The use of group decision support systems (GDSS) is rapidly growing. One key factor in the effectiveness of these systems may be the manner in which users are supported in their use of this technology This paper explores two types of GDSS facilitative support: chauffeur-driven and facilitator-driven. In the former case, a person is used to reduce the mystique of the GDSS technology for users. In the latter case, a person assists the group with its group process in addition to reducing the mystique of the technology. The work unfolds a research story in which the original thinking of the research team to the effect that facilitator-driven GDSS facilitative support is superior is proven incorrect. The results of a pilot study caused the research team to reverse its thinking and hypothesize that, given the nature of the facilitation used and the task faced by the group, chauffeur-driven facilitation would have an advantage. The results of the experiment reported in this paper support this hypothesis. Arguments are presented to the effect that, to be effective in a judgment task environment, facilitation must be open and adaptive rather than restrictive.