Interventions from within – The case of Action Research in the innovation ecosystem Open Arenas

Abstract Insider Action Research conducted within innovation ecosystems (IES) generates a qualitatively distinct set of methodological and ethical challenges that cannot be derived from single-organization or network-level action research alone. When the system being researched is simultaneously the system on which the researcher’s own organization depends for funding, legitimacy, and survival, a condition we term “double embeddedness” arises that transforms the foundational action research questions of “action for whom?” and “knowledge in whose interest?” from methodological concerns into existential ones. Drawing on a longitudinal reflective account of Project Open Arenas (POA), a government-commissioned action research initiative embedded in the western Swedish automotive IES during 2004–2005, this paper develops four propositions: (1) temporal boundaries may extend beyond the formal research project, requiring explicit temporal contracting; (2) multi-level embeddedness creates potentially competing organizational and system-level loyalties; (3) glocal dynamics – transnational actors redefining system boundaries mid-process – can override local intervention logic in ways current frameworks do not prepare the researcher for; and (4) IES-level action research may become entangled with system-level changes that have unforeseen consequences for the researcher's own organization. Two structural responses are proposed: an interdisciplinary research team configured to hold the insider/outsider tension productively, and a dynamic contracting process built to survive systemic disruption. Together, these propositions and responses point toward a more explicitly systemic and multi-level conception of the insider researcher’s role than the current literature provides.