An Interpretable Framework for Startup Success Assessment Using Entrepreneurial Ecosystem Intelligence

Abstract

Entrepreneurship Support Organisations (ESOs) need metrics to evaluate ventures within entrepreneurial ecosystems, but assessing startup readiness remains difficult. Current practice relies mainly on descriptive indicators that are hard to measure and to reproduce, and little is documented about how these indicators are operationalised or how well they perform. The Open Innovation Framework for Pitch Competitions (OIF-PC) (Movahedian Attar et al., 2025) takes a systems-thinking view in which entrepreneurial support is treated as a funnel whose design parameters shape venture development. Its constructs, however, are defined at programme level and have not yet been expressed as measurable attributes. Expressing them at venture level would give programmes a more useful basis for evaluation.

This study therefore operationalises the OIF-PC orientation as measurable attributes of startups and examines whether these attributes are associated with observed outcomes. A secondary dataset of 923 records describing 922 unique ventures, with 45 raw attributes covering funding, milestones, networks, and the ecosystem, was analysed using Logistic Regression, Random Forest and XGBoost. SHAP (SHapley Additive exPlanations) was used to interpret model outputs and to quantify the contribution of original and derived variables to the assessed readiness score.

Access to resources is associated with more favourable outcomes. How those resources are used, and how actively a venture engages with its entrepreneurial environment, account for a comparable share of the model's attribution, with network connectivity the single most influential attribute. This indicates that ecosystem constructs of the OIF-PC type can be operationalised through venture-level variables, but it is not a validation of the framework itself. On this basis, the study introduces the Startup Success Readiness Index (SSRI), which combines four pillars (Relational Strength, Milestone Achievement, Investor Diversity and Funding Efficiency) into a weighted score using SHAP-derived attribution. The SSRI gives ESOs a transparent readiness measure that they can use to profile ventures, find capability gaps and inform programme design.

Keywords: Startup Success Assessment; Entrepreneurship Support Organisations (ESOs); Open Innovation Framework for Pitch Competitions (OIF-PC); Machine Learning; Startup Success Readiness Index (SSRI); SHAP