-end completion time.
Amazon scraping benchmark results
Read Amazon scraping benchmark methodology for more details about our testing process.
You can also explore our full e-commerce scraping benchmark covering the top 100 e-commerce domains.
Amazon scraper performance by page type
Pricing analysis: Who is cheapest at scale?
What data you can scrape from Amazon
On Amazon product pages, only Bright Data and Apify return structured JSON (Decodo returns raw HTML through its unblocker). Bright Data returned 88 fields per product, while Apify returned 10.
The table below lists the fields that are unique to each provider, grouped by category; shared basics such as asin, brand, title, price, image, url, and rating are left out.
Product pages
Search pages
On Amazon search pages, all three providers return structured JSON. Bright Data returned 30 fields per listing, Decodo 18, and Apify 10. The table below shows the fields unique to each provider by category, again excluding the shared basics.
Amazon scrapers free trial
Detailed breakdown of Amazon scrapers
Bright Data succeeded on 94% of Amazon requests and was the fastest of the reliable providers, completing in about 14 seconds on average.
It also returned more data than any other provider, using a separate dedicated dataset for each page type: 88 fields on a product page (buybox and coupon data, best-seller ranks, answered questions) and 30 fields per listing on search. Its success was steady across domains, 98–100% on amazon.com, .co.uk, .de, .fr, .in, and .ca, with a single weak spot on amazon.com.mx (48%).
Beyond the two page types we tested, Bright Data offers 13 Amazon scrapers (product, reviews, sellers, best-sellers, and discovery by category, keyword, UPC, or brand) and 6 ready-made Amazon datasets in its Marketplace.
Oxylabs offers a general-purpose Web Scraping API with a dedicated Amazon endpoint that returns parsed Amazon data, rather than separate standalone APIs per page type. Through that endpoint it supports Amazon product, search, pricing, sellers, and best sellers pages by URL, and also exposes a universal source for arbitrary Amazon URLs alongside an Amazon-specific search source that targets each marketplace by domain.
Decodo succeeded on 91% of Amazon requests with an average completion time of about 19 seconds.
It handled the two page types differently: search went through a dedicated amazon_search endpoint that returns JSON with 18 fields per listing (price, rating, reviews, sponsored and best-seller flags, sales volume, and shipping), while product pages were fetched through its Site Unblocker as rendered HTML that you parse yourself.
Decodo’s success was high across Europe (98-100% on amazon.co.uk, amazon.de, and amazon.es) and low on amazon.co.jp (%64).
SerpApi provides two Amazon APIs, both returning JSON: Amazon Search API for organic results, filters, and product ads, and an Amazon Product API that returns product-page data queried by ASIN. Both support multiple Amazon domains and languages and accept delivery ZIP codes for location-specific pricing.
Zyte succeeded on 91% of Amazon requests with an average completion time of about 34 seconds.
It used a single API that renders each page in a headless browser and extracts it automatically, applying product extraction to product pages and productList extraction to search.
Performance varied more by domain than most: 98–100% on amazon.com, .co.jp, .it, .ca, and .com.mx, but noticeably lower on amazon.fr (58%), amazon.co.uk (82%), and amazon.de (86%).
Apify had the highest success rate of any provider at 95.3%, though it was slower, averaging about 43 seconds.
For the benchmark we used Apify’s E-commerce Scraping Tool actor (apify~e-commerce-scraping-tool), which takes product URLs in one input field (detailsUrls) and search URLs in another (listingUrls), with a country for localized results, and returned about 10 top-level fields per item (with more nested under offers and additionalProperties).
It was near-perfect on most domains (100% on amazon.com, .co.uk, .de, .co.jp, and .it), weakest on amazon.fr (66%), and slowest on amazon.com.mx (~65s). Apify also provides dedicated templates for Amazon product, review, and seller scrapers.
Nimble had a low Amazon success rate in our benchmark, around 26%.
It only returned usable data on the US-facing domains, amazon.com (92%), .com.br (98%), and .com.mx (94%), and failed almost entirely on the European, Canadian, and Japanese domains. Nimble provides an E-Commerce API for Amazon and Walmart with ZIP-code targeting and built-in residential proxies, offered on pay-as-you-go or subscription plans.
Add as preferred sourceAmazon scraping benchmark methodology
URL validation
Before the run, we checked that every product page was live and displayed a price, and that every search page loaded and returned several distinct products rather than a blank page or a single item.
Benchmark execution
We compared five providers on Amazon: Apify, Bright Data, Decodo, Nimble, and Zyte. Each provider fetched the same fresh set of URLs, split evenly between product and search pages and drawn in equal numbers from all 11 Amazon domains. Where a provider offered a dedicated Amazon scraper for a page type, we used it; otherwise we fell back to its universal source scraper or web unblocker. All providers ran from a single server location, so location gave no one an edge.
A valid HTTP status was never treated as a success on its own. We accepted responses in the 200-399 range, along with 404, as valid, and when a request returned HTTP 200 we ran a content check against ground truth defined in advance: for a structured (JSON) response we verified that the target field was present and valid, and for an HTML response we verified that the expected CSS selector was present. A request counted as a success only when that content check passed, so a block page, a captcha, or an empty shell scored zero even with a 200 status.
End-to-end response time is reported over successful requests. Where a provider returned no successful requests for a given Amazon domain, it is omitted from that domain rather than plotted at zero.
FAQs
Scraping publicly available data (such as prices, reviews, and titles) is legal, provided you do not access private user data behind a login. However, scraping violates Amazon’s Terms of Service. This is why using professional APIs with residential proxies is essential; they handle compliance and anti-ban measures for you.
The official Amazon Product Advertising API is designed for affiliates, not data analysis. It has strict rate limits and hides critical data points (like full review text or competitor stock levels). If you need deep market intelligence, the official API is usually insufficient.
Based on our data depth benchmark, Bright Data and Apify are the best choices for scraping reviews, as they capture user details, dates, and full text with high accuracy.
You can write a custom script using Python (Selenium/BeautifulSoup), but you will likely get blocked by Amazon’s CAPTCHA within minutes. To scrape at scale without paying for an API, you would need to manage your own massive pool of rotating proxies, which is often more expensive and time-consuming than buying a tool.
Cite this benchmark
Pick the format that matches where you're publishing. Pasting the link version into your CMS preserves the backlink.
@misc{sipi2026,
author = {Şipi, Nazlı},
title = {{7 Best Amazon Scrapers Ranked by Performance}},
year = {2026},
month = jul,
howpublished = {\url{https://aimultiple.com/amazon-scraper}},
note = {AIMultiple. Retrieved July 21, 2026}
}
Be the first to comment
Your email address will not be published. All fields are required. Comments are left in their original language.