For many software-intensive embedded systems, the cost of the system over its deployed life will be dominated by software-related costs. This is especially the case in high reliability systems, as software plays a key role in determining system failure rates and the effective cost of software-induced failures in these systems can be very high. Unfortunately, important decisions are often made early in system development projects without adequately considering the impact that these decisions will have on lifetime system costs. This paper shows how to use freely available, industry standard cost and quality estimation models such as COCOMO II and COQUALMO to quantitatively estimate software cost, schedule, required staffing levels, and defects introduced and removed over the life of a system.