![Allegro Scraper [Only $0.0014💰] | Prices | Sellers | GTIN avatar](https://images.apifyusercontent.com/ONDp4FXo1-HSiFD5G7Cbn6FDPVUS7vk49AMFAvFwwb0/rs:fill:250:250/cb:1/aHR0cHM6Ly9hcGlmeS1pbWFnZS11cGxvYWRzLXByb2QuczMudXMtZWFzdC0xLmFtYXpvbmF3cy5jb20vZHRSN3FuVGJRck5pU0NtajctYWN0b3ItbzVzSGE4V2N0R2RiNklVT20tc0NMZ25kaEk4ei1pbWFnZXNfJTI4MyUyOS5wbmc.webp){kind=avatar}
Allegro Scraper [Only $0.0014💰] | Prices | Sellers | GTIN
Pricing
from $1.40 / 1,000 results
Scrape Allegro.pl product data by keyword, category, seller or offer URL — pure HTTP, no browser. Each row: price, seller reputation, condition, parameters, Smart!/sponsored flags and sales velocity. Deep mode adds stock, GTIN and warranty.
Keywords to search on Allegro.pl. Each keyword runs its own paginated crawl, e.g. "laptop gaming", "iphone 15", "lego technic".
[ "laptop gaming", "iphone 15"]Direct Allegro URLs — offer pages (allegro.pl/oferta/…), search/category pages (allegro.pl/listing?string=… or allegro.pl/kategoria/…) or seller storefronts (allegro.pl/uzytkownik/…). Listing/category/seller URLs are paginated; offer URLs return a single product (always full detail). A p=N in the URL is respected and the crawl continues forward from that page — handy for splitting a huge storefront into page batches. Runs in addition to the search queries.
Upper bound of products to extract per search query or per listing/category/seller URL. Allegro serves ~60 products per page. Search and category views are capped by Allegro at 100 pages (~6000 products per query); seller storefronts have no such cap and are followed to their real last page.
Optional hard cap on the number of listing pages fetched per query/URL, independent of the item limit. Defaults to the number of pages needed to reach Max items per query (capped at 100 for search/category by Allegro).
Only return offers priced at or above this value in PLN. Maps to Allegro's price_from filter (applied to search and category URLs).
Only return offers priced at or below this value in PLN. Maps to Allegro's price_to filter (applied to search and category URLs).
Filter by product condition (Allegro's "stan" filter).
Order of the search/category results.
Visit each offer page for the deep dataset: full description, complete parameter table, rating + review count + score distribution, stock quantity, GTIN/EAN, warranty, return policy and full-resolution images. Costs one extra request per product (slower). Leave off for fast, rich listing-level data.
Attach the raw Allegro offer payloads (price / seller / rating / parameters boxes) to each detailed record under detail.raw. Only used when "Scrape full product details" is on. For debugging / power users.
When a search query is a pure EAN/GTIN barcode (8–14 digits) and "Max items per query" is 1, skip the extra per-offer fetch and echo the searched barcode into the record's gtin field. The listing data still gives price, seller, rating, images and parameters — making big barcode runs dramatically faster and cheaper. Turn OFF if you need descriptions, stock, warranty or return policy on barcode lookups. Only affects pure-EAN single-item queries; keyword and multi-item searches always fetch full details when deep mode is on.
Country the proxy request exits from (Allegro localises prices and availability by IP). Default "pl" (Poland) is correct for allegro.pl.
How many products/pages to fetch in parallel (1–50). UNBLOCKER re-renders each page in a real browser (~55–65 s), so parallelism is the main speed lever and it also cuts compute cost (cost ≈ allocated memory × wall-clock). Default 25; use 50 for large/deep runs.
Experimental speed boost: for each query, a patchright (engine-level undetectable) Chromium solves DataDome ONCE on a sticky Apify proxy session (same IP), then all pages of the query are fetched in parallel via curl-impersonate with the matching cookie — ~1–2 s/page instead of a fresh ~55–65 s UNBLOCKER render per page, no per-page solver fees. Put a STICKY group FIRST in apifyProxyGroups (e.g. ["RESIDENTIAL"]); if you also list UNBLOCKER, any page whose warm fetch fails automatically falls back to UNBLOCKER, so the run still completes reliably (note: pooled residential IPs are a lottery — some are pre-flagged, so solving may take a few attempts). If solving fails for every attempted IP, the whole query falls back to the per-page managed path. Recommended: "apifyProxyGroups": ["RESIDENTIAL", "UNBLOCKER"] for speed + reliability.
How many patched browsers may solve DataDome in parallel (1-6). Each browser costs ~300-500MB RAM; the 2-core container CPU-thrashes past ~3, slowing solves. Keep 3 unless the container has more cores. Many queries = many solves (DataDome re-challenges each new query string), so the efficient shape is FEW queries × DEEP pagination (one solve serves ~5+ pages at ~5 items/s).
Select proxies used by the scraper. You can list MULTIPLE groups (first one wins for routing):
warmup: false.super=true routing).Optional scrape.do managed-proxy API key. When set, every request is routed through https://api.scrape.do?url=…&geoCode=pl&super=true&render=false&sessionId=… (the official Allegro integration combo), which routes through Polish residential/mobile IPs and clears the DataDome challenge. Requires a scrape.do account with premium (super) routing enabled.