Innovation and patents in university rankings: measuring knowledge transfer

Why innovation and patent indicators matter

Research universities are increasingly expected to demonstrate economic and social impact beyond the academic sphere. Governments, funding agencies, and taxpayers want evidence that publicly funded research translates into innovation, economic growth, and solutions to societal challenges. University rankings have responded by incorporating innovation indicators, most notably patent counts, spin-off company formation, and industry research income. These metrics attempt to capture the third mission of universities—knowledge transfer and economic engagement—which sits alongside the traditional missions of teaching and research.

The inclusion of innovation metrics in rankings such as THE, U-Multirank, and Reuters Top 100 Innovative Universities reflects a broader shift in higher education policy toward valorization and impact. However, innovation is unevenly distributed across disciplines, institutional types, and national innovation systems, meaning that rankings containing these indicators can produce results that diverge substantially from those that focus purely on academic research. Understanding what patent and innovation indicators actually measure—and what they leave out—is essential for interpreting rankings that include them.

Patent-based indicators and their limitations

Patents are the most commonly used proxy for university innovation in rankings. The rationale is straightforward: patents represent the codification of new knowledge with potential commercial application, and universities that generate more patents are presumed to be more innovative. Rankings typically count patents filed by university-affiliated inventors, often normalized by institutional size or research expenditure, drawing data from patent databases such as the World Intellectual Property Organization, the European Patent Office, or the United States Patent and Trademark Office.

However, patents are a deeply imperfect innovation metric. First, patent propensity varies enormously by discipline. Biotechnology, pharmaceuticals, engineering, and computer science generate large numbers of patents because patentable inventions are a natural output of research in these fields. History, philosophy, literature, mathematics, and many social sciences produce essentially no patents regardless of the quality or impact of their research. A ranking that includes patent counts systematically penalizes institutions with strong humanities and social science programs. Second, patenting cultures differ across countries: universities in the United States have a long tradition of patenting following the Bayh-Dole Act of 1980, while European and Asian universities have historically had more varied approaches to intellectual property management.

Third, patent counts measure quantity, not quality or value. Most university patents are never commercialized and generate no licensing revenue. A simple count treats a patent on a minor incremental improvement the same as a patent on a transformative technology. Some rankings attempt to address this by weighting patents by forward citations—the number of times subsequent patents cite them—as a proxy for technological impact. Even with this refinement, patent data reflects only one form of innovation, leaving unmeasured the many other ways universities contribute to economic and social progress, including policy influence, cultural production, open-source software development, and public engagement.

Spin-off companies and industry income

Spin-off company formation is another innovation indicator used in rankings such as U-Multirank. The creation of a new company to commercialize university research is a tangible signal of knowledge transfer. However, counting spin-offs raises comparability issues: definitions of what constitutes a university spin-off differ across countries, reporting standards vary, and the economic value of spin-offs ranges from high-growth startups to small consultancies with limited economic impact. Some rankings attempt to track spin-off survival rates and venture capital investment, but these data are inconsistently available.

Industry research income, used in THE and U-Multirank, measures the amount of research funding universities receive from corporate partners. This indicator captures the commercial relevance of university research as judged by companies willing to pay for it. Industry income varies dramatically by field: engineering and biomedical sciences attract substantial corporate funding, while humanities and theoretical sciences attract little. The indicator also reflects national differences in corporate R&D spending and university-industry collaboration norms. A university in a country with strong government funding for basic research may show lower industry income than an equivalent institution in a country where businesses fund a larger share of university research.

Using innovation indicators wisely

Innovation metrics in rankings are best used for specific purposes. If you are a prospective graduate student in engineering or biotechnology interested in commercialization, rankings with patent and spin-off indicators can help identify universities that are active in translating research into technology. If you are a policymaker evaluating knowledge transfer systems, these metrics provide comparative benchmarks. However, if you are evaluating a university's overall quality or considering humanities or social science programs, innovation indicators should be given low weight in your assessment.

Always check the specific definitions and data sources behind the innovation indicators in any ranking you use. Verify whether patents are counted by application or grant, whether they are normalized by size, and whether field-weighting is applied. Consider the national context: a university with low patent counts may operate in a country with different intellectual property norms and an equally valid approach to knowledge dissemination. Supplement patent and innovation data with qualitative evidence about research impact, industry partnerships, and technology transfer outcomes. Rankings provide a useful starting point, but understanding a university's actual innovation performance requires a much richer picture than any single metric can provide.