Amazon Product Scraper - Search, Best Sellers

Amazon Product Scraper - Search, Best Sellers

This Amazon product scraper returns Amazon data as a clean spreadsheet or JSON file, with no coding needed. Pick what you want (a mode), enter a keyword, product or category, and press Start. It works on 20 Amazon country sites, and you only pay for results you actually get: if Amazon blocks a page, it costs you nothing.

Other tools make you rent a separate scraper for search, another for products, another for best sellers. This is one tool that does all six.

Quick start

1. Choose a mode: search, product, bestsellers, newreleases, seller or suggestions.
2. Fill the one input that mode uses: search terms, asins, category or sellerIds.
3. Pick the marketplace and set maxItems as a hard cap on cost.
4. Click Start, then export the dataset as JSON, CSV or Excel, or read it through the API.

Search amazon.com for "wireless headphones", first 2 pages:

{ "mode": "search", "search": ["wireless headphones"], "marketplace": "com", "maxPagesPerQuery": 2 }

Full detail for two products:

{ "mode": "product", "asins": ["B09XS7JWHH", "https://www.amazon.co.uk/dp/B08N5WRWNW"] }

Top 100 best-selling electronics in India:

{ "mode": "bestsellers", "category": ["electronics"], "marketplace": "in", "maxItems": 100, "maxPagesPerQuery": 4 }

Every product from a seller's storefront:

{ "mode": "seller", "sellerIds": ["A2L77EE7U53NWQ"], "marketplace": "com", "maxPagesPerQuery": 5 }

Keyword ideas for PPC:

{ "mode": "suggestions", "search": ["wireless head", "yoga"], "marketplace": "com" }

Modes (pick one per run)

Mode	What you get	Give it
Search	Product cards for a keyword: title, price, list price, rating, review count, image, Prime, sponsored flag, ASIN	Search terms
Product detail	Everything for specific products: title, brand, price, rating, review count, availability, Best Sellers Rank, feature bullets, all images, full spec table	ASINs or product URLs
Best sellers	A category's Best Sellers ranking (rank, ASIN, title, price, rating, review count)	Category
New releases	A category's newest products	Category
Seller storefront	Every product a seller lists	Seller IDs
Keyword suggestions	Amazon's autocomplete completions for a prefix, for keyword research and PPC	Search terms

What you get

One row per result in the dataset, tagged with type (search, product, bestsellers, new-release, seller or suggestion) so you can filter mixed exports. Prices come back both as a number (price) and with the currency symbol, ratings as a 0 to 5 number, and review counts as integers.

Filter the dataset view by type for a clean per-mode table, or export the whole run as JSON, CSV or Excel.

What this can do

Pull Amazon's public catalogue data, the things you can see on Amazon without logging in: product listings and prices, ratings and review counts, images, brand, best-seller rank, availability, seller info, keyword ideas and more.

What this cannot do

• It does not download the text of customer reviews. Amazon hides full reviews behind a login, so no tool can reliably scrape review text. You do get each product's star rating and number of reviews, just not the written paragraphs.
• It does not access anything private such as orders, seller dashboards or buyer information. Public pages only.

Reliability

• HTTP-based, with no browser, so it is fast and cheap.
• Every request is isolated, so one product failing never corrupts another's result, which is a common bug in other Amazon actors.
• Amazon occasionally rate-limits or serves a bot check, and the Actor retries on a fresh proxy session automatically. If a page stays blocked after retries it is reported as an error row and, because billing is pay per result, it costs you nothing.
• Residential proxies are the default and give the best success rate. Datacenter IPs get blocked quickly and are not recommended.
• The proxy exit is pinned to the marketplace's country automatically (UK IPs for amazon.co.uk, German IPs for amazon.de, and so on). Amazon serves a JavaScript bot challenge instead of results to out-of-country IPs on some marketplaces, amazon.co.uk in particular, and the Actor detects and retries those pages on a fresh session.

How much does it cost to scrape Amazon?

