The software ecosystem has become a central conceptualisation for characterising the contemporary software business world. To understand and evaluate ecosystems, the concept of 'ecosystem health' was borrowed from the field of biology. In a 'healthy' ecosystem, the participants will flourish and, vice versa, suffer in an unhealthy one. Yet, there is a lack of empirical validations for the current approach as well as certain limitations regarding the concept. This paper will present a critique on current ecosystem health measurement and evaluation approaches. In addition, there is discussion on three proposals that could help to refocus the academic research on software ecosystem health.