Recent years have witnessed a growing concern for the ability of firms to effectively manage their new product innovation in the face of disruptive technological changes, increased global competition, and rising costs of research and development. These concerns notwithstanding, firms are additionally required to launch radical new products to the market, as incremental new products provide their developers with only short-term sales and profitability. In response to these challenges, firms have entered into collaborative alliances to share the risks and costs involved in the new product development (NPD) process and to enhance their product innovation performance. Turning research discoveries into marketable radical new products through collaborative alliances is even more important for relatively small firms operating in technologically intensive industries. Such firms are often underfunded and unable to undertake a full NPD cycle internally due to an inability of assembling the right mix of internal capabilities. The inevitable need to access capabilities from alliance partners may lead some small firms to form collaborative alliances under unfavourable situations, which make alliances prone to failure (70% by some estimates) to reach new product innovation goals. The substantial rate of alliance failure is embedded in a clash between the logic of radical new product innovation management (the need for flexibility between alliance partners), and recommendations for alliance management (the need to determine the responsibilities of each partner from the onset of the alliance). Despite the benefits of alliances in providing required resources, alliances can impose substantial transaction costs to focal small firms. Thus, it is crucial to investigate how firms, particularly small firms, can make a balance between the benefits and costs involved in alliances, to mitigate alliance risks and increase the probability of new product radicalness. In this thesis, I introduce a new typology and demonstrate its application to product performance. The typology categorizes alliance partnerships along two dimensions of partnership timing (the stage of the NPD process during which alliance is formed) and partnership type (the role of alliance partner during the NPD process). I use this this typology to determine the interaction effects of partnership timing and type on the probability of product innovativeness (radicalness). To this end, I rely on insights from Transaction Cost Economics (TCE) and Resource Based View of the firm (RBV) theories as well as the absorptive capacity concept to develop testable hypotheses. I use a sample of 230 drugs developed by 85 biotechnology firms in collaborative alliances with 384 alliances in 1982-2016 with universities and research institutes, other biotechnology firms, and pharmaceutical firms formed during discovery, development, and prelaunch stages of the new drug development process. I find that the probability of drug radicalness increases when alliances with universities and research institutes, as well as other biotech firms, are formed during the discovery or development stages of the new drug development. However, results indicate that partnership with pharma firms during the discovery or development stages reduces the likelihood of drug radicalness. During the prelaunch stage, except for negative relation between alliances with other biotech and drug radicalness, results failed to find a significant relationship between university as well as pharmaceutical partnership and drug radicalness. By disintegrating alliances along two dimensions of partnership type and timing, this thesis substantially increases the understanding of the benefits and costs of each partnership type and during each stage of the NPD process. This helps relatively small firms to better understand when and with whom during the process of NPD they need to initiate alliances to increase their likelihood of product radicalness. This thesis also contributes to the current theoretical insights of TCE and RBV theories by considering costs and benefits of each partnership type variant along different stages of the NPD process. Methodologically, instead of focusing on analysis using firm level outcome variables (count number of new products), this thesis turns the unit of analysis to product level (innovativeness of the product) and links each product to its designated alliance attributes (timing and type) to provide more subtle and fine-grained implications. / Thesis / Doctor of Philosophy (PhD)