In striving for mass production of magnetocaloric materials for refrigeration applications, one has to face the fact that these will not be laboratory-grade materials, and will necessarily present (at the very least) composition gradients due to lower quality precursors. One can expect some probable consequences from this, namely that the maximum value of magnetic entropy change (ΔSM) will decrease compared to a pure material, but also some broadening of the ΔSM(T) curves should occur, as observed in some elementary ferromagnets. Some theoretical work has focused on this topic, for second-order phase transition systems via the Landau theory of phase transitions and the molecular mean-field model. Still, these theoretical considerations do not directly apply to first-order phase transition (giant magnetocaloric effect) systems. We here present a study on the effect of disorder on the magnetic and magnetocaloric properties of first-order phase transition systems. We employ the Bean–Rodbell model, and consider disorder to be described by a width of a Curie temperature distribution. We show how disorder effects “smooth” the discontinuities of magnetization and entropy change, and also affect magnetic hysteresis. We show how for sufficiently large disorder, the shape of the magnetic entropy curves approximate the distribution function. We discuss how the magnetic field dependence of magnetic entropy change is affected by disorder, in light of recent reports of “second-order like” dependence of magnetic entropy change on applied magnetic field, for disordered giant magnetocaloric effect La–Fe–Si-based samples.