Cities & Infrastructure · Master KPIs
Use a compact outcome framework that keeps service quality, equity, resilience and evidence visible across the pillar.
This page turns the subject into a public problem-solving framework: place, problem, causes, evidence, solutions, roadmap, results and contribution.
Problem
What is failing, for whom and where?
02 · Definitions change between places.
03 · Targets reward activity rather than durable public benefit.
Root causes
Test causes before selecting projects.
Each programme creates isolated metrics
Verify this mechanism against local institutions, assets, user experience and available evidence.
Inputs and outputs dominate
Verify this mechanism against local institutions, assets, user experience and available evidence.
No standard geography or denominator
Verify this mechanism against local institutions, assets, user experience and available evidence.
Missing data is hidden
Verify this mechanism against local institutions, assets, user experience and available evidence.
Targets lack baseline and method
Verify this mechanism against local institutions, assets, user experience and available evidence.
Indicator ownership is unclear
Verify this mechanism against local institutions, assets, user experience and available evidence.
Evidence
Record what is known, how it is known and what remains uncertain.
| Evidence area | What to establish | Key limitation |
|---|---|---|
| Definition | Numerator, denominator, unit and population | Small changes alter comparability |
| Baseline | Value, date, geography and source | Backfilled baselines need disclosure |
| Target | Expected value and rationale | Ambition must not replace feasibility |
| Actual | Measured result and distribution | Averages hide unequal impact |
| Confidence | Method, completeness and review status | Precision must match evidence |
Evidence rule: label verified evidence, partial evidence, illustrative analysis, community submission and insufficient evidence separately.
Solutions
Combine immediate action, controlled pilots and structural reform.
Define a limited master outcome set
Define owner, cost, dependency, safeguard, baseline and decision gate.
Publish full indicator dictionaries
Define owner, cost, dependency, safeguard, baseline and decision gate.
Pilot district and neighbourhood disaggregation
Define owner, cost, dependency, safeguard, baseline and decision gate.
Link every KPI to a decision owner
Define owner, cost, dependency, safeguard, baseline and decision gate.
Audit data and protect against gaming
Define owner, cost, dependency, safeguard, baseline and decision gate.
Retire indicators that do not support action
Define owner, cost, dependency, safeguard, baseline and decision gate.
Roadmap
Move from diagnosis to accountable improvement.
Define place, people and outcome
Record the responsible actor, evidence requirement and next decision.
Build a verified baseline
Record the responsible actor, evidence requirement and next decision.
Diagnose causes and constraints
Record the responsible actor, evidence requirement and next decision.
Compare options and pilot
Record the responsible actor, evidence requirement and next decision.
Deliver with safeguards
Record the responsible actor, evidence requirement and next decision.
Measure, improve and scale
Record the responsible actor, evidence requirement and next decision.
Results
Track outcomes people can experience.
Essential service reliability
Publish baseline, target, actual, date, geography, source and distribution.
Safe and affordable access
Publish baseline, target, actual, date, geography, source and distribution.
Infrastructure condition
Publish baseline, target, actual, date, geography, source and distribution.
Climate resilience
Publish baseline, target, actual, date, geography, source and distribution.
Distributional equity
Publish baseline, target, actual, date, geography, source and distribution.
Public response quality
Publish baseline, target, actual, date, geography, source and distribution.
Risks & safeguards
Address foreseeable harm before scaling.
Metric overload
Teams report without learning
Safeguard: Keep a small core plus diagnostic layersTarget gaming
Behaviour shifts to improve the number
Safeguard: Use balanced measures and auditsFalse comparison
Different methods are ranked together
Safeguard: Require comparability labelsData exclusion
Unmeasured groups vanish
Safeguard: Publish coverage and missingnessFAQ
Questions that should be answered before action.
Only enough to represent core outcomes; detailed diagnostics can sit underneath.
Definitions can be common while benchmarks and context differ.
When it no longer represents the intended outcome, with a documented version history.
Improve this Cities & Infrastructure Master KPIs analysis with local evidence.
Submit a place, source, correction, working practice, implementation lesson or measured result.