NexGenData

Posted on Jun 17 • Originally published at thenextgennexus.com

WIPO PATENTSCOPE API: Build a Global Patent Search Pipeline

#api #ip #patents #patentscope

TL;DR. The World Intellectual Property Organization (WIPO) operates PATENTSCOPE, a free public patent search system covering all PCT (Patent Cooperation Treaty) international applications plus more than 90 national patent collections — over 100 million documents. For IP attorneys doing prior-art searches, R&D analysts building patent landscapes, and competitive-intelligence teams tracking competitor filings, PATENTSCOPE is the single best free source for non-US patent data. The web interface is full-featured but the backend exposes a search endpoint that is queryable programmatically. This guide covers the PATENTSCOPE collection structure, the PCT workflow that drives international filings, and a working Python pipeline using the WIPO PATENTSCOPE Search actor.

What PATENTSCOPE covers

PATENTSCOPE contains:

All published PCT international applications — every WO-prefixed application since 1978
90+ national / regional collections — EPO (EP), USPTO (US, in summary form — full data via USPTO PatFT), JPO (JP), KIPO (KR), CNIPA (CN), and many more
Bibliographic data — title, applicant, inventor, IPC and CPC classifications, priority chain, publication date
Full-text searchable abstracts and descriptions — including machine translation across many languages
PDF documents for the published application

The PATENTSCOPE collections page documents which jurisdictions are included and the data freshness for each. WIPO PCT data is updated weekly on Thursday for newly published applications.

The PCT workflow — why it matters for IP intelligence

The Patent Cooperation Treaty lets an applicant file a single international application that preserves filing-date priority across 150+ contracting states. The lifecycle:

Priority filing — often a US provisional or national first filing
PCT international filing — within 12 months of priority. Filing receives a WO publication number.
International publication — 18 months from priority. This is when the application becomes searchable on PATENTSCOPE.
National-phase entry — 30 or 31 months from priority. Applicant chooses which countries to actually prosecute in. This is the most expensive decision in the PCT process and is a strong signal of how seriously the applicant values the invention.

For competitive intelligence the WO publication itself is the early signal — you see what a competitor is filing 18 months after priority but well before national-phase entry. The national-phase entries tell you which jurisdictions they're committing real money to defend.

Why IP attorneys need programmatic access

Prior-art landscape — for a new patent application, search all relevant IPC subclasses globally to identify potentially anticipating disclosures.
Freedom-to-operate — find granted or pending patents in target commercialization countries.
Competitive monitoring — track every PCT publication by a named competitor entity, weekly digest.
R &D trend analysis — aggregate IPC class filings over time to identify emerging fields.
Damages and licensing — patent-family mapping for valuation.

Working Python example

The WIPO PATENTSCOPE Search actor handles pagination, structured-field extraction, and family resolution. Curl:


    curl -X POST "https://api.apify.com/v2/acts/nexgendata~wipo-patentscope-search/run-sync-get-dataset-items?token=$APIFY_TOKEN" \
      -H "Content-Type: application/json" \
      -d '{"query": "EN_AB:(solid state battery) AND IC:H01M", "maxResults": 200, "since": "2024-01-01"}'

Python — competitor PCT monitoring digest:


    import os, json, urllib.request

    APIFY_TOKEN = os.environ["APIFY_TOKEN"]
    ACTOR = "nexgendata~wipo-patentscope-search"

    competitors = ["LG Energy Solution", "CATL", "Panasonic", "Samsung SDI"]
    results = []

    for company in competitors:
        payload = json.dumps({
            "query": f'PA:"{company}"',  # applicant name
            "since": "2026-01-01",
            "maxResults": 200,
            "collections": ["PCT"],
        }).encode("utf-8")
        url = f"https://api.apify.com/v2/acts/{ACTOR}/run-sync-get-dataset-items?token={APIFY_TOKEN}"
        req = urllib.request.Request(url, data=payload, method="POST",
                                      headers={"Content-Type": "application/json"})
        with urllib.request.urlopen(req, timeout=600) as r:
            pubs = json.loads(r.read())
            for p in pubs:
                p["_competitor"] = company
                results.append(p)

    print(f"{len(results)} PCT publications since Jan 2026")
    for r in sorted(results, key=lambda x: x["publicationDate"], reverse=True)[:20]:
        print(f"{r['publicationDate']} | {r['_competitor']:25s} | {r['publicationNumber']}")
        print(f"   {r['title'][:100]}")

Returned shape per publication:


