Kit Fine (2017) distinguishes between inexact and exact truthmaking. He argues that the former can be defined from the latter, but not vice versa, so truthmaker semanticists should treat the exact variety of truthmaking as primitive. I argue that this is mistaken. We can define exact truthmaking in terms of inexact truthmaking and we can't define inexact truthmaking in terms of exact truthmaking. I conclude that it's inexact truthmaking, rather than exact truthmaking, that truthmaker semanticists should treat as the primitive semantic relation.