Pyroelectric and electrocaloric characterization has been determined for $0.75\mathrm{Pb}\left({\mathrm{Mg}}_{1∕3}{\mathrm{Nb}}_{2∕3}\right){\mathrm{O}}_3–0.25\mathrm{Pb}\mathrm{Ti}{\mathrm{O}}_3$ relaxor based single crystal and ceramic. Differential scanning calorimetry was used for measuring the electrocaloric response for different electric fields in the vicinity of the Curie temperature. For both ceramic and crystals the maximum activity is found to be around the transition temperature. On the other hand hysteresis loops for different temperatures were used to predict the electrocaloric effect with very good qualitative agreements with direct measurements. Pyroelectric coefficient is found to be much larger for ⟨111⟩ single crystals reaching $1300\times {10}^{-6}\mathrm{C}{\mathrm{m}}^{-2}{\mathrm{K}}^{-1}$ whereas the ceramic reaches only $750\times {10}^{-6}\mathrm{C}{\mathrm{m}}^{-2}{\mathrm{K}}^{-1}$ . Higher pyroelectric coefficient and lower dielectric permittivity lead to outstanding figures of merits for sensors and energy harvesting, with a gain of 260% for voltage responsivity and more than 500% for energy harvesting. Although having a much larger pyroelectric activity, the electrocaloric effect is about the same for crystals and ceramics—around $0.40\mathrm{J}∕\mathrm{g}$ for $2.5\mathrm{kV}∕\mathrm{mm}$ electric field step. This result is interpreted by the decrease of the pyroelectric coefficient for high electric field. The electrocaloric activity is in fact limited by the saturation polarization and difference between Curie transition temperature and the working temperature. Those two parameters are very similar for crystals and ceramics. Single crystals are consequently very interesting materials in the framework of energy harvesting and sensor applications whereas no real improvement of performances can be expected for electrocaloric refrigeration devices.