The Hidden Cost of Over-Specified Requirements: Evidence from 40 Programs
A meta-analysis of 40 systems engineering programs finds that over-specification — requirements that constrain solution space unnecessarily — correlates more strongly with schedule overruns than under-specification. The study proposes a requirements specificity index and calibration approach.
Over-Specification Is Costing You More Than Under-Specification
The conventional wisdom in requirements management is to be as specific as possible. This study challenges that assumption with data from 40 programs.
The finding: Programs with requirements specificity scores in the 60–75th percentile had the best schedule and cost outcomes. Both under-specified and over-specified programs performed worse.
Why over-specification hurts: Unnecessarily constrained requirements limit contractor design freedom, increase change request volume as the design matures, and create brittle V&V matrices that require frequent revision.
Calibrating specificity: The paper proposes distinguishing "interface requirements" (high specificity justified), "performance requirements" (medium specificity, focus on what not how), and "derived requirements" (lowest specificity, allow design latitude).
Practical guidance: Review your requirements set for "how" language masquerading as "what" requirements. Each one is a potential unnecessary constraint.