The data presented in this chapter demonstrates two things.
Given the frequency of requirements errors and the multiplicative effect of the "cost to fix" factor, it's easy to predict that requirements errors will contribute the majority ”often 70 percent or more ”of the rework costs. Since rework typically consumes 30 percent to 50 percent of a typical project budget [Boehm and Papaccio 1988], it follows that requirements errors could consume 25 percent to 40 percent of the total project budget!
Our own experiences support this data, and that is the primary reason we wrote this book. If, with a small investment in a few key skills, we can do a better job in this area, we can save significant amounts of money, increase productivity, save precious time on the project calendar, and ultimately deliver higher-quality results to the customer, not to mention saving some of the wear and tear on the software team.