Shopify Products Scraper

Scrape the entire product catalog of any Shopify store — no API key, no login, no anti-bot. Every Shopify storefront exposes a public /products.json endpoint; this actor paginates it and returns one clean, flat row per product.

Point it at a competitor, a brand you resell, or your own store, and get structured data you can drop straight into a spreadsheet or feed.

What you get per product

store, productId, title, handle, url (canonical /products/{handle} link), vendor, productType, tags[], description (HTML stripped to plain text), createdAt, publishedAt, priceMin, priceMax, available (any variant in stock), variantCount, variants[] (title, sku, price, compareAtPrice, available), and images[] (URLs).

Input

Field	Notes
storeUrls	One or more stores. Bare domain (shop.com), full URL, or any store page (/collections/x) — each is normalized to the origin.
maxItems	Max products per store (default 1000). Paged 250 at a time until the cap or the catalog ends.
proxyConfiguration	Optional, off by default. The endpoint is public with no anti-bot, so no proxy is needed. Only enable one if you hit IP rate limits on very large catalogs.

Output

One dataset row per product, deduped by productId within each store. Each charged product is ok: true.

If a URL is not a Shopify store (no /products.json, returns HTML or 404), the actor emits a single clean diagnostic row (errorCode: "NOT_SHOPIFY") for that store and moves on — it never crashes, and it never charges for it. Empty catalogs return NO_RESULTS; bad input returns BAD_INPUT.

Example

{ "storeUrls": ["https://www.gymshark.com", "https://www.allbirds.com"], "maxItems": 500 }

Pricing

$1.00 per 1,000 products ($0.001 each), with no run-start fee. Flat rate — no volume tiers, no plan gates — and you are charged only for products actually returned.

You are charged once per product row (ok: true). Diagnostic rows (NOT_SHOPIFY, NO_RESULTS, BAD_INPUT, network/block errors) are ok: false and are never charged, so pointing the actor at a list of domains where half turn out not to be Shopify only bills you for the half that are. With no run-start fee, a daily catalog refresh of one store costs the same per product as a single bulk sweep of fifty.

Notes

• Shopify caps /products.json at 250 products per page; the actor handles pagination automatically.
• Only published products appear in /products.json — unpublished/draft products are not exposed by Shopify.
• inventory_quantity is intentionally not in the public feed; per-variant stock is reported as the boolean available.
• Some stores password-protect or disable the JSON endpoint; those come back as NOT_SHOPIFY rather than failing the run.
• Most fields mirror Shopify's feed directly, so a few can be null when the store omits them: title, handle, url, vendor, productType, and priceMin/priceMax (null when a product has no priced variants). tags, variants, and images are always arrays (possibly empty).