Many of our theories of international politics rely on microfoundations. In this short note, I suggest that although there has been increasing interest in microfoundations in international relations (IR) over the past 20 years, the frequency with which the concept is invoked belies a surprising lack of specificity about what microfoundations are, or explicit arguments about why we should study them. I then offer an argument about the value of micro-level approaches to the study of conflict. My claim is not that all theories of IR need to be developed or tested at the micro-level in order to be satisfying, but rather, that many of our theories in IR already rest on lower-level mechanisms—they either leave these assumptions unarticulated or fail to test them directly. In these circumstances, theorizing and testing micro-level dynamics will be especially helpful. I illustrate my argument using the case of resolve, one of the central explanatory variables in the study of international security. I argue that the absence of microfoundations for resolve is one reason why IR scholars have had difficulties testing whether resolve has the effects we often claim, and sketch out a two-stage research design political scientists can use to study unobservable phenomena.