Hydroprocessing stabilization has been assessed using experimental data acquired at transient conditions. These data were obtained from a hydrotreating pilot plant in a wide range of operating conditions. It has been found that the stabilization evolution follows a first-order response and could be characterized by a stabilization time τ. A linear model was developed to relate τ to its most influential parameters. The model can then be combined with online transient data to predict the steady-state performance. By testing against new data with other feedstocks, the model has been found to provide a good prediction of the stabilization evolution and the ultimate steady-state hydrotreating performance. It is, hence, possible to “online” calculate the steady-state performance without the need to reach this steady state.