Tracking Adoption of Research Recommendations: the Recommendation-Adoption Score

Why Read This from an HCI Perspective

This piece is meaningful for HCI/UX practitioners because it attempts to quantitatively track the path from research findings to actual product changes. Typically, the impact of research can fade after the presentation, but RAS reveals where recommendations get lost and structures definitions of status, ownership, and value—helping diagnose collaboration bottlenecks. The key is that it helps you see ‘delivered value,’ not just ‘good insights.’

CIT's Commentary

From a CIT perspective, RAS is not merely an operational metric; it’s a governance tool that exposes translation loss between research, design, and development. In particular, treating recommendations like an inventory and clearly separating statuses into Adopted/Committed/Communicated is practically effective—it reduces the illusion created by vague ‘in progress’ states. However, if this metric becomes a performance indicator used to evaluate teams, it could backfire by increasing easy-to-score tweaks rather than high-value work. That’s why CIT argues that RAS should be run as a way to diagnose system health—not as a tool for assigning blame. As the authors emphasize, trends matter, and you need an interpretive stance that separates researchers’ quality from an organization’s execution capability.

Questions to Consider While Reading

• Q.To keep RAS as a system-diagnosis metric rather than a team-evaluation metric, what operating principles and review cadence would be needed?
• Q.How well does the approach of assigning 0.66 points to Committed actually distinguish between ‘promised value’ and ‘realized value’?
• Q.In environments like Korean product organizations—where decision-making authority is distributed—what is the most realistic way to design clear ownership for recommendations?