Patent Intelligence (PATENTINT): Intelligence Discipline Guide

A patent is a bargain: an inventor gets a monopoly, and the public gets the working details. That trade makes the patent record the most technically candid public disclosure any organisation makes.

What Patent Intelligence is as a discipline

Patent intelligence is the analysis of patent and published application data to determine what an organisation is building, when it started, where it intends to sell and what it fears competitors will do. It uses bibliographic structure as much as text: priority dates fix invention timing, family members reveal target markets, classification codes place the work in a technical taxonomy, assignment records track ownership movement, and examiner citations map the prior art landscape. The result is a technology and intent picture that is dated, jurisdictionally specific and legally attested.

Sub-methods include landscape and white-space analysis, competitor filing trend tracking, inventor mobility analysis, family and priority mapping, citation and litigation analysis, and freedom-to-operate screening. In the cycle it acts as a mid-term warning source: filings sit roughly eighteen months behind invention because of the publication delay, but two to five years ahead of product release, which is the useful gap.

Why it matters

Only patent intelligence dates an invention with legal precision and shows which markets an organisation considered worth the cost of protection. It answers whether a claimed capability is real enough to have been reduced to practice, which engineers actually did the work and where they went afterwards, whether a technology space is crowded or open, and who holds blocking rights. Assignment and security interest records also expose financial distress and quiet ownership transfers that never appear in corporate filings.

What analysts actually look for

These are the concrete, observable signals that carry weight in this area of work:

Priority dates that fix when an organisation first claimed an invention, independent of when it announced anything publicly
Patent family geography showing which national markets an applicant paid to protect and therefore intends to enter
Classification code concentration revealing the specific technical subfields where filing effort is being spent
Named inventors and their filing histories, exposing team composition, seniority and movement between employers
Assignment and reassignment records showing ownership transfers, acquisitions and collateral pledges over patent portfolios
Examiner and applicant citations mapping which prior art and which competitors define the technical landscape
Continuation, divisional and accelerated examination behaviour indicating commercial urgency around a specific claim set
Abandonment and non-payment of maintenance fees signalling deprioritised programmes or financial strain

Where the data comes from

Authoritative and openly available collection points. Always confirm licensing and terms before operational or commercial use:

WIPO PATENTSCOPE — Free search across PCT international applications and dozens of national collections with full-text and family data
Google Patents — Free full-text search with machine translations, citation graphs, and linked assignment and litigation data
USPTO Patent Public Search and PatentsView — Authoritative US grants and applications plus a free bulk API for inventor, assignee and citation analytics
EPO Espacenet and OPS — Worldwide collection with INPADOC family and legal status data; OPS provides a free-tier programmatic interface
USPTO Patent Assignment Search — Recorded transfers, mergers and security interests over US patents, useful for ownership and distress signals
Lens.org — Free-tier linkage between patents and scholarly literature, valuable for bridging academic and patent intelligence
EPO Global Dossier and national registers — Prosecution file histories showing examiner objections and applicant amendments that narrow real claim scope

A working method

A repeatable sequence beats ad-hoc searching. This is a practical starting workflow:

Frame by claim, not by product — Translate the intelligence question into technical function and likely classification codes, since applicants deliberately avoid marketing vocabulary in claims.
Build the seed set — Find a handful of clearly on-target documents, then expand via classification codes, citations and family members rather than keyword search alone.
Normalise assignees — Consolidate corporate name variants, subsidiaries and holding vehicles, then check assignment records for post-grant ownership changes.
Date the timeline — Order the set by earliest priority date to reconstruct when the programme began and how the claimed scope evolved.
Read the file wrapper — Check prosecution history for the granted claims as amended; the abstract routinely overstates what the applicant actually owns.
Map the landscape — Plot filings by assignee, classification and jurisdiction over time to identify crowding, white space and abrupt strategy shifts.
Corroborate with adjacent sources — Cross-check against publications, regulatory approvals, job postings and procurement to confirm the filing reflects an active programme.

How this connects across the intelligence taxonomy

Intelligence work does not respect neat boundaries. The mission domain you are working, the disciplines you practise, and the data points you pivot on are one connected system. These are the direct relationships for this entry — every link is also a tag, so you can follow any thread across the whole library.

Applied in these mission domains

Operates on these data points

Patent — An intellectual property filing granting invention rights.
Company / Organization — A legal entity — corporation, LLC, NGO, or business.
Code Repository — A source-code repository — leaks secrets, reveals developers, and anchors supply-chain risk.
Software Package — A published dependency (npm, PyPI, Maven) — the vector for supply-chain compromise.
Credential / API Token — An exposed secret — API key, token, or JWT — granting access to systems and data.
Keyword / Narrative — A search term, topic, hashtag, or narrative tracked across media and platforms.

