Amazon Reviews Scraper — Extract Product Reviews & AI Summary as JSON

📮 Need help, hit a problem, want a new feature, or a field that's missing from the output? Email support@reefapi.com or open an issue on the Issues tab of this Actor — we reply fast and take feature requests seriously.

Scrape Amazon product reviews as clean, structured JSON. Pass an ASIN and get reviews with title, body, star rating, author, date, verified-purchase flag, variant, helpful votes and images — across 19 Amazon marketplaces — plus an AI "customers say" summary of the aspects buyers praise and complain about. No proxies, no CAPTCHAs, no login — just send an ASIN and get the data.

What is the Amazon Reviews Scraper?

The Amazon Reviews Scraper is an Amazon reviews data API that turns a product's review section into a structured dataset. Send an ASIN (and optionally a marketplace and sort order) and it returns the review list — each review's star rating, title, full body text, author, date, location, verified-purchase flag and the variant reviewed. On top of the raw reviews it returns Amazon's AI "customers say" insight: a plain-language summary of the themes buyers agree and disagree on (sound quality, fit, battery life, value). It's the tool you use to scrape Amazon reviews for sentiment analysis, product research and reputation monitoring — across all 19 marketplaces.

What you get per review

• ⭐ Star rating (1–5) and review title.
• 📝 Full review body text.
• 👤 Author name (and profile id).
• 🗓️ Date and reviewer location.
• ✅ Verified-purchase flag.
• 🧩 Variant reviewed (format — e.g. size / color / style).
• 👍 Helpful votes (when present) and 🖼️ review images (when present).
• 🧠 AI "customers say" summary — top praised and criticized aspects across all reviews.

What data can I extract?

Each review is one record. Key fields (from real output):

Field	What it is
review_id	Stable id for the review
author	Reviewer's display name
profile_id	Reviewer's Amazon profile id
rating	Star rating (1–5)
title	Review headline
body	Full review text
date	Date the review was posted
location	Reviewer's country/marketplace
verified	Verified-purchase flag
format	Variant reviewed (style/size/color)
helpful_count	"X people found this helpful" (when present)
images	Image URLs attached to the review
marketplace	Which Amazon site the review is from

The response also includes review_insights (the AI "customers say" summary), plus total_review_count and average_rating for the product.

Field availability can vary — not every review has images, helpful votes or a variant, and Amazon can change its layout. Some products expose only a limited public review set. Core fields (rating, title, body, author, date, verified) are present for real reviews.

How do I scrape Amazon reviews?

1. Open the Actor and go to the Input tab.
2. Enter an ASIN (the …/dp/<ASIN>/ segment of a product URL) and pick a Marketplace.
3. Choose a Sort — Most recent or Most helpful.
4. Click Start, then download the reviews as JSON, CSV or Excel from the Dataset tab.

Input

{ "asin": "B09B8V1LZ3", "marketplace": "com", "sort": "recent" }

• asin — the Amazon product id (from a /dp/<ASIN>/ URL).
• marketplace — com, co.uk, de, fr, it, es, co.jp, in, ca, com.au, com.mx, com.br, com.tr, ae, nl, se, pl, sg, sa.
• sort — recent or helpful.
• params — advanced JSON (star filter, verified-only, paging/cursor).

Output

Real response (trimmed — one representative review + the AI summary):

{
  "reviews": [
    {
      "review_id": "R149GR14HAGW5J",
      "author": "VOICECRACKCENTRAL",
      "profile_id": "amzn1.account.AGM7O4OPCBIBTVPHSRNJ4DDYQIMQ",
      "rating": 5.0,
      "title": "Absolutely amazing",
      "date": "June 7, 2026",
      "location": "the United States",
      "verified": true,
      "format": "Style: Pro Gen 2",
      "helpful_count": null,
      "body": "The sound quality is clear, the bass is nice, and the noise cancellation works better than I expected. They connect fast and feel comfortable…",
      "images": [],
      "marketplace": "com"
    }
  ],
  "review_insights": {
    "summary": "Customers praise the sound quality, noting they are loud and clear, and appreciate the noise cancellation. However, customers disagree on the fit, with some reporting they fall out. Battery life and connectivity are mixed…"
  },
  "total_review_count": 42718,
  "average_rating": 4.5
}

Use cases

• Sentiment analysis — feed review text into NLP/AI for themes, complaints and praise.
• Product research — read what real buyers say (and the AI "customers say" summary) before sourcing.
• Reputation & QA — monitor ratings, verified-purchase share and recurring issues over time.
• Review aggregation — build cross-marketplace review datasets for a product or brand.
• Competitor benchmarking — compare review sentiment and rating distribution against rivals.

FAQ

Is there an official Amazon reviews API? No public API returns product review text. Amazon's official programs don't expose reviews for arbitrary ASINs, so scraping the public review pages is the practical way to get this data as JSON.

Why scrape Amazon reviews? Reviews are the richest public signal of what buyers actually think. Pulled at scale, they power sentiment models, product-improvement research, QA monitoring and competitor benchmarking.

How much does it cost? Billed on Apify's pay-per-result model, benchmarked to be competitive with other Amazon review scrapers. See the Pricing section on this Actor's page for the current rate.

Can I integrate it? Yes — run it from the Apify API or the Python / Node.js clients, and connect the output to Make, Zapier, n8n, Google Sheets, webhooks or your database via Apify integrations.

Is it legal to scrape Amazon reviews? This Actor reads publicly visible reviews (no login, no accounts). Reviews contain personal data (author names), so handle the output responsibly and per GDPR/CCPA — anonymize where appropriate. Not legal advice.

---

Unofficial — not affiliated with Amazon. Public data only.