Amazon Reviews Scraper — review text, star ratings, verified purchases and review images

Amazon Reviews Scraper extracts customer reviews from Amazon product pages by ASIN or product URL. Every review comes back as one row with the star rating, review title, full review text, author, date, country, Verified Purchase badge, helpful votes, the product variant it was written for, the review's language, and any photos the reviewer attached. Export to JSON, CSV, Excel or XML, or pull it straight from the Apify API.

No Amazon account, no session cookie and no headless browser. You give it products, it gives you reviews.

This Actor is an independent tool and is not affiliated with, endorsed by, or sponsored by Amazon.

What does Amazon Reviews Scraper do?

It reads the customer reviews Amazon publishes on a product page and turns them into structured data. You can point it at:

• ASINs such as B07N7KFHVH, pasted as a list
• Amazon product URLs such as https://www.amazon.com/dp/B07N7KFHVH
• Amazon review URLs — the product ID is taken from the link

It works across Amazon marketplaces: amazon.com, co.uk, de, fr, it, es, ca, com.mx, in, co.jp and com.au. A URL's own domain always wins, so a list mixing amazon.com and amazon.de links is read on both sites without extra configuration.

What data does it extract from each Amazon review?

Field	Description
review_id	Amazon's own review ID, stable across runs and across marketplaces. Use it to deduplicate when monitoring.
rating	Star rating as a number, 1 to 5
title	Review headline
body	Full review text, with the reviewer's paragraph breaks kept
author_name	Reviewer's display name
author_profile_url	Link to the reviewer's public Amazon profile
review_url	Permalink to the review
date	Review date as ISO YYYY-MM-DD
date_raw	The date exactly as Amazon printed it
country	Country the review was written in, for example "United States"
verified_purchase	True when Amazon shows the Verified Purchase badge
is_vine	True for Vine Customer Review of Free Product
helpful_votes	Number of people who found the review helpful, as a number
variant	The variant the review is about, for example "Color: Large | Copper"
variant_asin	That variant's own ASIN
language	Language the review was written in, for example en-US
images	Reviewer photos, as full-size image URLs
asin	The product you asked for
source_asin	The product page this review was read from, which differs when variants are included
marketplace	Which Amazon site it came from, for example amazon.co.uk
product_title, product_rating, product_rating_count, product_rating_distribution	Product context on every row, so a single review row is enough for analysis
position, scraped_at	Order within the run and the UTC timestamp

How many Amazon reviews can you scrape per product?

Amazon publishes about 8 to 10 reviews per product page to visitors who are not signed in. The dedicated reviews pages, /product-reviews/<ASIN> and the star-filter links, redirect to a sign-in form, and the pagination endpoint the page used to call now answers 404. Measured on 2026-09-20 with a real browser as well as an HTTP client, so it is Amazon's behaviour, not a limit of this Actor. Any tool promising "all reviews" without a login is promising something Amazon stopped serving.

Amazon Reviews Scraper gets past that ceiling by merging sets of reviews instead of paginating one:

• Read the same product on several marketplaces. Amazon shows a different set of reviews per storefront. Adding co.uk, ca and com.au typically turns 8 reviews into 25 to 35. Reviews are deduplicated by review ID, so nothing is counted twice.
• Read the product's variants. Each colour or size has its own page with its own reviews. A product with ten variants has roughly ten times the reviews available.

Both are off by default, because both cost extra page fetches. Turn them on when you need depth.

How do I scrape Amazon reviews?

1. Add your ASINs, or paste Amazon product URLs.
2. Set Maximum reviews per product. It is a budget: the Actor stops fetching as soon as it has enough.
3. To go deeper, list extra marketplaces in Also read these marketplaces, or set Also read this many product variants.
4. Optionally filter by star rating, Verified Purchase or date.
5. Keep the residential proxy setting on. Amazon serves a stripped page with no reviews to datacenter addresses.
6. Run it, then download the dataset as JSON, CSV or Excel, or read it from the API.

Input options

Option	What it does
asins	Product IDs to scrape
startUrls	Amazon product or review URLs
marketplace	Which Amazon site to use for bare ASINs
maxReviewsPerProduct	Upper bound per product
alsoSearchMarketplaces	Extra storefronts to merge, for example co.uk, ca
maxVariantsPerProduct	How many colour or size variants to read as well
minRating, maxRating	Keep only reviews in a star range
verifiedPurchasesOnly	Keep only Verified Purchase reviews
reviewsStartDate, reviewsEndDate	Keep only reviews in a date range
includeProductDetails	Add product title, overall rating and star distribution to each row
concurrency	How many products to fetch at once
proxyConfiguration	Residential proxy settings

