Patent white space identification

Yes. For white-space analysis on current patent filings, I’d shortlist PatSnap, Questel Orbit Intelligence, and Clarivate Derwent Patent Search. The right choice depends on whether your priority is rapid innovation discovery, sophisticated patent landscaping, or defensible patent research.

1. PatSnap — my first tool to evaluate

PatSnap's landscaping functionality is explicitly designed to identify technology gaps: it clusters patents using classification and semantic information, visualizes patent density, and lets you investigate the areas between clusters as potential white spaces. It also has tools for putting a proposed idea/text onto the landscape to see where it sits relative to existing inventions.help.patsnap.com

Why I'd demo it: If the people doing the analysis are R&D/product strategists rather than patent-search specialists, the visual exploration is particularly useful.

2. Questel Orbit Intelligence — very strong for serious landscaping

Orbit's landscape maps use NLP/semantic concepts from patent full text, including claims and descriptions, and cluster patent families based on semantic similarity and classification. Its documentation specifically identifies subtopic and white-space identification as a use case.intelligence.help.questel.com+1

I'd lean toward Orbit if you have a sophisticated IP team that wants to construct and interrogate its own landscapes rather than simply consume an automatically generated opportunity map.

3. Clarivate Derwent Patent Search — strongest candidate for patent-quality research

Derwent combines global patent data with expert-created DWPI summaries and AI search. Its current product covers 165+ million patent records from 105+ jurisdictions, and its AI search is trained on Derwent's curated data.clarivate.com

Clarivate also explicitly offers technology-landscape analysis that looks at patent density and potential whitespace, including changes in filing activity and competitor positioning.clarivate.com

I'd favor Derwent when: the output needs to feed directly into patentability/FTO work or be defensible to patent counsel.

One important caveat

"White space" shouldn't simply mean an empty region on a patent map. An apparent gap can result from classification/search limitations, unpublished applications, terminology differences, or a technology being protected through claims that don't look obvious at the landscape level. PatSnap itself notes that white spaces are not necessarily completely empty.help.patsnap.com

For your use case, I'd therefore structure the workflow as:

Current filings → normalize into patent families → semantic clustering → technology/subtechnology taxonomy → identify low-density combinations → inspect claims → overlay assignees/competitors → check filing momentum → validate candidate white spaces manually.

What I'd do before buying

Run the same technology landscape through all three vendors using, say, the last 5–7 years of filings plus the most recent published applications. Give each vendor the same 5–10 innovation questions, such as:

"What combinations of X + Y + Z have relatively little patent activity but are adjacent to rapidly growing areas?"

Then compare how many genuinely interesting opportunities their systems surface, not just how attractive the maps look.

If you tell me what technology/domain you're innovating in (e.g., batteries, medical devices, AI, semiconductors, materials), I can do a more concrete comparison and identify which tool is likely to handle that particular patent landscape best, including pricing/enterprise considerations.