In best-value procurement, current practice shows that cost is frequently more influential than non-cost factors, and consequently, the lowest bidder is chosen in most of the cases; thus, a best-value selection is not achieved. Designbuilders cannot offer the best-value in their proposals if evaluation criteria do not show precisely what constitutes best-value and how best-value is scored. Thus, the aim of this research is twofold: first, to identify how highway agencies articulate evaluation criteria; and second, to propose a structured approach that enhances current practice on writing evaluation criteria. Through the lens of decision analysis, the researchers conducted a content analysis on 540 evaluation criteria included in 98 requests for proposals (RFPs) from 21 states across the United States (US). The study showed that 43% of evaluation criteria were generic, 53% used a generic constructed scale, and 4% assigned points or levels directly. These three groups represent different levels of specificity in writing evaluation criteria.Building upon these levels and on decision analysis theory principles, this research proposes a structured approach to support highway agencies in the process of crafting evaluation criteria. More precise and specific evaluation criteria