Comparative health systems research relies principally on two instruments, typological classification and indicator benchmarking, both of which observe systems from the outside. Both establish reliably that systems differ. Neither explains well why the same institutional instrument behaves differently in different settings. This reflective paper argues that a substantial part of the missing explanatory evidence is held by the people who operate these systems, and that the field does not systematically collect it. Writing from three vantage points that do not agree with one another, academic, commercial and lay governance, the paper identifies three areas in which practitioner knowledge appears to be load-bearing rather than merely illustrative: risk equalisation understood as a governance practice rather than an actuarial formula, purchasing leverage understood as a relational phenomenon, and sub-national institutional variation, particularly across Swiss cantons. It sets out the obvious objections to treating practitioner accounts as evidence, argues that the remedy is systematic collection and triangulation rather than deference, and locates the argument within the field's own longstanding self-criticism. It closes by noting that the three countries are currently renegotiating their respective settlements simultaneously and in public, and that the design logics of those reforms are observable now in a way they will not be in five years.