One of the most widely spread techniques to estimate the compressive strength of concrete is the rebound hammer test, also known as Schmidt Hammer test. In spite of a large number of scientific works trying to calibrate the test, to identify the parameters affecting its results and to estimate its reliability, the original Schmidt curve is still provided by the producers along with the hammer and is used in Structural Engineering Applications. This paper discussed an extensive research, and application, of this technique to a large number of cubes provided by the Laboratory for Building Materials of the University of Genoa, Italy, showing that several phenomena strongly affect the test: moisture content, maturity, stress state among the others. Strength estimates may differ as much as 70% if these parameters are not taken into account. Besides, several in situ investigations on existing buildings were affected by a large dispersion of data, so that we should conclude that the Rebound Hammer is unable of giving a reliable estimate of the concrete strength. This is probably due to the very limited area of the material on which the test is performed that allows also small local inhomogeneity to affect quite strongly the test. Therefore, the rebound hammer seems to be useless in the estimation of concrete compressive strength, being only a rough tool for estimating material homogeneity inside a specific concrete type.