Give it a list of products - ASINs or URLs - and get their reviews back as rows: the reviewer name, the review text, the star rating and the date. Sort and filter before you run, across seventeen Amazon marketplaces. Amazon publishes no free reviews API, so this is the practical way to get the full picture instead of the first page.
The input is the product, not the review - you name the items you want feedback on and the job collects what shoppers wrote about them.
One row per review, each tagged with the product it belongs to - so a run across many ASINs still reconciles back to your input list.
Amazon keeps review content in the page rather than exposing a public feed. That's precisely the gap this closes - without you writing or maintaining a parser.
Narrow the run up front rather than pulling everything and discarding most of it. The estimator shows the review count and credit cost before the job starts.
The same product is reviewed differently in different countries. Pull several storefronts for one ASIN and the regional split becomes visible in a single sheet.
Each row carries the review as Amazon displays it: the reviewer name, the review text, the star rating and the date it was posted - tagged with the product it belongs to so multi-ASIN runs stay sortable by item.
Review text comes back verbatim, with no preprocessing applied, so downstream sentiment analysis reads the shopper's own wording rather than ours. We don't publish a fixed column list here - a small free-tier run returns the definitive header row for the products and marketplaces you're actually pulling.
Each row carries the review, its author, and the product it was left on - so an export stands alone without a join back to the listing.
Three columns never filled, and four are misspelled at source. comments, official_comment_banner and autor_descriptor were empty on all 80 rows. And the export spells author as autor across four columns and badge as bage - reproduced above exactly as they appear, because your code has to match the real header rather than the correct English. Note date is a long-form string like "January 14, 2026" rather than ISO 8601, so it needs parsing before any time-series work.
Every regional storefront we support, by domain. A product's reviews differ per marketplace - the same ASIN can be loved in one country and returned in another.
A few examples of how teams use Amazon review data.
Pull the full review history for your own ASINs and read the one- and two-star text as a corpus. Recurring failure modes show up as repeated language - far more actionable than an average rating that moved 0.1.
Feed in a rival's ASINs and read what their buyers dislike. The gaps customers name in their own words are the sharpest input you'll get for positioning or a product change.
Take the top ASINs in a category and export the lot. With verbatim text and ratings in one file, you can run your own model or LLM over a real sample instead of a handful of reviews you read manually.
No subscription, no minimum. Your first 500 reviews are on us - after that, pay-as-you-go, and the rate halves at volume.
Every new account, one-time. No credit card required. CSV / XLSX export, sorting, filtering and API access included.
That's $2 per 1,000 reviews above the free tier. The pre-flight estimator shows the review count and credit cost before a run starts - no surprises, no compute units to translate.
For continuous category monitoring or very large historical pulls, talk to us about volume pricing, dedicated workers and an SLA.
LIVESCRAPER10 at checkout.Most teams run these two against the same ASIN list.
Same input, different output: title, description, price, availability and rating per ASIN. Run both and you have the listing and its reception side by side.
If you sell through a storefront as well as a marketplace, the same reputation work applies to your Google Business Profile - every review, every location.
Short answer: yes for the public review content - and that is all we ever collect.
Reviews are published to be read. Amazon shows the reviewer's display name, the text, the rating and the date to anyone visiting the page, signed in or not. Collecting publicly visible feedback for product and market research is long-established practice, and as long as the data is publicly available and the process doesn't disrupt the site there are no federal laws prohibiting it.
Two things worth stating plainly rather than burying. Amazon's terms of service restrict automated access, so this is a terms question as well as a legal one - if you have a contractual relationship with Amazon, check it. And reviews carry a reviewer's display name, which makes the output partly personal data: analysing sentiment across a category is a defensible purpose, building a picture of an individual shopper is not.
We touch nothing behind a login, run no third-party trackers on the data layer, and your exports auto-delete after 30 days.
The questions we hear most. Anything else? Talk to us - humans, not bots, write the answers.
500 one-time free reviews on every new account - no expiry. After that it's $0.002 per review, pay-as-you-go - no card on file until you say so.
Livescraper's Amazon Reviews Scraper turns a product list into review data. You submit ASINs or product URLs, choose how the reviews should be sorted and filtered, and download what shoppers wrote as a clean CSV or Excel file - or call the API and skip the download entirely.
Each row carries the review as Amazon displays it: the reviewer name, the full review text, the star rating and the posting date, tagged with the product it belongs to so a multi-ASIN run still reconciles against your input. Text is preserved verbatim, which matters if the next step is a sentiment model - the analysis reads the shopper's wording rather than ours.
Product teams pull their own ASINs and read the one- and two-star text as a corpus, where recurring failure modes appear as repeated language rather than a rating that drifted a tenth of a point. Competitive teams read what a rival's buyers dislike, in the buyers' own words. Research teams export a whole category to run their own analysis over a real sample instead of the few reviews a person can read by hand.
Amazon publishes no free reviews API - review content lives in the product page - so the alternative is a parser you maintain yourself. Seventeen regional marketplaces are supported, which is how the same ASIN's reception in Germany and Japan ends up comparable in one sheet. Start free: your first 500 reviews cost nothing and need no credit card, and the rate halves past five thousand.