Decisions for key technologies, like middleware, for large scale projects are hard, because the impact and relevance of key technologies go beyond their core technological field. E.g., object-oriented middleware has its core in realizing distributed object calls. But choosing a technology and product also implies to adopt its services, tools, software architectures, object and component paradigms, etc. Moreover, legacy applications and several other key technologies have to be integrated. And since no middleware product serves all requirements in the enterprise context, various middleware products have to be integrated, too. Another key problem of middleware evaluation is, that often the studies have to be performed very early in a project. In this paper we try to tackle these problems and describe how we can communicate the outcomes - which come from a technical viewpoint - to the management and other non-experts in the technological field.