We give the results from an evaluation of the dialogue manager of a developing, Slovenian and Croatian spoken dialogue system for weather-information retrieval. The results are based on two Wizard-of-Oz experiments and the application of the PARADISE evaluation framework. The only difference between both experiment settings is in the dialogue-management manner, i.e., while in the first experiment dialogue management was performed by a human, the wizard, in the second experiment it was performed by the newly-implemented dialogue-manager component.