What’s the Harm? “I need the software in three months, so I’ll tell the development staff that I need it in two months. I don’t think they can actually deliver it in two months, but at least that will ensure that I get it in three months.”
This rationale is intuitive and appealing. But it is ultimately destructive to software project costs and schedules. Projects that could have been completed in three months end up taking four or five months because of the problems caused by such reasoning. Underestimation causes a project to be under scoped and under planned. That increases the number of mistakes that occur upstream. These mistakes must be corrected eventually— at much greater cost than if they had been corrected earlier. These mistakes erode cost and schedule and virtually eliminate mid-range to long-range predictability.
Skillful project planners strive for accurate estimates, and they especially strive to avoid underestimating.