The COVID-19 pandemic necessitated urgent solutions to support governmental programs that were critical to mitigating unacceptable economic, social, and health risks. We conducted a multiple case study, confirmation workshop, and survey of Norwegian public institutions that had defied conventional wisdom about public sector inertia by successfully completing important, urgent, and unexpected development efforts (do-or-die projects). We found consistent patterns credited for their accomplishments, summarized as follows: (1) antecedent capabilities for collaboration, use of expertise, and interdisciplinary problem-solving; (2) characteristics of the project mandates including specificity of goals, importance, deadlines, and tie-in with societal mission; (3) emergent development practices that empowered a core team to make decisions, benefit from organizational support, and integrate effectively with diverse stakeholders; and (4) outcomes that reinforced the practices, such as success, pride in work, transparency, and continuous capability development. We examined our findings through the theoretical lenses of complexity leadership and goal-setting and found support for the relevance of do-or-die projects in both non-urgent and urgent contexts. The findings contribute to understanding how the public sector can perform under pressure by mobilizing latent capabilities and responding with flexibility and focus. Implications are outlined for diverse fields of agile practices and principles, organizational resilience, public sector interaction, complexity leadership, goal-setting, and project studies. Our study suggests that successful do-or-die projects result from a combination of contextual alignment, clear mandates, and empowered execution — offering insight into how such performance may be fostered in and beyond crises. Editor’s note: Open Science material was validated by the Journal of Systems and Software Open Science Board .

The uptake of practices labeled as “agile” is widely discussed in both research and practitioner communities, yet the scope, prevalence, and implementation of such practices remain unclear. A field-level hypothesis suggests that agility manifests in three distinct but related disciplines: software development, product orientation, and continuous delivery. This study tested that proposition through surveys of IT executives in the Norwegian public sector, which represents a substantial share of the country’s national digitalization efforts. Surveys were conducted in 2021 (136 invited, 78 responses) and 2024 (146 invited, 58 responses). The results suggest that agile practices are prevalent and perceived as growing, but systematic adoption patterns are not apparent. No stable factor structures emerged for the hypothesized disciplines, suggesting that agility manifests instead as a multifaceted, emergent phenomenon. Adoption appears to be enabled by team-level commitment and individual initiative, while constraints related to budgeting and governance inhibit it. These findings contribute to theory by questioning assumptions that agile practices are organized into stable domains, method by demonstrating the limitations of current indicators and survey approaches, and practice by underscoring that successful adoption requires enabling organizational conditions rather than prescriptive frameworks. Future research should focus on clarifying what constitutes an agile practice, developing robust indicators of adoption, and comparing implementation patterns across various sectors and countries.