Amazon Search & Product Scraper

Amazon Search & Product Scraper (Python + Playwright)

This Actor searches Amazon for given keywords and extracts product list data that is ready for analysis or export.

Typical use cases:

• Product research for marketplaces (e.g. iphone 17 case, usb c hub, gaming chair)
• Competitor and price monitoring across multiple Amazon marketplaces
• Feeding product lists into further analytics, dashboards, or LLM pipelines

Input

The input is defined in .actor/input_schema.json and exposed in Apify Console:

• keywords (array of strings, required)
  List of search keywords, e.g. ["iphone 17 case", "usb c hub"].
• max_items_per_keyword (integer, default 50)
  Maximum number of products to scrape for each keyword.
• max_pages (integer, default 3)
  Maximum number of result pages to crawl for each keyword (1–20).
• country (string, default "US")
  Amazon marketplace to target: one of US, UK, DE, FR, JP.
• min_rating (number, default 0)
  If > 0, products with rating lower than this value are filtered out.
• min_reviews (integer, default 0)
  If > 0, products with fewer reviews than this value are filtered out.
• exclude_sponsored (boolean, default false)
  If true, sponsored products are excluded from the results.
• fetch_details (boolean, default false)
  If true, the Actor will open product detail pages for the first max_detail_items results per keyword to enrich data (e.g. category path).
• max_detail_items (integer, default 5)
  Maximum number of products per keyword for which to open detail pages when fetch_details is enabled.

Output

Results are pushed to the default dataset. Each item contains (non‑exhaustive):

• keyword, country, pageIndex
• asin, title, brand, productUrl, imageUrl
• price, priceText, originalPriceText, currency
• rating, reviewsCount, isPrime, isSponsored, badges
• Optional when fetch_details=true: categoryPath, featureBullets

The dataset view is configured in .actor/dataset_schema.json so that the Overview table shows the most important fields directly in Apify Console.

Running locally

From the project root:

$apify run

The default local input is stored in storage/key_value_stores/default/INPUT.json.
You can edit it to test different keywords, marketplaces and filters.

Anti‑bot considerations

Amazon employs strong anti‑bot protections. This Actor includes:

• Realistic desktop browser profile (user agent, viewport, locale)
• Short navigation timeouts with retries and basic backoff
• Simple detection of common CAPTCHA / bot‑check pages

For production runs on Apify, it is strongly recommended to:

• Enable Apify Proxy and use residential or high‑quality datacenter IPs
• Keep concurrency reasonable (e.g. 1–3 browser contexts) to avoid rate limits

Future extensions

Planned improvements before public release:

• Additional fields (brand, badges, delivery info) when consistently available
• More robust selectors & fallbacks for Amazon UI changes
• Optional export helpers (e.g. sorting, filtering presets for typical workflows)