Related disciplines

Academic Intelligence — Research Output, Collaboration, and Expertise
Technical Intelligence — Technology Capability, Design, and Exploitation

Inside the platform: where Patent Intelligence lives

The Quantus platform is 204 pages behind a 147-item sidebar organised into six working groups: Command (24 items), Dashboards (15), Threat Theaters (14), Intelligence Domains (15), Investigate (34), and Administration (45). This entry is not a page in isolation — it is a thread running through several of them.

The modules that matter most here:

discipline.php?d=PATENTINT — Discipline hub
source-catalog.php?disc=PATENTINT — Source catalogue filtered to this discipline
search.php — Company / Organization profile
correlate.php — Correlation graph
cases.php — Case management

Each dashboard is local-first: it renders from the platform’s own database rather than depending on a live third-party call, so it still works when an upstream API is unreachable or rate-limited. Heavy aggregates are cached with a hard query time cap and degrade to the last good value instead of hanging the page.

Automation, playbooks and AI skills

Analysis that only happens when someone remembers to run it is not a capability. The platform ships a 30-step automation pipeline (cron.php) that collects, ingests, resolves, enriches, correlates and scores on a schedule — 25 seeders, 11 resolvers and 7 enrichment runners, all idempotent and cursor-based so a run can be interrupted and resumed without duplicating or losing work.

AI skills that apply

The 16 one-click operations in ai-skills.php are deterministic jobs, not free-text generation. The ones that matter here:

Threat Hunt
Auto-Collect Feeds
Enrichment → Local
Summarise (Copilot)
Generate Report

Alerting closes the loop: rules in alerts.php fire on new indicators matching a saved query, so a first sighting in this area raises a notification rather than waiting to be noticed at the next review.

Feeds, data sources and the API

The collection layer runs a feed registry of free, machine-readable sources — bulk blocklists and trackers (Maltrail, IPsum, FireHOL, the full abuse.ch corpora, phishing databases, Emerging Threats, Spamhaus, DigitalSide, ThreatView), authoritative government feeds (CISA KEV, OFAC, UN and EU sanctions lists), and reference datasets (RIR allocations, ip-to-ASN and geolocation tables, MITRE ATT&CK, EPSS). collect.php pulls them server-side on a schedule; feeds.php and source-catalog.php show what is registered, what it covers and when it last ran.

Anything the platform holds is reachable programmatically. The REST API in api.php exposes 11 endpoints — status, stats, search, lookup, recent, export, bulk_check, top_threats, by_category, categories, check — and export.php streams 18 formats in bounded chunks, so a million-row export neither exhausts memory nor times out:

STIX 2.1, MISP, OpenIOC 1.1, CEF (ArcSight), LEEF 2.0 (QRadar), Zeek/Bro intel, Snort/Suricata rules, Palo Alto EDL, BIND RPZ, hosts blackhole, iptables, CSV, JSON, NDJSON/JSONL, XML.

That covers the CTI standards (STIX 2.1, MISP, OpenIOC), SIEM ingestion (CEF, LEEF, Zeek), detection engines (Snort/Suricata), and direct enforcement (Palo Alto EDL, BIND RPZ, hosts, iptables) — so intelligence developed here can be actioned in the tools you already run, without a manual reformatting step. A TAXII 2.1 server and a MISP/RSS feed are also served for pull-based sharing.

Use cases

Three ways this entry earns its keep in day-to-day work:

Triage under time pressure. An artifact or report lands and you need a defensible read in minutes, not days. Frame by claim, not by product is the first move; the platform pre-computes the enrichment so the analyst spends the time on judgement rather than lookups.
Building the picture. A single indicator is rarely the story. Normalise assignees turns one artifact into a network — shared infrastructure, repeated selectors, the same operator behind different names — via the correlation graph and the cross-entity link engine.
Producing something actionable. Analysis that ends in a document nobody can use is wasted. Corroborate with adjacent sources feeds the case file, the detection rule, the block list or the referral — with sourcing attached so the recipient can verify it.

Case management (cases.php), watchlists, saved searches and scheduled reports mean the work persists between sessions and survives an analyst leaving the team.

How each sector uses Patent Intelligence

The same entry is worked very differently depending on who you are, what authority you hold, and what you are ultimately producing. A military analyst is supporting a commander’s decision; a journalist is meeting a publication standard; an NGO caseworker is protecting a person. The underlying artifacts are shared — the constraints, outputs and thresholds are not.

