Review & ratings analytics API

A review and ratings analytics API extracts customer feedback from marketplace listings so you can measure sentiment and run voice-of-customer analysis at scale. ShopAPIS powers it by returning rating, review_count, and a reviews[] array — each entry carrying star score, title, body, date and a verified_purchase flag — across 70+ marketplaces in 30+ countries. You send a product URL or identifier; ShopAPIS returns the aggregate scores plus the underlying review text to feed your NLP pipeline.

Voice-of-customer is only as trustworthy as the reviews behind it — which is why ShopAPIS flags verified_purchase on every review, so your sentiment model can weight or filter out unverified noise.

Fields that power review analytics

These belong to the same 40+ field object on the e-commerce data API pillar, so reviews arrive alongside price, seller and attribute context.

Why an API beats DIY

Review data is paginated, lazy-loaded, and locale-specific — collecting it at scale means clicking through dozens of pages per listing while staying ahead of anti-bot defenses — and with Akamai measuring bots at about 42% of web traffic, nearly two-thirds malicious , those defenses push block rates high. Worse, raw scraped reviews mix verified and unverified buyers, so a sentiment model trained on them learns from noise. ShopAPIS paginates the full review set, normalizes scores and dates across marketplaces, and tags verified purchases, so your data scientists analyze clean voice-of-customer signal instead of building crawlers. The collection mechanics are on the marketplace scraping API pillar.

Sample request and response

GET /v1/reviews?marketplace=amazon&country=US&id=B0CHX1W1XY&limit=3
Authorization: Bearer YOUR_API_KEY

{
  "marketplace": "amazon",
  "country": "US",
  "product_id": "B0CHX1W1XY",
  "rating": 4.7,
  "review_count": 128344,
  "rating_breakdown": { "5": 0.78, "4": 0.13, "3": 0.05, "2": 0.02, "1": 0.02 },
  "reviews": [
    { "score": 5, "title": "Great little speaker", "body": "Sound is fuller than the last gen...",
      "author": "J. Marsh", "date": "2026-05-28", "verified_purchase": true },
    { "score": 2, "title": "Wi-Fi drops", "body": "Keeps losing connection in another room...",
      "author": "anon", "date": "2026-05-21", "verified_purchase": false },
    { "score": 4, "title": "Good value", "body": "Does what it says, setup took two minutes...",
      "author": "R. Patel", "date": "2026-05-19", "verified_purchase": true }
  ],
  "fetched_at": "2026-06-05T08:30:00Z"
}

Use cases

• Voice-of-customer analysis — mine reviews[] for recurring complaints and feature requests.
• Sentiment tracking — trend rating and verified sentiment over time and against rivals.
• Product development — surface the defects driving 1★ and 2★ reviews.
• Listing health — pair with catalog enrichment to fix what reviewers flag.

Marketplaces commonly analyzed

Related solutions