Pay per event. Current prices:

Event	Free and Bronze	Silver (10% off)	Gold and above (20% off)
Result (result-item), from Search, Best sellers, New releases, Seller or Suggestions	$0.002	$0.0018	$0.0016
Product (product-detail), one full product record	$0.0075	$0.00675	$0.006
Actor Start (apify-actor-start)	$0.01	$0.01	$0.01

The Actor Start event is charged once per run, one event per GB of memory, minimum one event, and carries no Store discount. So 1,000 search rows cost about $2 on Free and Bronze, $1.80 on Silver and $1.60 on Gold and above, and 1,000 full product records cost $7.50, $6.75 and $6.

You are charged only for results actually returned. A run that gets blocked and returns nothing costs only the Actor Start fee, and maxItems is a hard cap on the rest.

Input

• mode: which data type to scrape (see the table). Each mode uses one input field:
  • Search and Keyword suggestions use search (a list of keywords)
  • Product detail uses asins (ASINs or full product URLs, where a URL's marketplace wins)
  • Best sellers and New releases use category (a slug such as electronics, books or toys-and-games, or a full best-sellers URL)
  • Seller storefront uses sellerIds (the value after seller= or me= in a storefront URL)
• marketplace: which Amazon site, 20 in total from com and co.uk to in, co.jp and com.br. A full product URL overrides it.
• maxItems: hard cap on total result rows, for cost control.
• maxPagesPerQuery: how many pages to follow for Search, Seller, Best sellers and New releases. Each page holds roughly 16 to 60 items. Ignored by Product detail and Keyword suggestions.
• maxConcurrency: parallel requests.
• proxyConfiguration: residential proxy by default, strongly recommended because Amazon blocks datacenter IPs. The Actor adds the marketplace's country to a residential Apify proxy configuration unless you set one yourself.

FAQ

Can it scrape Amazon review text?

No, and no tool can do it reliably, because Amazon puts full review bodies behind a login. You do get each product's average star rating and its total review count in every mode that returns products, which covers most rating and reputation analysis.

Which Amazon marketplaces are supported?

Twenty: amazon.com, co.uk, de, fr, it, es, in, ca, com.au, com.br, com.mx, co.jp, nl, se, ae, sg, pl, com.tr, sa and eg. Set marketplace, or paste a full product URL and its marketplace wins.

Do I need an Amazon API key or a seller account?

No. The Actor reads public Amazon pages over plain HTTP, so there is no API key, no login and no seller account. It never touches orders, seller dashboards or buyer information.

How do I find the category slug for best sellers?

Open the Best Sellers page on the marketplace you are targeting and copy the slug out of the URL, for example electronics or toys-and-games. Slugs differ between marketplaces, and you can also paste the full best-sellers URL into category instead.

What happens when Amazon blocks a page?

The Actor retries it on a fresh residential proxy session. If it stays blocked, the run emits an error row instead of a result, and because billing is pay per result that page costs you nothing.

Can I use it for PPC keyword research?

Yes. Run mode: "suggestions" with prefixes in search, for example wireless head, and you get Amazon's own autocomplete completions for that marketplace, which are the phrases shoppers are actually typing.

Why does this Actor run with limited permissions?

Least privilege. It only reads and writes its own run storages, the dataset and key-value store Apify creates for each run, so it never needs access to the rest of your account. It reads only public Amazon catalogue pages.

Notes and limits

• This scrapes public Amazon listing and catalogue data. It does not scrape individual review text, because Amazon login-walls full reviews. Average rating and review count are included, the review bodies are not.
• Price is returned for the vast majority of products. A small number legitimately have no price (pre-orders, "see buying options", currently unavailable items), and Amazon occasionally serves a page variant where the price loads dynamically. Product mode automatically re-fetches to recover it, but price can still be null in those cases. All other fields are unaffected.
• Category slugs vary by marketplace, so copy them from the Best Sellers URL on the marketplace you are targeting.
• Respect Amazon's Terms of Service and applicable laws. You are responsible for how you use scraped data.

More data tools by Automation Craft

• ShopBack Scraper - Cashback Rates, Affiliates
• Google Trends Scraper - Compare and Trending Now
• G2 Reviews Scraper: Ratings, Pros and Cons
• Meta Ad Library Scraper - All Placements, Filters
• Facebook Ads Library Scraper - Page Ads, No Login
• Airbnb Scraper: Prices, Calendar, Reviews
• Website Screenshot Scraper: Full Page, Bulk
• Substack Scraper: Posts, Notes and Profiles