🎖 Military and defence

Defence acquisition and science staff use patent intelligence to date the start of an adversary or supplier programme, to judge whether a claimed capability has been reduced to practice, and to see which markets a producer intends to serve. Family filings reveal export ambition; classification codes place work in a technical taxonomy comparable across countries. Products feed technology watch, industrial base assessment and counter-proliferation support. Constraints are real: some jurisdictions impose secrecy orders on defence-relevant filings, so absence of filings by a national champion may mean concealment rather than inactivity, and technical detail extracted from patents can still be export controlled when aggregated and transferred.

🕵 National intelligence

National intelligence uses patents as a dated, legally attested disclosure stream that is fully open and therefore shareable. Requirements typically concern programme timing, technology transfer between entities, and quiet changes of ownership. Assignment and security interest records expose transfers to foreign holding vehicles and lending against IP, both of which precede corporate events that never appear in company filings. All-source fusion pairs filings with procurement, publications and trade data. Handling is straightforward because the material is public, but analysts must caveat that non-filing is a strategy: sensitive programmes protect through trade secret, and the record then goes quiet by design.

👮 Law enforcement

Investigators use patent records in counterfeiting, trade secret theft, export control and fraud cases. Inventor names and filing dates place individuals with specific technical knowledge at specific employers at specific times, which is directly relevant to misappropriation allegations. Assignment records establish who owned what and when, supporting title and standing questions. Evidentially, certified copies from the issuing office are what a court accepts, not database screenshots; obtain them through the office's official channel. Freedom-to-operate and infringement conclusions are legal opinions and must come from counsel. Prosecution file wrappers can also contain admissions about prior art and capability that assist an investigation.

🔍 Private investigation and corporate security

Corporate intelligence teams use patents for competitor programme mapping, technology due diligence in mergers and acquisitions, white space analysis and expert identification. Assignment and lien records give an early read on financial distress. Inventor mobility analysis shows where a rival's engineering talent went after a programme ended. What a private actor must not do is present analysis as a freedom-to-operate or validity opinion, which is regulated legal advice, or redistribute bulk content from a licensed database. Inventor-level analysis involves personal data and should remain proportionate to the commercial question rather than becoming a profile of an individual.

📰 Journalism and OSINT media

Journalists use patents to verify or puncture technology claims, because the claims section states what the applicant could actually defend rather than what marketing asserts. Verification requires reading the granted claims as amended rather than the abstract, checking whether the patent is granted or merely published as an application, and confirming current ownership through assignment records. A patent is not evidence a product exists or works. Give the assignee a right of reply, and be careful about naming inventors, who are usually employees with no control over how the filing is used and may face professional consequences from coverage.

🌍 NGO, humanitarian and human rights

Civil society uses patent data in access to medicines, agricultural biodiversity, environmental technology and surveillance technology accountability work. Family analysis shows where a medicine is protected and where generic production is legally possible, which directly supports procurement advocacy. Surveillance technology filings document capability that vendors decline to describe. Do no harm applies when filings identify individuals in jurisdictions where association with a controversial technology carries risk. Documentation for accountability should preserve the original office record with its identifier and date, since applications are amended, abandoned and reassigned, and screenshots without identifiers are worthless later.

🎓 University and research

Researchers use patent data for innovation studies, technology diffusion and science-to-technology linkage work. Methodological care centres on assignee harmonisation, family definition and the difference between simple and extended families, all of which change results substantially. Office coverage differs by era and jurisdiction, and classification schemes have been revised, so longitudinal work needs concordance tables. Ethics review is rarely required for bibliographic patent data but applies when inventor-level mobility is linked to employment or personal records. Reproducibility demands publishing the office, database version, family definition, classification version and extraction date alongside every count.

Playbook: working Patent Intelligence end to end

A repeatable sequence, from the moment the requirement lands to the moment a product is delivered and the case is closed out. Each phase states what you are trying to establish, not merely what to click — the point is a defensible chain of reasoning, not a checklist.

Phase 1 — Frame by technical function

Translate the intelligence question into technical function, effect and structure rather than product names, because applicants deliberately avoid marketing vocabulary in claims and use generic terminology to broaden scope. A good output is a function statement with candidate classification codes in both CPC and IPC, plus a synonym set covering the terms different national drafters use. Stop when a subject matter expert agrees the function statement covers the technology as a competitor would describe it in a filing rather than in a brochure.

