artsper-scraper

Scrapes both artworks and artist profiles from Artsper.com, including per-artist rollup stats (average price, medium mix, sold count, latest artwork) computed from the artist's full catalog — not a sample.

Contents

• Key features
• Output
• Input
• Input examples
• Incremental (delta) mode
• How to scrape Artsper.com
• You might also like
• FAQ

🔑 Key features

• Two entity types, one Actor. Set entityType to artworks or artists — no need for two separate scrapers to build a full picture of an artist and their work.
• Real per-artist rollup stats, not a sample. Average price, unique medium mix, sold-artwork count, and latest artwork are computed from an artist's entire catalog, enumerated via Artsper's own catalog pagination (paginated to completion), not just whatever a single search page happens to show.
• Delta mode built in. Every run classifies each item as new, changed, unchanged, or delisted against a persisted baseline — pay for what changed, not a full re-scrape every time.
• Exhaustive when you ask for it. With no startUrls/artistUrls, discovery is the full site sitemap — every artwork, every artist — but maxItems defaults to 50 (a fast preview, and what keeps an unconfigured run within Apify's automated 5-minute QA check), so clear it (null) for a real full crawl.
• Rich data straight from the site's own structured markup. Every artwork/artist page ships clean Schema.org JSON-LD (price, medium, dimensions, artist nationality, birth year, full biography) — read directly rather than scraped from display text, so it doesn't break on cosmetic page redesigns the way plain-HTML scraping would.
• A verified, trackable artist list. artistUrls lets you scrape exactly the artists you care about — each URL is checked against the real artist sitemap first (typos/removed artists are logged and skipped), and unlike a generic startUrls list, it gets its own persisted tracking scope: run the same list on a schedule and get real new/changed/delisted detection for just those artists.
• Progress visible in the run log. Every ~10 items or ~20 seconds, the log reports how many items have been processed (with a known total for bounded runs), pushed, and errored, plus the observed rate — and for entityType: artists, a running count of artworks fetched across all rollups so far.
• Confirmed-gone URLs are never re-fetched. Artsper.com doesn't return an HTTP error for a removed artwork — it silently serves the generic catalog page instead (still 200 OK). This Actor detects that content signal, persists it, and skips the URL on every future run — no input needed to enable it.
• Resilient to a real, confirmed rate limit — bounded, not stuck. Artsper.com's Cloudflare protection sometimes throttles a large artist rollup (occasional 429s that mostly resolve on retry, just slowly). Rather than one artist's catalog silently stalling a run for a long time, a wall-clock budget cuts a rollup short after 180 seconds and still returns a real, useful partial result (flagged via rollup_rate_limited in the run log), and a separate failure-count breaker catches a harder, sustained block the same way.

📋 Output

One dataset item per artwork or artist, depending on entityType — see .actor/dataset_schema.json for the full field list, or the Output tab's Artworks / Artists views for a readable table.

Example artwork record:

{
  "source": "artsper",
  "entity_type": "artworks",
  "external_id": "artsper_2462506",
  "url": "https://www.artsper.com/fr/oeuvres-d-art-contemporain/peinture/2462506/toile-lamour-dans-la-nature-jerome-mesnager",
  "title": "Toile L'amour dans la nature",
  "artist_name": "Jérôme Mesnager",
  "artist_url": "https://www.artsper.com/us/contemporary-artists/france/17/jerome-mesnager",
  "category": "Painting",
  "medium": "acrylic",
  "surface": "canvas",
  "date_created": "2022",
  "price": 2800,
  "currency": "EUR",
  "availability_status": "InStock",
  "width_cm": 81,
  "height_cm": 100,
  "condition": "Artwork sold in perfect condition, framed, ready to hang",
  "signature": "Hand-signed by artist",
  "gallery_name": "At Down",
  "gallery_country": "France",
  "change_type": "new"
}

Example artist record:

{
  "source": "artsper",
  "entity_type": "artists",
  "external_id": "artsper_artist_17",
  "url": "https://www.artsper.com/fr/artistes-contemporains/france/17/jerome-mesnager",
  "name": "Jérôme Mesnager",
  "nationality": "France",
  "birth_year": 1961,
  "total_artworks": 45,
  "sold_artworks": 0,
  "avg_price": 1037.0,
  "mediums": "Fine Art Drawings, Painting, Print, Sculpture",
  "change_type": "new"
}

Results can be downloaded as JSON, CSV, or Excel from the Console's Output tab, or pulled via the Apify API/dataset endpoint.

