The recovery of offshore wind farm wakes in the German Bight was analyzed by a unique in situ data set, measured on‐board the research aircraft Dornier Do‐128 during the WIPAFF project in 2016 and 2017. These observations were used to validate a simple analytical wake recovery model in five case studies. The observed recovery rates were compared with the results of the mesoscale Weather Research and Forecasting (WRF) model. The airborne data show that the wake recovery can be described by an exponential function as expected by the analytical model and strengthens the hypothesis that the vertical downward moment flux has an important influence on the wake recovery. However, the predicted wake recovery rates (by the analytical model) do not always fully agree with the observations. Although, as a first‐order approximation, the model seems to perform well, further optimization has to be implemented to account for wind park layout, turbine induced turbulence, and horizontal momentum flux. WRF simulations reveal an exponential recovery, although the mesoscale model does not reproduce the correct atmospheric conditions for most of the cases. Therefore, the wake recovery rates estimated by WRF disagree with the measured data in most of the studied cases.