Phase 2 — Build and validate a seed set

Find between five and twenty documents that are unambiguously on target, verified by reading the claims. These anchor everything downstream. A good output is a seed list where you can state, for each document, why it is on target. Stop when adding another seed does not introduce a new classification code, applicant or citation cluster, and record any technology sub-branch you deliberately excluded so the scope is visible to the reader of the final product.

Phase 3 — Expand by classification, citation and family

Expand the seed set using CPC and IPC codes, forward and backward examiner citations, and family members rather than keyword search alone, which misses filings drafted in different vocabulary or other languages. A good output is a corpus with the expansion route recorded per document. Stop when new documents are overwhelmingly off-target on claim reading, and quantify precision on a sample so the reader knows how noisy the corpus is.

Phase 4 — Harmonise assignees and inventors

Consolidate corporate name variants, transliterations, subsidiaries, joint ventures and holding vehicles into canonical entities, and normalise inventor names carefully because they are the weakest field in most databases. Cross-check against corporate registries. A good output is an entity table where each canonical assignee lists its observed variants and its registry identifier. Stop when the top twenty applicants by volume are all resolved to a legal entity, since the tail rarely changes structural conclusions.

Phase 5 — Check assignment and encumbrance records

Pull assignment history and security interests. Ownership at grant is frequently not ownership today, and recorded liens against patent portfolios are an early and public indicator of financial distress or of a quiet transfer to a foreign holding company. A good output is a current ownership statement per key family with the recorded date of each transfer. Stop when the chain of title is continuous or a gap is documented as unrecorded, which is itself worth reporting.

Phase 6 — Establish the priority timeline

Order the corpus by earliest priority date, not publication date, to reconstruct when the programme actually began and how claimed scope evolved. Remember the roughly eighteen month publication delay, which means the newest visible filings describe work already done. A good output is a dated programme timeline with inflection points marked. Stop when the timeline covers the full period of interest and gaps are explained by either publication lag or genuine inactivity.

Phase 7 — Read granted claims and file wrappers

Read claim one as granted, then the prosecution history to see what was surrendered during examination. Abstracts and titles routinely overstate what the applicant owns, and continuation practice can leave a broad-sounding family with narrow enforceable scope. A good output is a plain language statement of what each key patent actually covers. Stop when the key families are characterised; do not read wrappers for the entire corpus, which is rarely worth the time.

Phase 8 — Map the landscape and white space

Plot filings by applicant, classification, jurisdiction and priority year to reveal crowding, abandonment, sudden entries and abrupt strategy shifts. Genuine white space is usually either technically infeasible or commercially uninteresting, so treat an empty region as a question rather than an opportunity. A good output is a landscape with an explanation for each empty area. Stop when the map answers the customer question rather than when it looks complete.

Phase 9 — Analyse jurisdictional intent

Read family member selection as market intent: where an applicant pays for national phase entry, translation and maintenance is where it expects to sell or to enforce. Withdrawal or non-payment of renewal fees signals abandonment of a market or of the technology. A good output is a market intent map by family with fee status. Stop when the pattern is consistent across the applicant's portfolio, since single-family choices can reflect budget rather than strategy.

Phase 10 — Track inventor mobility

Follow named inventors across assignees and over time to see where technical talent moved, which frequently identifies a new programme before the new employer files. Keep this proportionate: it is personal data, and the analytical purpose should be organisational capability, not surveillance of individuals. A good output is a movement summary at team level. Stop when the movement pattern is established, and avoid building persistent profiles of named engineers beyond the question at hand.

Phase 11 — Corroborate with adjacent sources

Confirm that filings reflect an active programme using publications, regulatory approvals, standards contributions, job advertisements, procurement notices and trade data. Filing without any other footprint may indicate defensive or speculative filing. A good output states, per key finding, whether independent corroboration exists. Stop when the highest-consequence findings are corroborated or explicitly flagged as filing-only evidence.

Phase 12 — Report with legal boundaries stated

Deliver dated findings on programme timing, ownership, scope and market intent with the office identifiers for every cited document, and state plainly that the product is not a legal opinion on validity, infringement or freedom to operate. Note secrecy order regimes and trade secret strategy as reasons absence of filings is not absence of capability. A good output can be verified independently from the identifiers given. Stop before characterising infringement risk in any form.

The platform ships this as a step-checked workflow in playbooks.php, so progress is recorded against a case rather than held in someone’s head.

Source register: what to collect from, and how

Sources are listed with their access model so you can plan around cost and licensing before you build a dependency on them. Open means no account required; registration means a free account or API key; licensed means paid or institutional access. Always confirm current terms — licensing changes, and a source that was free for research may not be free for commercial or evidential use.

