Skip to main content
. 2023 Oct 13;4(10):100842. doi: 10.1016/j.patter.2023.100842

Table 2.

Design process

1. Understand the system being measured, including both technical55 and organizational20 considerations.
Determine scope
  • What is included in the system?

  • What will the metrics be used for?

Understand the causal structure of the system
  • What is the logic model or theory?56

  • Is there formal analysis57 or expert opinion58 that can inform this?

Identify stakeholders59
  • Who will be affected?

  • Who will use the metrics?

  • Whose goals are relevant?

2. Identify the goals
  • What immediate goals are being served by the metric(s)?

  • How are individual impacts related to performance more broadly?60

  • What longer-term or broader goals are implicated?

3. Identify relevant desiderata
  • Availability

  • Cost

  • Immediacy

  • Simplicity

  • Trust

  • Fairness

  • Non-corruptibility

4. Brainstorm potential measures for metrics
  • What outcomes are important to capture?

  • What data sources exist?

  • What methods can be used to capture additional data?

  • What measurements are easy to capture?

  • What is the relationship between the measurements and the outcomes?

  • What is not captured by the measurements?

5. Consider and plan
  • Understand how measurements will be used to build metrics.61

  • Consider how the metrics will be used to diagnose issues or incentivize people.30

  • Consider the use of soft metrics to triangulate.41

  • Consider avoiding the “reward/punish” dichotomy.62

  • Identify and mitigate likely failure modes with pre-mortems.34

6. Plan to revisit the metrics
  • Set a specific date or interval for routine re-evaluation.

  • Identify additional triggers for non-routine re-evaluation.