Classical thermodynamics provides a powerful description of energy conservation, irreversibility, and entropy production, but it does not explicitly formalize information as a state-level quantity capable of governing the emergence and persistence of organization. In this article, I present a unified framework linking information, entropy, and system dynamics for a broad class of physical, biological, and computational systems. I argue that organization is not fully captured by entropy alone, and I formalize organization as a dynamical balance between usable information and effective entropy. I introduce (i) operational definitions of usable information and effective entropy that can be adapted across domains, (ii) a general organizational potential describing whether a system tends toward stable structure or toward disorder, and (iii) a dynamical formulation that predicts transitions between organization, dynamic equilibrium, and disorganization. I show how this framework recovers known results from statistical physics and nonequilibrium thermodynamics, while also providing a coherent language for biological maintenance of order and formachine learning dynamics. I conclude with testable predictions and a roadmap for empirical validation.