Ignorance of user factors can be seen as one of the nontechnical issue
s contributing to expert system failure. An expert advisory system is
built for nonexpert users; the users' acceptance is a very important f
actor for its successful implementation. If an expert advisory system
satisfactorily represents the expertise in tire domain, there still re
mains tire question: ''Will the end-users use the system?'' This paper
aims to address users' issues by analysing their reactions towards an
expert advisory system called ADGAME, developed to help its users mak
e better decisions in playing a competitive business game. Two experim
ents with ADGAME have been carried out. The research results show that
, when the use of the expert advisory system is optional, there is con
siderable reluctance to use it, particularly amongst the ''worst'' pot
ential users. Users also doubt the potential benefits in terms of impr
oved learning and confidence in decisions made. Strangely, the one pos
itive expectation that users had, that the system would save them time
, proved not to be the case in practice; ADGAME appears to improve the
users effectiveness rather than their efficiency.