Known gaps:

• price/currency are native, unconverted values, and the artist rollup's avg_price/ sold_avg_price/latest_artwork_price assume a single currency across an artist's whole catalog (the common case) — mixed currencies aren't detected or converted.
• latest_artwork_* assumes the catalog's default sort is newest-first; not independently confirmed against a date field.
• A large artist's rollup can be cut short by Cloudflare rate-limiting (see Key features) — when that happens, total_artworks/avg_price/etc. reflect only the artworks fetched before the 180-second budget ran out, not the artist's full catalog. Check the run log for a "cut short by sustained rate-limiting" warning if a rollup's numbers look lower than expected.

⚙️ Input

See .actor/input_schema.json for the full JSON schema.

Parameter	Type	Default	Description
entityType	String	artworks	artworks or artists.
startUrls	Array of strings	(none)	Specific artist/artwork URLs, or a category/search catalog URL, to scrape directly instead of discovering via the full site sitemap. Scope is always "custom", with no persisted tracking across runs — for persisted per-artist tracking, use artistUrls instead. A catalog URL's unfiltered form self-caps at 100 pages regardless of true match count — narrow with a category/filter for large result sets.
artistUrls	Array of strings	(none)	Artists only. Paste artist URLs, one per line (a .txt file's contents paste in directly, no upload needed) to scrape exactly those artists + their artworks. Verified against the real artist sitemap first (see verifyArtistUrls); gets its own persisted scope, so re-running the same list on a schedule gives real new/changed/delisted tracking for just those artists. Takes priority over startUrls.
verifyArtistUrls	Boolean	true	When artistUrls is set, check each URL against the real sitemap first (roughly a minute, one full sitemap walk, regardless of list size) and skip any not found. Turn off to scrape the list unverified.
maxItems	Integer	50	Stop after pushing this many dataset items. Defaults to a fast, cheap preview (also what keeps an unconfigured run within Apify's automated 5-minute QA check). A full crawl covers the entire site — raise this or clear it (set to null) for that; note a capped run never updates the incremental baseline (see below), so any real tracking run needs this cleared.
mode	String	auto	auto (recommended): full scan on the first run, incremental after. full: always push every item and refresh the baseline. incremental: always push only new/changed items. Only a plain, unscoped sitemap crawl or a verified artistUrls list can detect delistings or update the baseline.
impersonate	String	chrome (internal)	curl_cffi TLS-impersonation target. Artsper's Cloudflare protection challenges plain requests, so this is set internally by default — override only if that stops working.
proxyConfiguration	Object	{"useApifyProxy": false}	Apify Proxy config. Off by default — the sitemap/profile/detail pages this Actor relies on worked fine without one while building it.

🧪 Input examples

Full artwork catalog scan (exhaustive, slow — every artwork's detail page is fetched; maxItems: null explicitly overrides the 50-item default):

{ "entityType": "artworks", "maxItems": null }

Bounded preview (the default — fast, but not a curated subset, just the first N artworks the sitemap yields):

{ "entityType": "artworks", "maxItems": 200 }

Full artist directory, with rollup stats:

{ "entityType": "artists", "maxItems": null }

A specific, tracked list of artists (verified against the sitemap, own persisted scope — maxItems: null so the whole list is scraped even if it's longer than the 50-item default):

{
  "entityType": "artists",
  "artistUrls": [
    "https://www.artsper.com/fr/artistes-contemporains/france/17/jerome-mesnager"
  ],
  "maxItems": null
}

Scheduled tracking run — full, uncapped run (maxItems cleared — required for the baseline to save and delistings to be detected):

{ "entityType": "artworks", "mode": "incremental", "maxItems": null }

🔄 Incremental (delta) mode

Every run classifies each item as new, changed (price or availability moved, for artworks; catalog size/sold-count/avg-price moved, for artists), unchanged, or delisted, using a state baseline persisted in a named Apify Key-Value Store scoped to entityType.

