We each have a self-concept that includes multiple, interconnected identities which form and evolve over time. These identities can become differentially active, or salient, and in doing so guide our behavior in a seemingly predictable way. Although identity researchers know how to make identities salient, and understand how identities should guide behavior when salient, we currently struggle to accurately predict when a particular identity will drive a consumer’s real-world behavior. This issue stems from the unnatural interventions that identity researchers typically use in their studies to make identities salient. The author reviews the current state of identity salience research, and argues that we need to embrace subtle interventions; practical methods to predictably make identities salient that are simple for marketers to use in a natural setting. This will enable us to study a wider range of identities, better understand the process by which particular identities become more (or less) salient, and be more useful to practitioners.