This research examines the impact of two potential antecedents of growth in high-technology, network industries. The first antecedent, installed base, is traditionally framed as a key driver of growth in these settings. The second, product quality, is thought to have only random influences on growth. This work tests the influence of both of these variables on growth in the packaged application software industry, and proposes that their relative influence may be contingent upon the network intensity of a given market or segment. In turn, installed base and timing of product release are proposed to have a significant impact on quality. The sample for this study encompassed five segments of the packaged application software industry from 1986-1998, including word processing, spreadsheets, desktop publishing, CAD, and personal finance. Several results from the empirical analysis offer useful insights into the nature of network industries, and implications for strategic management in these domains. First, installed base size exhibits a negative relationship with growth that becomes more positive as size increases. Second, product quality is positively and significantly associated with installed base growth. Third, the impact of size on growth is shown to vary across industry segments, suggesting that the influence of network effects may not be as homogeneous as extant theory implies. Finally, installed base is associated with higher baseline product quality within periods, consistent with the notion that a large installed base confers learning-based advantages to quality. Together, these findings indicate that competitive dynamics in network industries may be more complex than previously thought. This research extends existing theoretical and empirical work on network effects and positive feedback, and suggests new avenues for effective strategies in emerging high-technology settings.