• mode: auto (default) — first run for a scope pushes everything (full); later runs push only new/changed/delisted (incremental).
• A startUrls-scoped run is always partial and never updates the baseline. artistUrls is different: it's treated as a complete, closed universe (like a full sitemap crawl), so its baseline does get saved and delisted-detection does run — as long as no item hit a genuinely ambiguous error that run (a confirmed-gone URL doesn't count against this — see below; a timeout, 5xx, or parse error does, and skips the baseline-save/delisted-detection that run to avoid false "delisted" reports).
• Every bounded run (startUrls or artistUrls) logs a quick "X new, Y already tracked, Z already known-dead" preview right after startup, before the full fetch loop begins — new/already-tracked is informational only (every URL is still fully fetched and classified regardless, real price/status-change detection isn't skipped for "already tracked" items); already-known-dead URLs genuinely are skipped, see below.
• A URL confirmed gone is recorded and never fetched again on any future run, regardless of mode or scope — logged distinctly from other errors. Doesn't count toward the baseline-save/ delisted-detection gate above, since it's a trustworthy, already-understood outcome rather than an ambiguous failure.

🚀 How to scrape Artsper.com

1. Open the Artsper Artist & Artwork Scraper in Apify Console and go to the Input tab.
2. Pick entityType (artworks or artists).
3. maxItems defaults to 50 (a quick preview) — clear it (set to null) for a full, uncapped run.
4. Click Start.
5. When the run finishes, browse results in the Output tab, or download as JSON/CSV/Excel, or fetch them via the API.
6. To track over time instead of scraping once: create a Schedule with mode: auto — the first run does a full scan, every run after only bills what actually changed.

🔗 You might also like

• SaatchiArt Artist & Artwork Scraper — the same entity-type/rollup-stats design pointed at a second, independent art marketplace, for expanding artist coverage or cross-referencing the same artist's presence and pricing across platforms.
• Artmajeur Artist & Artwork Scraper — a third, independent art marketplace source, same design.
• Artfinder Artist & Artwork Scraper — a fourth, independent art marketplace source, same design.
• Artsy Artist, Artwork & Gallery Scraper — a fifth, independent art marketplace source, plus a genuine dealer/gallery angle (real represented-artist rollups per gallery) this collection doesn't otherwise have.
• Bluethumb Artist & Artwork Scraper — a sixth, independent art marketplace source (Australia's largest), with real per-artist sold/available/total counts read from a single profile-page fetch, no rollup fan-out needed.
• Riseart Artist & Artwork Scraper — a seventh, independent art marketplace source, built entirely on the site's own public GraphQL search API (no HTML scraping, no per-item detail fetch), with a real per-artist rollup (total artworks, avg price, for-sale vs. enquire-only split).
• Kunsthaus ARTES Artist & Artwork Scraper — an eighth, independent art marketplace source (German gallery), with a real four-state availability signal (for sale, price-on-request, reserved, sold) read from the site's own schema.org microdata.
• Ars Mundi Artist & Artwork Scraper — a ninth, independent art marketplace source (a sibling German gallery running the same underlying platform as Kunsthaus ARTES), also selling jewelry and decorative objects alongside paintings/sculptures.
• Artprice.com Artist & Artwork Scraper — a tenth, independent art marketplace source, built on Artprice.com's own public unauthenticated JSON APIs (artist directory + current marketplace listings), with real per-artist rollup stats and multi-currency pricing (EUR/USD/GBP/CNY/BTC/ETH) on every artwork.

❓ FAQ

Is it legal to scrape Artsper.com? It's legal to collect publicly available marketplace data such as artwork listings, prices, and public artist profile information. Scrape it only with a legitimate purpose under GDPR.

How do I get only new/changed items? Use mode: auto (or incremental) on a schedule — see Incremental mode.

Does this Actor detect sold artworks? Yes — sold_artworks/sold_avg_price on artist records are computed from real availability_status values read off each artwork's own page, confirmed live (a real artist sampled came back with sold_artworks: 13 out of 54 fetched). If a large artist's rollup gets cut short by rate-limiting before reaching every artwork, these numbers reflect only what was fetched — see the rate-limiting note under Known gaps.

Search keywords

artsper scraper, art marketplace scraper, artist data scraper, artwork data scraper, art price data, art market analytics, art collector data feed