Output example

{
  "review_id": "R3QZ12747962W7",
  "asin": "B07N7KFHVH",
  "variant_asin": "B0D5DM136R",
  "rating": 5,
  "title": "great product",
  "body": "I am very pleased with this garlic press. It presses garlic quickly with very little effort.\n\nMy only regret is that I wish I had purchased the larger model.",
  "author_name": "George Pinell",
  "author_profile_url": "https://www.amazon.com/gp/profile/amzn1.account.AE3SWAAR6UPKOHMUEPU2I3SUZUNA",
  "review_url": "https://www.amazon.com/portal/customer-reviews/srp/-/R3QZ12747962W7",
  "date": "2026-07-20",
  "date_raw": "Reviewed in the United States on July 20, 2026",
  "country": "United States",
  "verified_purchase": true,
  "is_vine": false,
  "helpful_votes": 5,
  "variant": "Color: Large | Copper",
  "language": "en-US",
  "images": [],
  "marketplace": "amazon.com",
  "product_title": "Zulay Kitchen Premium Garlic Press Set",
  "product_rating": 4.6,
  "product_rating_count": 51884,
  "product_rating_distribution": { "5_star_pct": 78, "4_star_pct": 13, "3_star_pct": 5, "2_star_pct": 2, "1_star_pct": 2 },
  "position": 1,
  "scraped_at": "2026-09-20T15:40:12.001Z"
}

Can I filter Amazon reviews by star rating, verified purchase or date?

Yes. Set a star range, a Verified Purchase toggle, or a date range, and only matching reviews are written to the dataset.

One thing worth knowing: Amazon no longer applies those filters on the public product page, so no tool can ask Amazon for one-star reviews only. Filtering happens after the page is read. You get exactly the reviews you asked for; they simply cost the same page fetch as an unfiltered run.

How much does it cost to scrape Amazon reviews?

You pay per review returned, plus Apify platform usage. See the Pricing tab for the current rate. Because the price is per review, a product with four reviews costs a fifth of one with twenty, and a run that returns nothing costs nothing but platform time.

Is it legal to scrape Amazon reviews?

Scraping publicly available data is broadly lawful in the US and the EU, and this Actor reads only what Amazon shows to any visitor without an account. It does not log in, does not bypass a paywall and does not collect anything behind authentication.

Reviews do contain personal data, such as a reviewer's display name and profile link. If you are in the EU or handling EU residents' data, you need a legal basis under the GDPR for storing it, and you should drop the author fields when you do not need them. Use the data responsibly, and follow Amazon's terms and applicable law. This is not legal advice.

What can you do with Amazon review data?

• Product research. Read what buyers complain about before sourcing a product, and size the complaint by how many people found the review helpful.
• Sentiment analysis. Feed review text to a language model or a classifier. Each row carries the rating as a label.
• Competitor monitoring. Track a rival's new reviews on a schedule and alert on one- and two-star arrivals.
• Quality control. Catch a defect pattern early by filtering to critical reviews from the last 30 days.
• Listing optimisation. See which variant attracts the complaints and fix the listing or the packaging.

Using Amazon Reviews Scraper with the Apify API

Start a run and collect the dataset:

curl -X POST "https://api.apify.com/v2/acts/automly~amazon-reviews-scraper/run-sync-get-dataset-items?token=<YOUR_API_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{"asins":["B07N7KFHVH"],"maxReviewsPerProduct":10}'

The Actor also works with the Apify Python and JavaScript clients, with scheduled runs, and with integrations such as Zapier, Make, Google Sheets, Slack and webhooks.

Frequently asked questions

Do I need an Amazon account or a session cookie? No. Amazon Reviews Scraper reads only the reviews Amazon publishes to anonymous visitors. You never hand over credentials.

How many reviews will I actually get per product? About 8 to 10 from a single marketplace. Adding three more marketplaces usually brings 25 to 35, and reading variants adds roughly another 8 per variant. The numbers depend on how many reviews the product has.

Why do some products return fewer reviews than the product page shows? The rating count on a listing counts ratings without text as well. Only reviews with text are published on the page, and only those are returned.

Can it scrape reviews in other languages? Yes. Reviews are returned exactly as written, and each row carries a language field, so a pull across marketplaces stays sortable by language.

Does it return review photos and videos? Photos, yes, as full-size image URLs. Videos are not returned.

Can I scrape reviews for a whole category or a search term? Not directly. Give it products. Pair it with an Amazon product scraper to turn a search or a category into a list of ASINs, then feed those here.

How do I avoid being blocked? Keep the residential proxy setting on and leave concurrency at the default. Amazon serves a stripped page without reviews to datacenter IP addresses, and this Actor reports that case as an error rather than silently returning nothing.

Can I run it on a schedule and only get new reviews? Yes. Schedule the Actor, then deduplicate on review_id, which is stable across runs and marketplaces.

What happens if a product has no reviews? You get no rows for it and a line in the log. A product whose page came back without the review section is reported separately as an error, so a block never looks like an empty product.

Support

Found a product that does not parse, or a field you need? Open an issue on the Actor's Issues tab with the ASIN and the marketplace, and it will be looked at.