The five PII submetrics
PII.ST — Security track record · 25%
Measures resistance to exploits and failures together with detection, containment, remediation, recurrence and disclosure quality. No reported incidents does not automatically mean strong security.
PII.CD — Consensus and control decentralisation · 20%
Measures effective control rather than nominal counts, using architecture-appropriate evidence such as stake, work, block production, sequencer authority, emergency keys and governance power.
PII.ID — Infrastructure diversity · 15%
Measures correlated failure risk across clients, hosting, geography, operating systems, RPC dependencies, oracles, bridges and critical maintainers.
PII.AF — Availability and finality · 20%
Measures whether the network reliably produces and finalises valid state, including halted time, missed blocks or slots, finality delay, reorganisations, degraded operation and recovery.
PII.GI — Governance and upgrade integrity · 20%
Measures whether protocol change is transparent, reviewed, constrained and recoverable, including specifications, testing, timelocks, emergency powers, key management and rollback procedures.
PII is not sentiment
Edition 3.1 replaced the earlier Public Trust Index with Protocol Integrity Index because the current framework prioritises objective protocol evidence over subjective reputation or sentiment.
PII versus ERI
PII measures current structural integrity. ERI measures how well the network retains function and recovers through adverse conditions over time. They answer related but different questions.
Separate dimensions matter: high throughput cannot cancel weak availability, and strong current integrity cannot prove future resilience.
See the complete framework or the controlling Edition 3.2 specification.