Objective: Background and Objective: Technology-based startups play a significant role in entrepreneurial ecosystems due to their contribution to innovation development, knowledge commercialization, specialized employment, and economic transformation. Nevertheless, a considerable proportion of these businesses fail during the early stages of their operations. The failure of technology-based startups is a complex and multidimensional phenomenon that cannot be attributed solely to insufficient financial resources, managerial weaknesses, or lack of market access, as institutional, ecosystem, strategic, financial, and market-related factors interact with one another and may reinforce each other through feedback mechanisms. Accordingly, this study aims to identify and model the causal structure of factors affecting the failure of technology-based startups within the entrepreneurial ecosystem of Markazi Province, Iran.


Method: In terms of purpose, this study is applied and developmental, while in terms of nature and methodology, it adopts a mixed-methods (qualitative–quantitative) approach. In the qualitative phase, data obtained from semi-structured interviews were analyzed using thematic analysis and coded with MAXQDA software. In the quantitative phase, the causal relationships among the identified factors were analyzed using the Fuzzy Cognitive Mapping (FCM) approach and modeled using Mental Modeler. The data collection instrument was in-depth, semi-structured interviews. The study was conducted during the spring and summer of 2026. The statistical population consisted of technology-based startups located in the Markazi Province Science and Technology Park. Purposive and judgmental sampling was employed, and interviews were conducted with 24 startup managers and experts until theoretical saturation was reached. In addition, a pairwise comparison questionnaire was used to examine the causal relationships among the identified factors and determine the strength of their effects using fuzzy values. Subsequently, a fuzzy relational matrix was constructed, and the centrality, influence, and dependence of the variables were analyzed. MICMAC analysis was then conducted to identify driving, linkage, and dependent variables, while dynamic simulation was employed to examine the behavior of the system over time. Sensitivity analysis and scenario analysis were also conducted to identify leverage points and assess the consequences of alternative interventions.


Results: Overall, 18 key variables were incorporated into the model and organized into five overarching themes: unfavorable institutional atmosphere, provincial ecosystem deficiencies, strategic incapacities, market interaction deadlock, and intensifying events. The results indicated that the failure of technology-based startups is not a linear process but rather the outcome of a network of interactions among multiple factors. Macroeconomic turbulence, monetary and banking instability, and geographical isolation were identified as driving factors. In contrast, capital imbalance, industry trust deadlock, and local brain drain were classified as linkage and critical factors and emerged as particularly important leverage points within the system. Furthermore, incorrect pricing, validation failure, and resistance to adoption were primarily positioned as outcome variables within the network. The results of the dynamic simulation showed that, if the existing conditions persist, the system moves toward an undesirable equilibrium state, with variables such as resistance to adoption, validation failure, incorrect pricing, capital imbalance, and industry trust deadlock stabilizing at relatively higher levels. The findings also demonstrated that interventions focused solely on macro-level factors are insufficient to break the failure cycle and that simultaneous interventions targeting leverage points at the financial, market, industry, and human-capital levels are necessary.


Conclusion: The findings indicate that the failure of technology-based startups in Markazi Province should be understood as a systemic, multilevel, and self-reinforcing phenomenon. The main implication of the study is that isolated and short-term support policies, such as financial assistance or general training programs, are unlikely to generate sustainable effects unless the interrelationships among capital, validation, industry trust, market access, and human capital are addressed simultaneously. Accordingly, the development of stage-based financing mechanisms, strengthening startup–industry linkages, establishing systematic market-validation mechanisms, improving pricing practices, retaining human capital, and developing an early-warning system for startup failure are recommended as key practical interventions. The proposed model can serve as a useful tool for diagnosing systemic problems, identifying leverage points, and supporting the decision-making of policymakers and managers within the startup ecosystem of Markazi Province.