Source	Access	What it gives you	How it is used here
WIPO PATENTSCOPE	Open	International PCT applications plus dozens of national collections with full text search, family data and machine translation	First stop for global family coverage and for non-English filings that Western databases index poorly
Google Patents	Open	Full text of many national collections with OCR, machine translation, citation links, litigation flags and similar document search	Fast seed set building and readable full text, including scanned older documents other systems do not render
Espacenet	Open	EPO worldwide collection of over a hundred million documents with classification search, INPADOC family and legal status data	Authoritative family construction and legal status checking, which determines whether a right is still in force
USPTO patent public search and open data	Open	Official US granted patents, published applications, examination data and bulk data products including assignment records	Authoritative US text, prosecution status and recorded ownership transfers with dates
PatentsView	Open	Disambiguated US patent data with cleaned inventor, assignee, location and citation tables and a query API	Inventor mobility and assignee-level quantitative analysis without building your own disambiguation pipeline
European Patent Office Open Patent Services	Registration	Programmatic access to Espacenet bibliographic data, families, legal status and full text where available	Automated landscape refresh and legal status monitoring across large corpora without manual retrieval per document
Lens.org	Registration	Combined patent and scholarly corpus with linkage between publications, patents and institutions, and portfolio tools	Science-to-technology bridging, connecting a research group's papers to its filings in one query
China National Intellectual Property Administration	Open	Official Chinese patent and utility model records including examination status and Chinese language full text	Essential for coverage of the largest filing jurisdiction, including utility models absent from many datasets
Japan Platform for Patent Information	Open	Japan Patent Office search platform with Japanese and English abstracts, legal status and machine translation	Access to Japanese filings and their prosecution status, often the earliest disclosure in electronics and materials
Korean Intellectual Property Rights Information Service	Open	Korean patent, utility model, design and trademark records with English search and status data	Coverage of Korean industrial filings in displays, batteries, semiconductors and shipbuilding
USPTO Patent Trial and Appeal Board records	Open	Inter partes review, post grant review and appeal proceedings with filings, evidence and decisions	Reveals which patents competitors consider threatening enough to attack, and exposes technical evidence
EPO Register and Global Dossier	Open	European prosecution history, oppositions, appeals and linked file wrappers from participating offices	Shows what was surrendered during examination and who filed oppositions, which names commercial rivals
WIPO IP statistics data centre	Open	Aggregated filing, grant and family statistics by office, origin, technology field and year	Provides denominators so an applicant's filing surge can be judged against overall office growth
Cooperative Patent Classification scheme	Open	Joint EPO and USPTO classification taxonomy with definitions, concordances and revision history	Defines the search space and enables longitudinal comparison across classification revisions
Derwent Innovation or Orbit Intelligence	Licensed	Commercial platforms with rewritten abstracts, curated assignee harmonisation, legal status and analytics	Faster harmonisation and landscape production where budget exists, with redistribution limits on output
Espacenet legal status and INPADOC data	Open	Consolidated legal event data covering grant, lapse, opposition, transfer and renewal fee payment across offices	Determines whether a right is live in a given market, which is the question most customers actually mean

Prefer sources that publish a methodology and a revision history. A dataset that changes silently is a liability in any product that has to survive challenge.

Tooling

Tools commonly used against Patent Intelligence. None of these replace judgement, and each carries its own failure modes — know what a tool infers versus what it observes.

Espacenet classification search — Browses and searches CPC and IPC definitions to build code-based queries; authoritative, though the taxonomy takes practice to navigate efficiently.
PatentsView API and bulk downloads — Structured US patent data with disambiguated inventors and assignees; excellent for quantitative work but US only and lagging live records.
EPO Open Patent Services client libraries — Automates family, bibliographic and legal status retrieval; rate limited and requires careful handling of family definitions.
OpenRefine — Clusters and harmonises assignee name variants at scale; effective, but transliterated and joint venture names still need manual adjudication.
VOSviewer or Gephi — Visualises citation and co-classification networks across a landscape; readable output, though layout choices can imply structure that is not there.
Python with pandas and matplotlib — Builds filing trend, jurisdiction and applicant charts reproducibly from exports; requires the analyst to handle family deduplication explicitly.
Google Patents similar documents and prior art finder — Surfaces conceptually related filings without exact terminology; useful for seed expansion, opaque about how similarity is computed.
PDF and OCR tooling for file wrappers — Extracts text from scanned prosecution documents and older grants; accuracy varies sharply with scan quality and non-Latin scripts.
Machine translation with domain glossaries — Renders Chinese, Japanese and Korean filings readable for triage; adequate for screening, unreliable for claim scope, which needs a qualified translator.

