Soil slips and slides are a common occurrence during heavy precipitation in many areas of the world. One of these places is the Ligurian region in northwestern Italy where, during intense precipitation in September 1993, more than 350 soil slips occurred in a small basin near the city of Genoa. Criteria for determining the occurrence of soil slips are usually based on estimates of average intensity and storm duration, to be compared with limiting values, usually represented by a line in a scatter diagram. These criteria, however, can only be used where a large amount of data makes the scatter diagram feasible; moreover, they do not take into account the degree of moistness of the soil. Thus they fail to explain some observed phenomena, e.g. soil slip occurrence under storms with lower total depth (i.e. lower average intensity) with respect to storms which did not produce soil slips. In this paper an alternative approach is presented in which the criteria for determining the occurrence of a soil slip is based on the comparison between soil slip producing and not producing events, their interpretation by using a conceptual model for infiltration, the use of a storm model; the final result is again a threshold line in an average intensity-duration diagram, with specified values of the initial moisture condition and of the return period of the storms which produce soil slips. The approach was tested in the said basin near Genoa with satisfactory results.