Nowadays, more and more, the monitoring of concrete's setting and hardening as well as concrete's condition assessment and mechanical characterization is realized with the Ultrasonic Pulse Velocity technique. However, despite its increasing use, the high potential and the vast applicability over a wide range of materials and structures, the aforementioned nondestructive testing technique is only partially exploited since a) a default pulse usually not selected by the user is transmitted, b) a single frequency band dependent on the testing equipment (pulse generator and sensors) is excited and c) usually the first part of the signal is only considered. Moreover, the technique, as defined by its name, is based on pulse velocity measurements which strongly rely on a predefined threshold value for the calculation of the travel time between the transmitting and receiving sensor. To overcome all these issues, in the current experimental campaign, user-defined signals are generated, a broad range of ultrasonic frequencies is excited, while the full length of the signal is also taken into account. In addition, the pulse velocity measurements are replaced by the more advanced phase velocity calculations determined by reference phase points of the time domain signals or by phase differences of the signals transformed in the frequency domain. The experiments are mainly conducted in hardened concrete specimens but the aggregates are substituted by spherical glass beads of well-defined sizes and contents in order to better control the microstructure. Reference liquid media are also examined for comparison purposes. The results in both cases show strong dispersive trends indicated by significant changes in the phase velocity.