How GIPResearch.org conducts and evaluates research

Research Methodology: evidence-led reference

This reference explains how a public IP research conclusion can be traced from a precisely framed question to independently identifiable documents. It is an editorial method, not a claim that a particular search was carried out. The central distinction is between retrieving documents, interpreting those documents and reaching a legal or business conclusion. A bibliography alone does not make an analysis reproducible: the author must explain the corpus, dates, search choices, exclusions, and how contrary results affected the answer. The WIPO Guide to Using Patent Information and WIPO's patent landscape guidelines provide relevant external methodological reference points.

Define a repeatable research question

Start with the decision a reader is attempting to make, such as whether a technical feature is already disclosed, which organizations appear in a published patent corpus, or which source describes a filing route. Record jurisdiction, time horizon, unit of analysis, and what constitutes a sufficient answer. A question about an inventor's technical contribution is different from a legal opinion about patentability; one cannot silently substitute the other. Specify whether the output is a source map, an analytical comparison, a search report, or a professional interpretation. If the question changes after discovery, keep the earlier question in the change log and explain the revision.

Use identifiable primary records

For patents, capture publication or application identifiers, country or office, kind codes where available, dates and official record links. Bibliographic records, published patent documents, claims, office-action histories and legal-event feeds are related but not interchangeable. WIPO's PATENTSCOPE addresses published PCT documents and participating collections; the EPO's Espacenet provides a different retrieval interface. Academic papers require identifiers such as DOI, authorship and publication venue. A search engine snippet is only a locator, not the evidence itself. Preserve the original passage or bibliographic field on which each consequential statement relies.

Primary reference: WIPO Guide to Using Patent Information. Interpret only within the source’s published jurisdiction and scope.

Document queries and classification editions

Describe exact keyword combinations, alternative technical terms, field restrictions, and relevant classification symbols. Patent classifications change with revisions, so record whether a query used IPC 2026.01, CPC 2026.08, or a historical version; do not infer crosswalk equivalence merely from similar symbols. Reproducibility also requires a search date, data coverage statement, and exclusions such as unpublished applications or non-English material. A reviewer should be able to rerun the query or understand why the original result count is not reproducible after an underlying database update. Keep unsuccessful variants because they help explain the search boundary.

Separate evidence strength from author interpretation

Make four columns in the working evidence log: exact proposition, primary record, inference drawn, and confidence or limitation. A patent publication can establish that disclosed text was publicly available by a particular date; it does not by itself establish that a product infringes a granted claim. A citation count can identify relationships in one database, but cannot independently measure invention quality or economic significance. An official classification symbol can organize a corpus without proving that every listed document is technologically relevant. These distinctions should appear next to the conclusion, rather than be hidden in a generic disclaimer.

Primary reference: WIPO Guidelines for Preparing Patent Landscape Reports. Interpret only within the source’s published jurisdiction and scope.

Use counterexamples to test analytical robustness

Challenge the selected terms and filters with cases near the boundary: patents using unusual terminology, cross-disciplinary inventions, later patent-family members, and records lacking abstracts or machine-readable claims. For comparative research, test whether different deduplication rules or filing-date choices change the observed pattern. Record at least one plausible alternative interpretation and explain why it was included or excluded. A conclusion that survives a sensitivity check is more useful than a precise-looking chart with an undocumented population. Where results cannot be replicated with an accessible interface, publish the limitation, not an implied complete dataset.

Record review and meaningful updates

The editorial workflow should reserve distinct checkpoints for source retrieval, factual verification, domain-specialist review where needed, and final publication. This page describes the recommended standard; it does not certify that each prior page has completed every step. A later correction should identify the affected proposition, previous wording, corrected support and substantive edit date. A cosmetic update must not reset every sitemap date. For evolving law or classification schemes, establish an event-driven review trigger connected to the official issuing authority; for stable definitions, periodic verification may suffice.

Publish a bounded conclusion

An effective report tells the reader both what the identified evidence supports and what was not tested. Include a reproducibility packet with query date, database, query string, classification edition, document IDs, evidence notes, rejected alternatives, and final analytical question. Source maps can provide substantial public value without claiming the completeness of a commissioned freedom-to-operate search or the authority of a legal opinion. Refer users requiring a specific professional determination to appropriately qualified advisers. This .org method page therefore documents method governance; the separate /methodology/ hub helps choose among different patent-search methods.

Primary reference: WIPO PATENTSCOPE. Interpret only within the source’s published jurisdiction and scope.

Illustrative research scenario: applying the distinctions

Consider a hypothetical analyst comparing patent publications about a mechanism used in compact water connectors. First describe the mechanism without assuming it is patentable. Search the official records using a documented technical phrase set and a separate classification-assisted query, preserving both result sets. Select three representative documents and record publication identifiers, relevant figures or claims, dates and the passages relevant to the technical feature. If one record describes a similar geometry but a different locking element, note the precise difference instead of asserting novelty from a visual impression. Search non-patent literature for earlier technical disclosure. Then repeat the query using a broader functional term and explain whether new records changed the conclusion. This would establish a repeatable *search illustration*, not a completed patentability or infringement opinion. The resulting publication can include its search protocol, a small evidence matrix and a clear unresolved question. If the documents are merely illustrative rather than collected, label them as an example and never fabricate identifiers. The major output is an explainable reasoning trail, not a persuasive count.

Practical decision and evidence matrix

What this reference establishes — and what it does not

The material above is a source-bounded reference concerning research methodology. It distinguishes official documentation from editorial method and provides practical decision boundaries; it does not claim an exhaustive global search, an empirical dataset that was not collected, a legal opinion, or certified implementation of publisher policy. Readers seeking to apply a source to a current filing, claim or professional obligation should inspect the applicable original authority, record its effective date, and obtain the necessary qualified advice. A page is updated substantively only when its evidence, method or answer materially changes.