AI skills and automation in detail

These are deterministic jobs with defined inputs and outputs, not open-ended prompting. Each is idempotent and cursor-based: interrupt one and it resumes where it stopped rather than duplicating work or losing progress.

Threat Hunt — Runs saved hypotheses against the corpus and surfaces what matches, with the query preserved as a versioned artifact.
Auto-Collect Feeds — Pulls the registered feed set server-side on a schedule, recording per-feed status so a silently dead feed is visible.
Enrichment → Local — Materialises enrichment into the local store so dashboards render from your own database instead of a live third-party call.
Summarise (Copilot) — Produces a narrative summary beside the underlying records. It explains; it never creates indicators or assigns attribution.
Generate Report — Assembles a sourced product from the current case or query, with provenance attached to each element.

A note on the boundary: the only skill that involves a language model is Summarise (Copilot), and it writes prose about records that already exist. Nothing else on this list involves generation of any kind. No indicator, relationship or attribution in the platform originates from a model. See the full skill list.

Tradecraft notes

The distinctions that separate a competent analyst from a fast one:

Read claim one as granted, not the abstract. Abstracts describe ambition and are drafted for searchability. The granted claim, narrowed through prosecution, is the only text that states what the applicant can actually enforce, and the gap between the two is often enormous.
Priority date, not publication date, dates the programme. Publication lags roughly eighteen months, so a landscape built on publication dates systematically misplaces the start of every programme and makes fast movers look slower than they are.
Where an applicant pays is where it intends to sell. National phase entries, translations and renewal fees are expensive and deliberate, so family geography is a better statement of commercial intent than any executive interview.
Non-filing is a strategy. Process innovations, algorithms and defence work are frequently protected by secrecy rather than patents, and some jurisdictions impose secrecy orders. Silence in the record can mean the technology matters more, not less.
Utility models and design rights are routinely omitted from Western datasets and dominate filing volume in some jurisdictions. Excluding them makes an active applicant look dormant and distorts every national comparison you draw.
Examiner citations are analytically richer than applicant citations. They are added by a professional searching for the closest prior art, so they reveal genuine technical adjacency rather than the applicant's chosen framing.
Assignment and lien records are financial intelligence. Recorded security interests over a portfolio, or transfers to an unfamiliar holding vehicle, frequently precede distress, acquisition or export of a technology, and they appear months before any corporate announcement.

Measuring whether it is working

Capability claims should be falsifiable. These are the measures that show whether work on Patent Intelligence is producing anything, and they are worth baselining before you change process or tooling.

Lead time between first priority filing detected and the corresponding product, procurement or capability appearing publicly, tracked per programme.
Precision of the landscape corpus, measured by expert claim reading on a random sample, reported alongside every landscape product.
Assignee harmonisation completeness, measured by the share of filings assigned to a canonical entity with a registry identifier.
Proportion of key findings verified against the issuing office record rather than an aggregator, which is the difference between citable and unusable.
Number of commercial or programme decisions changed by a patent finding, such as a partnership reassessed or a technology path abandoned.
Rate at which reported ownership proves current when re-checked at delivery, since assignments recorded late are a persistent error source.
Time to refresh a monitored landscape after a scheduled data update, tracked to show whether monitoring is genuinely continuous.

Beware of measuring volume alone. Indicator counts and report counts rise easily and say little; time-to-attribution, proportion of findings that survive review, and how often a product changed a decision say a great deal.

Common pitfalls

Reading the abstract instead of the granted claims, which are usually far narrower after examiner amendments
Ignoring the roughly eighteen month publication delay and reporting a filing landscape as if it were current activity
Treating filing volume as innovation quality when defensive, blocking and vanity filings inflate counts substantially
Missing filings held under subsidiary, holding company or inventor names, producing a falsely thin portfolio picture
Overlooking that unfiled trade secrets are the deliberate alternative, so absence of patents can mean secrecy rather than absence of capability
Assuming a granted patent is valid or enforceable; many are narrowed, invalidated or lapsed for unpaid fees

Legal and ethical considerations

Patent documents are published to be read, so analysis and citation are unproblematic, but bulk redistribution of commercial database content is restricted by licence. Freedom-to-operate and infringement opinions constitute legal advice in most jurisdictions and must come from qualified counsel, not analysts. Some jurisdictions impose secrecy orders on defence-relevant filings, and technical detail extracted from patents can still fall under export control when combined and transferred. Keep inventor-level analysis proportionate, since it involves personal data.

