🛒⭐ Amazon Product Scraper with Review Insights

🛒⭐ Amazon Product Scraper With Review Insights

An Amazon product scraper with real customer review insights built for Apify. Give it Amazon category, search, or product URLs and it returns detailed product data (title, price, brand, images, description, bestseller ranks, offers, seller info, and more) plus real per-review text, star rating, review date, verified-purchase flag, and helpful-vote count for each product — not just aggregate star counts.

This actor is a superset of a general-purpose Amazon product scraper: every input and output field of the base scraper still works exactly the same, with review-text fetching added as a genuinely new capability on top.

---

📦 What It Extracts

🧩 Data Type	💡 Description
Product Name & ASIN	Title and the unique Amazon Standard Identification Number
Price & Stock	Current price, list price, currency, in-stock status
Ratings & Review Counts	Average star rating, star breakdown, total review count
Real Review Text (new)	Per-review title, body text, rating, date, verified-purchase flag, helpful-vote count
Product Details	Features, description, product overview / spec table, A+ content
Category & Breadcrumbs	Product's category path on Amazon
Images	Thumbnail, gallery, and high-resolution image URLs
Seller Information	Seller name, seller URL, business details (when "Scrape sellers" is on)
Offers	Buy-box and marketplace offers with price and condition (when "Maximum scraped offers" is set)
Bestseller Ranks	Category rank(s), e.g. "#12 in Electronics"
Variant Data	Variant ASINs and per-variant prices (when variant options are enabled)

---

⚙️ Key Features

• 🔗 Flexible input — accepts Amazon search URLs, category/browse URLs, or direct product (/dp/) URLs, and mixes multiple Amazon marketplace domains (.com, .de, .co.uk, etc.) in one run.
• ⭐ Real customer review text (this variant's core addition) — fetches actual review title, body, rating, date, verified-purchase status, and helpful-vote count per product, not just the star-count summary.
• 🗂️ A dedicated reviews dataset — every fetched review is also pushed to a separate per-run child dataset (visible under the run's Storage tab), so you can query review rows without unpacking nested JSON.
• 🧭 Search, category, and direct-product modes — scrape an entire category, a keyword search, or a specific product list.
• 💰 Offers & seller data — optionally pull marketplace offers and seller details for the buy box and every offer.
• 🔄 Variant support — collect variant ASINs, per-variant prices, and optionally expand variants into separate result rows.
• 🌍 Multi-marketplace & delivery location — set a proxy country, delivery country/zip code, and language per run; every follow-up request (offers, sellers, variants, reviews, pagination) stays on the SAME Amazon marketplace domain as the URL it came from.
• 🚀 Real-time output — every scraped row is pushed to the dataset as soon as it's collected, so results appear in the Output tab while the run is still in progress.

---

🧭 How to Use

1. Add URLs — paste one or more Amazon category, search, or product URLs into "Enter one or more Amazon category or product URLs".
2. Set limits — choose how many results per URL and (optionally) how many reviews per product.
3. Choose review settings (new) — set "Max reviews per product" and pick "Top (most helpful)" or "Most recent" ordering under the ⭐ Review Insights section.
4. Optional extras — enable seller/offer scraping, variant prices, delivery location, or a proxy country override.
5. Run — click Start. Results (including review text) stream into the dataset in real time.
6. Export — download the dataset as JSON/CSV/Excel, or read it via the Apify API; open the run's Storage tab for the separate reviews child dataset.

---

🧾 Input

Example JSON input:

{
  "categoryOrProductUrls": [
    "https://www.amazon.com/s?k=keyboard",
    "https://www.amazon.com/dp/B08N5WRWNW"
  ],
  "maxItemsPerStartUrl": 10,
  "maxSearchPagesPerStartUrl": 9999,
  "maxConcurrentRequests": 10,
  "proxyCountry": "AUTO_SELECT_PROXY_COUNTRY",
  "maxReviewsPerProduct": 10,
  "reviewSortOrder": "top",
  "proxyConfiguration": {
    "useApifyProxy": true,
    "apifyProxyGroups": ["RESIDENTIAL"]
  }
}

