Software Design as an Investment Activity: A Real Options Perspective

Key software design principles‐e.g., information hiding, spiral processes, and guidelines for timing design decisions—remain idiosyncratic, ad hoc, and not unified or explained in theory. This makes them hard to teach, learn, and use in principled ways. We see the potential to build stronger foundations for softwar e design by viewing it as an investment activity and by relating design concepts direct ly to well developed financial investment theories. In this paper we focus on the specific is sue of the value of flexibility in the face of uncertainty. Uncertainty puts a premium on flex ibility to change products and plans, but flexibility also incurs costs. The theories on which earlier approaches to software engineering economics were based—primarily static net present value—cannot account for the value of flexibility. Real options theory, wh ich seeks to make this value tangible in capital investment through an analogy to financi al (e.g., stock) options, provides a framework within which we analyze software design concepts. We sketch our interpretation of important design principles in real opti ons terms, making the idea precise for the specific case of optimal timing of design decisio ns. We discuss the validity of the approach for software; impediments to quantitative application; and how qualitative options thinking can nevertheless improve design decision making.