Data integrity: no fabrication, no drift, no hallucination

Intelligence that cannot be traced back to a source is not intelligence, it is assertion. Everything in this entry — and everything in the platform behind it — is built on a small number of non-negotiable rules.

Provenance on every record

Every indicator carries the source that supplied it, a first-seen and last-seen timestamp, and a sighting count. Where several feeds report the same artifact, each contribution is recorded separately rather than collapsed, so you can see whether a finding rests on one source or twelve. Source attribution travels with the data into every export, so a recipient can audit a claim without asking you for the working.

Nothing is invented to fill a gap

If the platform has no data for Patent Intelligence, it says so. Empty is displayed as empty — never padded with plausible-looking placeholder values, sample records or illustrative examples that a reader might mistake for observations. A dashboard with no rows is a true statement about collection coverage, and it is treated as a gap to close, not a blemish to hide.

Scoring is deterministic and reproducible

Threat scores, reputation grades and risk tiers are computed from stated inputs with fixed weights, not estimated. The same inputs always produce the same output, and the formula is visible rather than a black box. Aggregates are cached with an explicit time-to-live so a figure on screen is never silently stale — and when a heavy query exceeds its time budget the platform serves the last known-good value and labels it, rather than inventing a fresh number or hanging.

Where AI is used, and where it is not

Language models summarise and explain. They do not create indicators, assign attribution or manufacture relationships. No IP address, wallet, hash or identity in the platform originates from a model — every one is ingested from a named feed, resolved from a reference dataset, or entered by an analyst with a source recorded. Copilot output is presented as narrative alongside the underlying records, never in place of them, so a reader can always check the summary against the evidence.

Guarding against drift

Enrichment is additive and timestamped rather than overwriting. Reference data — sanctions lists, allocations, taxonomies — is re-synchronised from the authority on a schedule instead of being edited in place, so local copies cannot quietly diverge from the source of truth. Attribution is recorded with a confidence level and the reporting it rests on, and inferred relationships are labelled as inferred. When a source retracts or corrects, the correction propagates rather than leaving a stale assertion behind.

What this means for you

You can put a finding from this platform in front of a regulator, a court, a board or a partner agency and show where each element came from. That is the standard the tooling is built to — because in this work, being confidently wrong is more damaging than being usefully uncertain.

By the numbers

The taxonomy this entry belongs to is not a marketing list — it is the actual structure of the platform: 52 mission domains, 52 intelligence disciplines and 65 data points, each with a live dashboard behind it. Supporting that: 18 indicator types, 14 playbooks, 16 AI skills, 18 export formats and a 30-step automated pipeline.

This particular entry connects directly to 6 data points, 3 mission domains, 2 closely related entries — every one of them a tag you can follow, and a dashboard you can open.

Questions analysts actually ask

Does a patent prove a company can build the thing?

No. A patent proves someone drafted claims that an examiner allowed over the prior art, which requires enablement on paper rather than a working production line. Many filings are speculative, defensive or intended to block rivals. Treat a patent as evidence of intent, of a technical approach considered worth protecting, and of a date. To establish real capability, corroborate with procurement records, regulatory approvals, publications, standards contributions, job advertisements or physical evidence. The converse also holds: absence of filings does not prove absence of capability, because trade secret protection and secrecy orders both produce silence.

How do I choose between simple and extended families?

Use simple families, which share exactly the same priority set, when counting distinct inventions and comparing applicants, because extended families inflate counts by chaining loosely related filings. Use extended INPADOC families when tracing where a technology has been protected across markets or when following continuations and divisionals that carry claims in different directions. Whichever you choose, state it explicitly in the product, because family definition alone can change a headline count by a factor of two and makes numbers from different analysts incomparable.

Can I do a freedom-to-operate analysis?

Not as an intelligence analyst. Freedom to operate, validity and infringement are legal opinions that in most jurisdictions may only be given by qualified counsel, and an unqualified opinion can expose your organisation to willfulness findings and your client to costs. What you can lawfully deliver is a landscape: who holds rights in the relevant space, in which jurisdictions, with what legal status, and with claim scope summarised factually. Hand that to counsel as input. State the boundary in the product itself so no reader mistakes the deliverable for legal advice.

How do I handle Chinese filings properly?

