Despite actionable insight being widely recognized as the outcome of data analytics, there is a lack of a systematic and commonly-agreed definition for the term. More importantly, existing definitions are generally too abstract for informing the design of data analytics systems. This study proposes a definition of actionable insight as a multi-component concept comprising analytic insight, synergic insight, and prognostic insights. This definition is informed by a conceptual framework, which also can be used to systematically understand actionable insight, both at the concept-level and component-level. Each component is explained from the analytical, cognitive, and computational perspectives and relevant design considerations are suggested. We hope this study could be a rudimentary step toward the realization of decision-centric data analytics that can deliver the promised actionable insight.