Complexity has become a major keyword when it comes to organization issues in Construction and Real Estate Management. Expressed via many different theoretical definitions, complexity in terms of management represents the part of organization which is simply not manageable. Therefore, reduction of complexity by separating a system into independent well-defined and well-controlled subsystems becomes a major task. However, since construction and real estate projects are becoming larger, encompassing higher volumes as well as higher numbers of participants and are nonetheless subjected to strongly limited time-frames and budgets on tight markets, efficient organization developed into the crucial issue to stand a competition successfully. On this background, engineering naturally focuses on saving costly resources, where the explicit value of a single measure can easily be derived from the cost of the resource and the duration of the therewith reduced time floats. This leads to well-known concepts, e.g. just-in-time-delivery, where a system is optimized with respect to physical resources as well as virtual resources like storage space or reserve time. However, as this strategy clearly saves explicit local resources, concurrently the coupling of processes via the required availability of physical and virtual resources, ranging from pre-products to decisions, plans and responsibilities, is strongly increased and, thus, complexity is reintroduced to a significant degree. This paper proposes an approach on the basis of Systems Theory providing an explicit measure to evaluate the increase of complexity in relation to possibly saved resources. Since cost of complexity are not given a priori but result from possible deviations, this article investigates the propagation of virtual uncertainties of real and abstract pre-products through a network of given complexity. On this basis, some general rules are derived allowing to maintain the balance between saving resources and the therewith increasing cost of the consequently rising complexity.