Search CNIPA directly as well as through international aggregators, include utility models, and treat volume with care because filing subsidies have historically inflated counts without corresponding technical substance. Use grant status, family extension abroad, renewal payment and citation by foreign examiners as quality filters. Machine translation is adequate for triage but unreliable for claim scope, so commission a qualified translation for any filing that carries a conclusion. Note also that inventor and assignee transliteration is inconsistent, so harmonisation needs the Chinese characters rather than the romanised strings.

What does a lapsed patent tell me?

That the owner stopped paying to maintain it, which is a deliberate commercial decision. Lapse in some jurisdictions but not others maps where the owner still expects revenue. Wholesale lapse across a family usually means the programme was abandoned, the technology was superseded, or the company is short of cash, and clustering lapses in time is a useful distress indicator. It also means the disclosed technology has entered the public domain in those markets, which is directly relevant to procurement, generic manufacture and access advocacy. Always check status per jurisdiction rather than per family.

How far can I go with inventor analysis?

Far enough to answer an organisational question, and no further. Tracking which team moved from one employer to another, or that a named inventor group stopped filing after a programme ended, is legitimate competitive intelligence. Building a persistent profile of an individual engineer, inferring their nationality or personal circumstances, or using the data to target them personally goes beyond the purpose and engages data protection law in most jurisdictions. Keep individual-level detail out of the delivered product where a team-level statement answers the question, and document the proportionality decision.

Which database should be authoritative when they disagree?

The issuing office, always. Aggregators normalise, translate and occasionally mis-parse, and legal status data in third party systems can lag by months. Use Espacenet or Google Patents for discovery and speed, then verify every number, date, claim and ownership statement that carries a conclusion against the office register: USPTO for US, EPO Register for European, CNIPA, JPO or KIPO for Asian filings. For any evidential use, obtain certified copies through the office's official channel, because database exports are not accepted as proof of the record.

Standards, frameworks and further reading

Work that references a recognised framework is easier to defend, easier to hand over, and easier for a partner to consume:

Patent Cooperation Treaty, which governs international applications, the eighteen month publication rule and national phase entry deadlines that structure the visible record.
Paris Convention priority rules, which fix the twelve month priority period and therefore the relationship between first filing and family members.
Cooperative Patent Classification and International Patent Classification schemes, which define the technical taxonomy used for landscape search and comparison.
WIPO Standard ST.16 and related kind codes, which distinguish published applications from grants and are essential to reading a document identifier correctly.
INPADOC family definitions maintained by the EPO, which determine how documents are grouped and therefore how counts are produced.
National secrecy order regimes such as the US Invention Secrecy Act and equivalent provisions elsewhere, which suppress defence-relevant filings from publication.
Export control regimes including the EU dual-use regulation and ITAR, which can control technical data extracted and aggregated from published patents when it is transferred.
TRIPS Agreement, which sets minimum standards of protection and the flexibilities relevant to access to medicines and compulsory licensing analysis.

References

Primary sources and authoritative references for this entry. Publishers revise and retire material, so treat the retrieval date as part of the citation and re-check before relying on any of it in a formal product.

PATENTSCOPE international patent search system — World Intellectual Property Organization. Global search across PCT applications and national collections with family and translation support
Espacenet worldwide patent database — European Patent Office. Over a hundred million patent documents with classification, family and legal status data
USPTO open data and patent public search — US Patent and Trademark Office. Official US patent text, examination data, assignment records and bulk data products
PatentsView disambiguated patent data — US Patent and Trademark Office and research partners. Cleaned inventor, assignee and citation tables supporting quantitative innovation analysis
Google Patents full text collection — Google. Searchable full text across many national collections with OCR and machine translation
Cooperative Patent Classification — EPO and USPTO. The joint classification taxonomy and its definitions, concordances and revision history
WIPO IP Statistics Data Center — World Intellectual Property Organization. Official filing and grant statistics by office, origin and technology field
European Patent Register — European Patent Office. Prosecution history, oppositions and appeals for European applications and patents
Lens.org scholarly and patent linkage — Cambia. Combined corpus linking publications, patents and institutions for science-to-technology analysis

Link integrity: every reference above was verified with a live request when this page was generated. Where a publisher had moved or withdrawn a document, the link was repointed at a preserved copy in the Internet Archive and marked as archived. Anything with no reachable copy anywhere had its link removed rather than left to rot — the source is still credited, it simply cannot be linked.

Put it into practice

The Quantus Intel threat intelligence platform operationalises this entry: tracks filings, families and assignments by assignee and technology, alerting on portfolio and ownership shifts. Explore the platform, or browse the rest of the library by following any tag above.