Amazon Product Scraper — Reliable, with Price Intelligence

Scrape any Amazon product page into clean, structured data — and get the answer, not just the dump: a price-anomaly flag, a deal score, and a "what changed since last run" diff. Works across 19 Amazon marketplaces (.com, .co.uk, .de, .in, .ae, .co.jp and more). No login, no cookies.

The difference from every other Amazon scraper: this one never returns a silent blank. Every row tells you what was found, what was missing, and why — and a run-level quality report proves it. Empty cells with no explanation are the #1 reason scrapers get abandoned; this fixes that.

What you get

Product data	Intelligence & reliability
Title, brand, price, list price, discount %	dealScore (0–100 buyer-value verdict)
Rating, review count, star histogram	isPriceAnomaly (price >2σ from its own history)
Best Sellers Rank (+ per-category)	priceVsAvgPercent, priceLow, priceHigh
Variants (color/size + child ASINs)	whatChanged since last run (price/stock/rating/BSR)
Feature bullets, description, image gallery	amazonsChoice, bestSeller badges
Seller, ships-from, FBA/FBM fulfillment	completeness %, fieldsFound, fieldsMissing
Availability, in-stock, currency, marketplace	run-level QUALITY_REPORT in the key-value store

How to use it

1. Paste one or more Amazon product URLs.
2. Set the proxy country to match the marketplace (prices differ by delivery location).
3. Leave Track price history and Detect changes on, and schedule it daily — that's how the anomaly flags and whatChanged build up. Run once for a snapshot; schedule for a monitor.

Input

{
  "productUrls": [
    { "url": "https://www.amazon.com/dp/B09XS7JWHH" },
    { "url": "https://www.amazon.co.uk/dp/B08N5WRWNW" }
  ],
  "proxyCountry": "US",
  "trackPriceHistory": true,
  "detectChanges": true,
  "maxItems": 100
}

Output (one product)

{
  "asin": "B09XS7JWHH",
  "title": "Sony WH-1000XM5 Wireless Headphones",
  "brand": "Sony",
  "price": 328.00,
  "listPrice": 399.99,
  "discountPercent": 18,
  "currency": "USD",
  "priceStatus": "found",
  "rating": 4.4,
  "reviewCount": 12847,
  "starsBreakdown": { "fiveStar": 70, "fourStar": 20, "threeStar": 5, "twoStar": 3, "oneStar": 2 },
  "bestSellersRank": 12,
  "bsr": [{ "rank": 12, "category": "Electronics" }, { "rank": 1, "category": "Over-Ear Headphones" }],
  "amazonsChoice": true,
  "bestSeller": false,
  "variantAxes": ["color_name"],
  "variantCount": 3,
  "variants": [{ "asin": "B09XS7JWHH", "dimensions": ["Black"] }],
  "featureBullets": ["Industry-leading noise cancellation", "30-hour battery life"],
  "imageCount": 7,
  "soldBy": "Amazon.com", "shipsFrom": "Amazon.com", "fulfillment": "FBA",
  "dealScore": 71,
  "priceVsAvgPercent": -12, "priceLow": 298.00, "priceHigh": 399.99, "isPriceAnomaly": true,
  "whatChanged": [{ "field": "price", "from": 379.00, "to": 328.00 }],
  "isNew": false,
  "completeness": 100, "fieldsFound": "title,price,listPrice,rating,reviewCount,brand,availability,image", "fieldsMissing": null,
  "url": "https://www.amazon.com/dp/B09XS7JWHH",
  "scrapedAt": "2026-09-03T09:00:00.000Z",
  "status": "ok"
}

Why the intelligence fields matter

• isPriceAnomaly + priceVsAvgPercent — schedule daily and the actor learns each product's normal price, then flags a genuine drop or spike (>2σ) automatically. Arbitrage sellers and deal sites live on this.
• whatChanged — turns a one-off scrape into a monitor: it reports exactly what moved since the last run (price, stock, rating, review count, BSR). Wire the webhook and you have price-drop alerts.
• dealScore — one 0–100 number combining discount, rating and review volume, so you can rank a list of products by "is this actually a good buy?" without post-processing.
• QUALITY_REPORT (key-value store) — field coverage %, blocks encountered, anomalies, and dataFaked: 0. Honest accounting competitors hide.

Pricing

Pay-per-result: $4.00 / 1,000 products ($0.004 per result), plus a near-zero start fee. You are never charged beyond your maxItems, and blocked/failed pages are reported, not billed as data.

FAQ & limitations

• Do I need cookies/login? No. Product-page data is fetched logged-out. (Full review text is login-walled — use the companion Amazon Reviews actor for that.)
• Why is a price null? The row's priceStatus says why — unavailable_in_marketplace, no_seller_for_this_location, or page_not_a_product. Never a silent blank.
• Do I need to schedule it? Only if you want price history / anomaly / whatChanged. A single run still returns everything else. History needs ≥4 prior runs before isPriceAnomaly becomes non-null.
• Variant prices? Variants list child ASINs + their option values; per-variant price needs each child ASIN as its own URL (add them to productUrls).
• Blocks? Amazon throttles datacenter IPs; if you see captcha blocks in the log, turn on Use residential proxy. Blocks are retried with fresh IPs and reported honestly.
• Which marketplaces? 19, auto-detected from the URL; currency and country are set accordingly.
• Integrations? Output goes to Zapier, Make, n8n, or any webhook via Apify's integrations.

Related actors

• Amazon Reviews (Honest) — review text + rating summary at the scope Amazon serves without login.
• eBay Sold Listings — real sold comps for resale pricing.
• Shopify Product Scraper — the same reliability approach for Shopify stores.
• Portfolio Health Monitor — smoke-tests your scheduled actors and alerts when one degrades.