Feature: amazon

Amazon Reviews

Pull structured Amazon reviews with star ratings, verified purchase flags, helpful votes, and review excerpts.

What is Amazon Reviews?

Scavio's Amazon reviews endpoint returns paginated reviews for an ASIN, each with the reviewer name, star rating, title, body, date, verified purchase flag, helpful vote count, and variant purchased (color, size). We expose rating distribution so you can see the percentage split from 1 to 5 stars at a glance, and we return the list of top positive and top critical reviews that Amazon surfaces on the page. Pagination tokens let you walk through thousands of reviews per product without tripping rate limits. Reviews are returned in the requested locale, so you can collect German reviews from amazon.de and English reviews from amazon.com in separate calls.

Example Response

JSON
{
  "asin": "B0E2X4K9PQ",
  "rating": 4.8,
  "review_count": 1042,
  "rating_distribution": {
    "5": 0.82,
    "4": 0.11,
    "3": 0.04,
    "2": 0.02,
    "1": 0.01
  },
  "reviews": [
    {
      "reviewer": "Alex P.",
      "rating": 5,
      "title": "Best ANC headphones I have owned",
      "body": "The new V2 chip is noticeably better than the XM5. Flights are silent.",
      "verified": true,
      "helpful_votes": 142,
      "variant": "Black",
      "date": "2026-03-08"
    },
    {
      "reviewer": "Morgan L.",
      "rating": 2,
      "title": "Great sound, bad call quality",
      "body": "Music is great, but people say I sound muffled on calls in noisy rooms.",
      "verified": true,
      "helpful_votes": 58,
      "variant": "Silver",
      "date": "2026-02-19"
    }
  ],
  "next_cursor": "ref=cm_cr_arp_d_paging_btm_next_2"
}

Use Cases

  • Voice-of-customer analysis for product teams
  • Feature request mining from negative reviews
  • Comparative review summarization between SKUs
  • Detecting review bombing and authenticity issues
  • Training LLM shopping assistants with real customer language

Why Amazon Reviews Matters

Reviews are the single highest-signal source of product feedback on the internet, but they are buried under Amazon's rendering logic and aggressive bot protection. Scavio returns them as clean, paginated JSON with verified-purchase flags and variant mapping, so product teams can run sentiment, theme, and sku-level drift analysis without building scraping infrastructure. For ecommerce brands, this is the fastest way to find out what customers actually say about your product in their own words.

LangChain Example

Drop amazon reviews data into your LangChain agent in a few lines:

Python
from langchain_scavio import ScavioAmazonReviewsTool
from langchain_anthropic import ChatAnthropic

tool = ScavioAmazonReviewsTool(api_key="your_scavio_api_key")
llm = ChatAnthropic(model="claude-opus-4-6")

reviews = tool.invoke({"asin": "B0E2X4K9PQ", "pages": 3})
negatives = [r for r in reviews["reviews"] if r["rating"] <= 3]

bodies = "\n".join(r["body"] for r in negatives)
themes = llm.invoke(f"Cluster these complaints into themes:\n\n{bodies}")
print(themes.content)

Frequently Asked Questions

Send a search request with the appropriate platform (amazon) and Scavio returns amazon reviews data in the response. See the example above for the exact field path.

Yes. Scavio fetches amazon reviews data in real time on each request. There is no caching layer and no stale data.

Scavio's Amazon reviews endpoint returns paginated reviews for an ASIN, each with the reviewer name, star rating, title, body, date, verified purchase flag, helpful vote count, and

Amazon Reviews data is returned as part of the standard search response. Each request costs 1 credit. Free tier includes 500 credits/month.

Start Using Amazon Reviews

Pull structured Amazon reviews with star ratings, verified purchase flags, helpful votes, and review excerpts.