This paper and another one preceding it investigate the potential of computer technologies like Geographical Information Systems (GIS) and Expert Systems (ES) to help with Impact Assessment (IA), suggesting that one way to optimise the effectiveness of GIS is to embed these systems in a more user-friendly environment. This paper explores the experience and potential of ES to provide such decision support environment, extending the argument further into the realm of Decision Support Systems (DSS). The proposition behind this paper is that these areas (IA, GIS, ES and maybe DSS) are potentially complementary and there can be mutual benefits if they are brought together. Certain tasks in IA — like screening, scoping, or statement review — already have potential for substantial automation, as an opportunity for technology transfer within organisations involved with IA, so that relatively scarce experts can be freed up for more difficult jobs while less expert staff are promoted to answer more sophisticated questions like these. More technical tasks like impact and environmental prediction also show scope for some degree of modelling-based automation, while other less routinised and more open-ended tasks involved with the management of the whole IA process probably lend themselves better to automation of a different kind, which works more as an "aid" than a substitute for the expert, for which the more flexible DSS seem more appropriate.