Children's ability to master skills for coordinated conversations is crucial for their healthy social and cognitive development. However, until recently, researchers lacked the tools to adequately characterize this complex phenomenon whereby children must learn to coordinate across multiple levels, use multimodal signaling, and adapt to their communicative cultural conventions and the diversity of conversational contexts. This paper proposes that the community capitalize on new technological opportunities — namely, unmoderated online data acquisition methods and high-scalability Machine Learning tools — to build a quantitative framework that will not only provide a deeper understanding of how children grow to become competent communicators but also constitute a theoretical foundation for applications in health, education, and child-oriented AI.