Data has become an increasingly important resource in professional esports, supporting match preparation, tactical discussions, and performance assessment. In practice, however, esports analytics remains fragmented. Access to gameplay data differs strongly between games, tools rarely integrate well; and analysts still rely heavily on manual workflows to collect, process, and interpret information. At the same time, advances in sensing technologies enable the collection of psychophysiological signals such as heart rate variability, electro dermal activity, and eye tracking, which may offer additional insight into player focus and stress. Despite this potential, their role in everyday coaching practice remains unclear. This chapter presents findings from the BEAT (Better Esports Analytics for Teams) project, an applied research project examining how analytics is currently used within professional esports contexts. Drawing on interviews with analysts, coaches, players, and analytics providers, combined with an analysis of existing platforms, we identify recurring challenges related to data fragmentation, limited automation, lack of standardization, and concerns around privacy and trust. Based on these observations, we outline design implications for future esports analytics systems that better align with team workflows and support actionable decision-making in competitive contexts.