Innovation has become central to achieving high-value growth, particularly in economies seeking to escape the middle-income trap where competitiveness still rests on low wages and process cost-cutting. In triple helix innovation ecosystems, technology transfer is seen as a crucial link enabling public–private value co-creation. Yet the efficiency of these processes remains poorly understood. University technology transfer efficiency (UTTE) — the relationship between inputs devoted to transferring research into commercial applications and resulting outputs — poses significant conceptual and measurement challenges. Inputs are ambiguous and multidimensional (research expenditure, technology transfer office resources, organizational capabilities); outputs are heterogeneous (patents, licenses, spin-offs, licensing income, broader impacts); and no consensus exists on which efficiency concept should formally link them. Despite a substantial body of empirical literature, UTTE measurement remains fragmented across incompatible units, anchored to transactional proxies, and disconnected from the economic outcomes that justify the investment. What is being measured is not what matters; what matters is not being measured.
This paper investigates whether the existing literature offers an adequate conceptual and empirical basis for measuring UTTE — and if not, what is missing. Our systematic review examines 71 peer-reviewed studies published between 2006 and 2026, analyzing: (a) how inputs and outputs are defined and operationalized; (b) which methodological frameworks relate inputs to outputs; (c) how studies address data limitations and cross-institutional comparability; and (d) to what extent emerging metric frameworks from professional associations (e.g., ASTP) are reflected in recent empirical work.
We find that DEA-type non-parametric frontiers dominate methodologically, with stochastic frontier analysis providing a parametric alternative; that data limitations severely constrain cross-institutional and cross-country comparability; and that emerging professional frameworks have only partially penetrated empirical practice. Taken together, these findings reveal a field built around what is measurable and institutionally available rather than what is theoretically adequate — and point toward a reoriented research agenda that moves upward in aggregation level, from institutions to ecosystems; forward in the value chain, from transfer outputs to enterprise outcomes; and deeper in explanatory mechanism, from quantitative proxies to qualitative process understanding.