Strategic Complement Certification in Platform-based Markets : evidence from the Mobile App Industry

In many platform-based markets, thousands of third-party complementors offer their products. To align their activities and augment the value of their ecosystems, platform owners have several strategic options. In this dissertation, I study an empirically prevalent strategy for governing platform-based markets: the selective endorsement of complements via certification. Previous research has found that complement certification increases platform value by reducing customers search costs and supporting high-potential products, which may otherwise go unnoticed. However, there are critical knowledge gaps regarding the strategic role of certification in platform-based markets, particularly how platform owners select which complements to endorse, what factors influence these decisions, and how the complementor ecosystem responds to certification. Using a dataset of app developer populations from Apples App Store, I combine and extend prior work on platform governance and certification to address these gaps. First, I employ a novel theory-building approach based on machine learning methods to detect patterns in the platform owners selection criteria. I find certification is awarded not only to encourage users to discover new apps but also to induce desired behaviors among developers. Second, I examine how the complementor ecosystem responds to certification. Using a neural-network-based text-embedding algorithm, I cluster similar apps in their respective app category. I find that certification has a generative effect on developers, which increases their innovation efforts after a competing app has been certified. My results also suggest that developers adjust their pricing strategies in response to certification. However, these effects vary across app categories, contingent on their level of competitive crowding. Together, these studies build a framework for future research studying certification as a strategy to govern complementor activity on platform-based markets. I discuss the implications of my findings for the literature on platform governance, certification, and studies using machine learning methods in strategy research.

In many platform-based markets, thousands of third-party complementors offer their products. To align their activities and augment the value of their ecosystems, platform owners have several strategic options. In this dissertation, I study an empirically prevalent strategy for governing platform-based markets: the selective endorsement of complements via certification. Previous research has found that complement certification increases platform value by reducing customers search costs and supporting high-potential products, which may otherwise go unnoticed. However, there are critical knowledge gaps regarding the strategic role of certification in platform-based markets, particularly how platform owners select which complements to endorse, what factors influence these decisions, and how the complementor ecosystem responds to certification. Using a dataset of app developer populations from Apples App Store, I combine and extend prior work on platform governance and certification to address these gaps. First, I employ a novel theory-building approach based on machine learning methods to detect patterns in the platform owners selection criteria. I find certification is awarded not only to encourage users to discover new apps but also to induce desired behaviors among developers. Second, I examine how the complementor ecosystem responds to certification. Using a neural-network-based text-embedding algorithm, I cluster similar apps in their respective app category. I find that certification has a generative effect on developers, which increases their innovation efforts after a competing app has been certified. My results also suggest that developers adjust their pricing strategies in response to certification. However, these effects vary across app categories, contingent on their level of competitive crowding. Together, these studies build a framework for future research studying certification as a strategy to govern complementor activity on platform-based markets. I discuss the implications of my findings for the literature on platform governance, certification, and studies using machine learning methods in strategy research.