Real-Time Amazon Product Data

Live Amazon search, product details, reviews, offers, bestsellers, and seller data — pulled fresh on every request through geo-targeted residential proxies, returned as clean JSON.

Real-Time Amazon Product Data scrapes Amazon storefronts on demand and parses the messy HTML into flat, predictable JSON objects — no Amazon account, PA-API keys, or HTML wrangling required. Every endpoint accepts a country code so you can query the US, UK, German, French, and other marketplaces from a single API. Requests are routed through rotating residential proxies with per-marketplace geo-targeting, so you get the same listings, prices, and rankings a real shopper in that country would see.

Endpoints

Endpoint	What it returns
`GET /search?q=&country=US&page=1`	Search results array. Each item: `asin`, `title`, `price`, `rating`, `rating_count`, `thumbnail_url`, `url`, `is_sponsored`, `prime`. Plus `next_page_token` for pagination.
`GET /product?asin=&country=US`	Full product detail: `asin`, `title`, `description`, `price`, `currency`, `availability`, `rating`, `rating_count`, `images[]`, `features[]`, `category`, `best_seller_rank`, `brand`.
`GET /product/reviews?asin=&country=US&sort=recent&page=1`	Customer reviews array (`sort` = `recent` or `helpful`). Each review: `review_id`, `author`, `rating`, `title`, `body`, `date`, `verified_purchase`, `helpful_votes`. Plus `next_page_token`.
`GET /product/reviews/top?asin=&country=US`	The most-helpful reviews for a product (same review fields as above).
`GET /product/offers?asin=&country=US`	All seller offers for an ASIN: `price`, `seller_name`, `condition`, `shipping`.
`GET /category?id=&country=US&page=1`	Products within an Amazon category / search index (same item shape as `/search`).
`GET /bestsellers?category=&country=US`	Ranked bestseller list for a category slug (e.g. `electronics`, `books`). Each item: `rank`, `asin`, `title`, `price`, `rating`, `thumbnail_url`.
`GET /seller?id=&country=US`	Seller storefront info: `name`, `rating`, `rating_count`, `about`.
`GET /scrape?url=`	Pass any Amazon product / search / reviews / offers / bestsellers / seller URL; the page type is auto-detected and parsed into the matching shape above.
`GET /health`	Liveness check, returns `{"status":"ok"}`.

Supported country codes: US, UK, DE, FR, CA, JP, AU, IT, ES (each maps to the native Amazon storefront, e.g. UK → amazon.co.uk, DE → amazon.de). Default is US.

Why this API

Residential-proxy reliability. Requests exit from rotating residential IPs geo-matched to the marketplace you're querying, so Amazon serves the real localized storefront instead of anti-bot 503s — far more consistent than datacenter scraping.
Sponsored ads are correctly flagged, not mislabeled. Search results carry an honest is_sponsored boolean and the real product title, instead of the "Sponsored" badge text leaking into the title field. Filter ads out yourself, or keep them.
Canonical product URLs. Every search/result url is built deterministically from the ASIN (/dp/<asin>), so you never get stuck with throwaway ad-redirect links.
Multi-marketplace from one call. A single country parameter switches between nine Amazon storefronts with the correct TLD and Accept-Language for each.
Clean, flat JSON with safe nulls. Parsing is fully defensive — a missing field comes back as null rather than throwing, so your integration never breaks on an odd listing. Prices and ratings are normalized (e.g. abbreviated review counts like 33.4K become integers).

Typical use cases

Price tracking and repricing tools that need live Amazon prices and per-marketplace availability.
Product research and market-intelligence dashboards (ratings, review counts, bestseller rank, brand).
Review mining and sentiment analysis over recent or most-helpful customer reviews.
Competitor and offer monitoring across third-party sellers for a given ASIN.
Catalog and lead enrichment — turn an ASIN or Amazon URL into structured product data.

Good to know

Fresh on every request, then briefly cached. Responses are served live and cached server-side for a short window (search ~1h, products ~6h, reviews/offers ~30–60m, categories/bestsellers ~2h) to keep things fast. Data reflects Amazon at fetch time.
Transient empties self-heal. Amazon occasionally returns a thin page; the API never caches an empty result, so simply retrying a call that came back with zero items will typically return full data on the next attempt.
rating_count can be null. Products with no reviews yet legitimately have no count — treat rating_count (and rating) as optional. Other fields like price, availability, or description may also be null when Amazon doesn't show them on the page.
Anti-bot pages are handled, not leaked. If Amazon serves a CAPTCHA/robot-check, the API returns a 502 upstream unavailable rather than garbage — retry the request. A missing product returns 404, and an unsupported country returns 422.
Best for live lookups, not bulk dumps. This is a real-time scraping API; design around on-demand calls (optionally with your own queue) rather than crawling tens of thousands of pages in a tight loop.
is_sponsored is provided so you decide. Results include sponsored placements flagged with is_sponsored: true; filter them out when you want organic-only rankings.