Definition
A composite indicator of how much risk an institution carries by relying on a system, defined as dependency depth multiplied by substitution cost multiplied by correction latency. Depth is the number of dependents whose criticality is high or critical. Substitution cost is the estimate in the unit the dependency record states, such as person-weeks. Correction latency is in hours from a decision to intervene to the intervention taking effect. The score is read as a trend, and compared across deployments only where substitution cost is stated in the same unit.
It matters because the three factors are usually owned by different teams and never multiplied. A system can look acceptable on each factor alone and still leave the institution unable to act, which is exactly what dependency state tracking exists to surface.
Tests that show it is in place
- All three factors are measured from records rather than estimated in a workshop.
- The score is recomputed on a fixed cadence and its trend is reported, not only its level.
- A rise driven by any single factor triggers the same review as a rise driven by all three.
Evidence to keep
- Artifact: A scored dependency record showing depth, substitution cost, and correction latency inputs.
- Behavior: A quarterly recomputation whose rise opened a reversibility investment decision.
- Metric: Exposure score per deployment, trended quarter over quarter.
- Metric definition: Structural risk carried by an institution's reliance on one deployment.
- Metric unit: high- or critical-criticality dependents × the record's substitution-cost unit × hours
- Data source: Dependency record, substitution study, and intervention drill timings.
- Calculation: dependency depth (dependents whose criticality is high or critical) x substitution cost (in the unit the dependency record states) x correction latency (hours).
- Threshold: No absolute target; a rise of more than 25% quarter over quarter opens a review.
References