We study the possibility of identifying dark matter properties from direct (XENON100) and indirect (GLAST) detection experiments. In the same way, we examine the perspectives given by the next generation of colliders (ILC). All this analysis is done following a modelindependent approach. We have shown that the three detection techniques can act in a highly complementary way. whereas direct detection experiments will probe efficiently light WIMPs, given a positive detection (at the 10% level for mχ ∼ 50 GeV), GLAST will be able to confirm and even increase the precision in the case of a NFW profile, for a WIMP-nucleon crosssection σχ-p < 10-8 pb. However, for heavier WIMP (∼ 175 GeV), the ILC will lead the reconstruction of the mass.