Cone penetration tests, CPTs, are extensively used in the Netherlands to assess the stability of fourteen thousand kilometres of dykes protecting the country from flooding. On the regional dykes, site testing is planned and executed only from spring to autumn. The data collected in the drier season of the year must be used then in safety factor calculation for dyke stability with reference to the worst expected conditions, including the highest weights and the highest water pressures over the year. Inferring reliable values of the shear strength in a different season implies understanding the unsaturated response of the dyke material and the effect of variable water content on the CPT response. In previous studies referring to CPTs in unsaturated soils, it was observed that both the cone resistance and the sleeve friction depend on suction, however, only the cone resistance was used to determine the shear strength in combination with water content or suction probes installed into the ground. In this contribution, we analyse an extensive set of data, coming from repeated CPTs performed over one year on the Maasdijk near Oijen in the Netherlands. The data are elaborated to investigate whether the entire set of data can be exploited to try to derive the water content and the constant water content shear strength at the same time, if the test is repeated in different seasons.