Parameter	Type	Description
categoryOrProductUrls	Array	Required. Search URLs (/s?k=...), category URLs, or product URLs (/dp/ASIN).
maxItemsPerStartUrl	Integer	Max products to scrape per URL.
language	String	Language to request from Amazon (per-domain support varies).
proxyCountry	String	Proxy country, or AUTO_SELECT_PROXY_COUNTRY / NO_PROXY.
maxSearchPagesPerStartUrl	Integer	Max search pages to walk per URL.
maxProductVariantsAsSeparateResults	Integer	Expand this many variants into their own dataset rows.
maxOffers	Integer	Max marketplace offers to scrape per product.
scrapeSellers	Boolean	Fetch seller detail pages for the buy box and offers.
ensureLoadedProductDescriptionFields	Boolean	Retry to make sure seller-editable fields (description, A+ content) are loaded.
useCaptchaSolver	Boolean	Auto-solve Amazon captchas using your own CapSolver key (see below).
capsolverApiKey	String (secret)	New. Your own CapSolver API key - only used when useCaptchaSolver is on. No key is bundled; leave empty to keep captcha solving off.
scrapeProductVariantPrices	Boolean	Fetch a price for each variant ASIN.
scrapeProductDetails	Boolean	Full product-detail mode (default) vs. lightweight search-result-only mode.
maxConcurrentRequests	Integer	Concurrent product-page requests.
countryCode / zipCode	String	Delivery location for localized prices/offers/availability.
locationDeliverableRoutes	Array	Which page types the delivery-location setting applies to.
maxReviewsPerProduct	Integer	New. Real reviews to fetch per product (title, body, rating, date, verified flag, helpful votes). 0 disables review fetching.
reviewSortOrder	String	New. top (most helpful first) or recent (most recent first).
proxyConfiguration	Object	Apify proxy settings (residential recommended).

---

📤 Output

Dataset: one row per product (or per review, in the separate reviews dataset — see below).

Each product row includes (abridged; the full field set matches the table above plus dozens of detail fields such as productOverview, bestsellerRanks, delivery, returnPolicy, variantAsins):

Field	Type	Description
asin	String	Product ASIN
title	String	Product name
url	String	Product page URL
price	Object	{ "value", "currency" }
listPrice	Object	List/crossed-out price
inStock	Boolean	Availability
brand	String	Brand name
stars	Float	Average rating
reviewsCount	Integer	Total review count (aggregate, from the product page)
productPageReviews	Array	New - real reviews: { reviewId, title, body, rating, author, date, isVerifiedPurchase, helpfulVotes, reviewUrl }
breadCrumbs	String	Category path
thumbnailImage / galleryThumbnails	String / Array	Image URLs
seller	Object	Seller id, url, name
offers	Array	Marketplace offers (when maxOffers > 0)
bestsellerRanks	Array	{ "rank", "category" }

Reviews child dataset: when maxReviewsPerProduct is above 0, every fetched review is ALSO pushed as its own row to a separate per-run dataset (named <runId>-reviews, visible under the run's Storage tab), tagged type: "review", isChild: true, with parentAsin, productUrl, and productTitle linking it back to its product.

---

❓ FAQ

1. What's different from a plain Amazon product scraper?

This actor adds real, per-review customer text (title, body, rating, date, verified-purchase flag, helpful-vote count) on top of standard product data, plus a dedicated reviews dataset - while keeping full compatibility with a standard Amazon product-scraper input/output shape.

2. Can I extract reviews and ratings?

✅ Yes. Star ratings and review counts come from the product page as before, and this actor now fetches the actual review text too (title + body), along with rating, date, verified-purchase status, and helpful-vote count per review.

3. How many reviews can I get per product?

Set maxReviewsPerProduct (default 10, max 200). Reviews are fetched in pages of about 10 from Amazon's own review listing, in either "most helpful" or "most recent" order.

4. Does it work on non-US Amazon domains?

Yes - pass a URL on the domain you want (e.g. amazon.de, amazon.co.uk). Every follow-up request for that item (offers, sellers, variants, reviews, pagination) stays on the same marketplace domain as the URL it came from.

5. Can I scrape just search results without full product details?

Yes - set scrapeProductDetails to false for a faster, lightweight mode that returns search-result-level data only (no review fetching in this mode).

6. Is it safe to scrape Amazon?

Scrape only publicly available data (prices, titles, ratings, reviews) and use it for research, analytics, or price comparison in line with Amazon's terms.

7. Do I need a CapSolver account?

Only if you enable useCaptchaSolver. Provide your own capsolverApiKey - the actor does not include or use any bundled key.