Abstract
Youth entrepreneurship is a critical component of economic development in Africa, yet the factors enabling youth-led ventures to scale within entrepreneurial ecosystems remain inadequately understood. Existing literature often lacks a longitudinal perspective on venture growth trajectories. This paper examines the mechanisms through which youth-led ventures in Kenya achieve scalability, analysing the interplay between venture characteristics, ecosystem support structures, and founder capabilities over time. A longitudinal, mixed-methods design was employed. Quantitative data on venture performance metrics were collected from a panel of 120 youth-led enterprises at three intervals. These were triangulated with in-depth qualitative interviews conducted with a stratified subsample of 30 founders and 15 ecosystem stakeholders. Analysis revealed that ventures achieving significant scale (defined as a threefold increase in revenue and employment) were consistently characterised by founders' strategic use of blended finance and active mentorship. Notably, 65% of scaled ventures had accessed both grant and debt financing, compared to only 22% of non-scaled ventures. Scalability is not serendipitous but is catalysed by a deliberate combination of financial strategy, targeted human capital development, and embeddedness within formal and informal support networks. The ecosystem's role evolves from providing basic access to facilitating strategic partnerships. Policymakers and development finance institutions should design integrated support packages that combine staged financing with structured mentorship. Microfinance providers should develop financial products specifically tailored to the growth phases of youth-led businesses. Youth entrepreneurship, venture scalability, entrepreneurial ecosystems, longitudinal study, Kenya, blended finance This study provides novel longitudinal evidence on the sequential process of scaling for youth-led ventures, introducing a phased model of ecosystem engagement that moves beyond static analysis.