Data Doesn’t Drive Decisions — Delivery Does

Data is frequently described as a driver of better decisions. In public-sector health programs, however, data alone rarely changes outcomes. Experience across multiple implementation and evaluation efforts suggests that delivery structures, not analytic products, determine whether information informs action.

While supporting a contract focused on monitoring and evaluation for a large public-sector health initiative, GPHS observed that analytic reports were produced on schedule and met technical requirements, yet were rarely used in program decisions. The issue was not data quality or analytic rigor. Instead, decision timelines, accountability structures, and operational roles were not aligned with when and how information was delivered.

This pattern is common across public-sector programs. Evaluation findings, surveillance summaries, and performance metrics often arrive after key decisions have already been made. When data functions independently of delivery processes, it becomes retrospective rather than actionable.

Programs that successfully integrate data into decision-making tend to share several characteristics:

• Analytic activities are embedded within delivery teams rather than isolated

• Outputs are designed around specific operational decisions

• Data interpretation incorporates program context and constraints

• Feedback loops allow findings to inform adjustments in real time

In response and operational settings, decision support works best when analytics are treated as part of the delivery system itself. This requires close coordination among analysts, program managers, and operational staff, as well as clarity around decision authority and use.

Experience across applied program delivery efforts demonstrates that strengthening the link between data and delivery does not require more dashboards. It requires intentional alignment between information, decision points, and operational accountability.