Gravitational waves have been directly detected by the LIGO experiment in 2015. Since then, five black holes and one neutron star binaries merging have been observed during the two observational runs. The measured signals already provided a large amount of physical results, from the mass distribution of stellar-masses black holes to the short gamma-ray burst mechanisms through measurement of the expansion rate of the Universe and tests of general relativity. Both detection and analysis of gravitational waves rely on the simulation of the expected signal to be extracted from the noise. As the exact computation of gravitational waveforms is a computationally expensive resource, physicists are building phenomenological models. I will review the process of designing such models as well as their impact on the physics extracted from the signals.