    {
      "publicationNumber": "WO 2026/123456 A1",
      "publicationDate": "2026-05-15",
      "title": "Solid-State Battery with Sulfide Electrolyte Composition",
      "applicants": ["LG Energy Solution, Ltd."],
      "inventors": ["KIM, Jihoon", "LEE, Soyoung"],
      "ipcClasses": ["H01M10/0562", "H01M10/0525"],
      "cpcClasses": ["H01M10/0562"],
      "priorityClaims": [{"date": "2024-11-21", "country": "KR", "number": "10-2024-..."}],
      "abstract": "...",
      "pdfUrl": "https://patentscope.wipo.int/search/en/detail.jsf?docId=WO2026123456"
    }

Comparison: WIPO PATENTSCOPE vs commercial patent databases

Source	Coverage	Family resolution	Cost
WIPO PATENTSCOPE (web)	PCT + 90 national	Yes, via family link	Free
EPO Espacenet	100M+ docs, EPO-curated	Yes, DOCDB / INPADOC	Free
USPTO Patent Public Search	US only, granted + pubs	US continuity only	Free
Derwent Innovation	Global + manually-tagged abstracts	Yes, Derwent families	$3,000–$8,000/seat/yr
PatSnap	Global + AI valuation overlays	Yes	$10,000–$50,000/yr
LexisNexis IP / Questel Orbit	Global	Yes	Enterprise quote
PATENTSCOPE actor (Apify)	PCT + national via PATENTSCOPE backing	Yes, via PATENTSCOPE family link	PPE per result

For a small patent-prosecution shop, $10K–$50K/year for PatSnap is hard to justify. The free PATENTSCOPE backbone with a PPE programmatic layer handles 80–90% of prior-art and competitive-monitoring use cases at a small fraction of that cost. For litigation-grade family analysis, commercial databases still earn their keep — but most of the day-to-day work doesn't need them.

Combining with national feeds

PATENTSCOPE is the right starting point but national patent offices publish richer data. For US-specific work pair with our USPTO Patent Search actor. For China specifically, CNIPA has its own filings surface. The competitive-intelligence pattern is usually PATENTSCOPE for PCT and global early signal → national offices for prosecution-stage detail in the countries you care about.

R&D / corporate use patterns

Weekly competitor PCT digest — every Thursday after the WIPO publication cycle, scan named competitors' new PCT pubs.
Field-trend dashboard — track IPC subclass filings over time (e.g., H01M for batteries, A61K for pharma) to spot emerging research clusters.
Inventor-mobility signal — track named inventors across assignees to spot key R&D staff moves.
National-phase entry tracker — monitor which countries a competitor is committing prosecution budget to.

Search-syntax notes

PATENTSCOPE uses a field-prefix syntax that's worth knowing:

PA:"Acme Corp" — applicant exact name
IN:"Smith, John" — inventor
EN_AB:(keyword) — English abstract keyword
IC:H01M — IPC main group
CTR:JP — country of publication
DP:[2024-01-01 TO 2026-05-01] — date range on publication date

Combining fields with boolean operators (AND, OR, NOT) and parentheses follows the standard Lucene-style syntax. Most prior-art searches end up as something like IC:H01M10/0562 AND DP:[2020-01-01 TO 2026-01-01] AND (EN_AB:sulfide OR EN_AB:argyrodite).

Family resolution and continuity

A single invention typically generates a family of related applications across jurisdictions. PATENTSCOPE's family-view links members via shared priority claims. The actor returns the priority chain on each result — joining across results by priority number reconstructs the family. For freedom-to-operate work you often want the family view: any grant in any target jurisdiction matters; pre-grant publications in other jurisdictions usually don't.

Machine translation caveats

PATENTSCOPE's WIPO Translate covers many language pairs to acceptable but not authoritative quality. For high-stakes use (litigation, freedom-to-operate opinions on foreign-language documents) the machine-translated text is a triage tool, not a final source. Original-language native review by qualified counsel is still the standard. For landscape and trend analytics the machine translation is fine and unlocks the full PATENTSCOPE corpus.

For US-only patent research, see our USPTO actor. For broader corporate intelligence pairing patents with SEC filings, see our SEC 8-K guide and unified SEC event router. For pairing patents with corporate press releases (often the public-facing companion to a patent grant), see our press-release ticker extraction guide.

Get started: Run your first PATENTSCOPE search programmatically at the WIPO PATENTSCOPE Search actor. Bring an IPC class, an applicant name, or a keyword query.

DEV Community