Applications
Applications
What decision needs a more complete balance sheet?
An application is a decision context: a question, a boundary, evidence, and people with responsibilities. GDR is designed to keep these connected through scenarios, action, and verification.
Application study 001
The Amazon bioregion
What would it take to decide well here? The Amazon makes the problem visible at planetary scale: a system whose meaningful boundary crosses administrative borders, whose value is mostly not priced, and whose depletion is not recoverable on a policy timescale. It is the first worked application of the method, not the boundary of it.
Status
Exploratory. No partnership, engagement, or implementation is claimed.
Evidence
Source-labelled proxy snapshot with versioned records; observed and modelled values kept distinct.
Not connected
Governed actor assignments, execution, and outcome verification.
Application families
The framework can be applied at different scales without changing the evidence rules.
Nations and regions
Public decisions that depend on long-run ecological, social, and capital capacity.
Bioregions
Systems whose meaningful boundary crosses or differs from administrative borders.
Organizations
Institutions examining dependencies, effects, resilience, and future capacity.
Portfolios and capital
Capital allocation examined alongside the capacities it depends on and influences.
Programs and interventions
Specific actions evaluated through baseline, intended change, evidence, rights, and verification.
Decision context / intended working model
From a consequential question to a verifiable result.
Question
What decision needs a more complete account? What evidence would change it?
Boundary
Which place, period, people, and rights belong in the account? Who has authority to decide?
Evidence
What supports the decision, which version is being used, and what remains unknown?
Scenario
Compare possible paths, assumptions, constraints, and trade-offs before committing.
Action
Connect responsible people and organizations with resources, dependencies, and milestones.
Verification
Define how change will be checked, then carry the result and its limitations into future work.
Hard cases
Every framework eventually meets a subject its standard method was not built for.
Most indices handle this badly or silently, either forcing an ill-fitting number or quietly dropping the case from a published table without explanation. These four recurring cases resolve to three different protocols rather than one uniform rule, and the variation is the point.
The city-state with no agricultural sector
Where an indicator has no domestic analog, it is excluded from the pillar's aggregation rather than scored at zero, the indicator count is adjusted, and the substitution is disclosed exactly as every other imputation is. A city-state sits at the boundary between the national and municipal tracks and should be read on whichever indicator set actually applies to its economic structure.
The nation at war
No composite is published for a nation in active, large-scale armed conflict where the social pillar's underlying survey data cannot be verified as current. Ecological indicators, where remote sensing remains available, may be reported individually and labelled as a partial indicator profile, not as an account. When the data honestly does not support a number, the position is to say so rather than compute one from whatever fragments remain.
The sanctioned economy
Scoring prioritizes indicators sourced from third-party remote sensing and independent international bodies over self-reported national statistics wherever both exist, and any account published carries an explicit flag naming which indicators rely on self-reported data that could not be independently verified. Not refused outright, unlike the war case, but visibly lower in confidence.
The microstate
Several indicators are rarely fielded at a population of a few tens of thousands, and single-year fluctuations that would be noise for a larger nation can swing an indicator dramatically. Accounts at this scale are reported with a wider uncertainty band, and flagged where they rely on a peer-group imputation the framework does not otherwise use.
Declining to score
A framework that always produces a number is optimizing for coverage over honesty.
Declining to score, stated plainly rather than quietly omitted from a table, is itself a form of rigor. The standing commitment is to state a limitation directly rather than resolve it by force: where a public series does not exist for a cell, that cell is reported as unpopulated rather than imputed; where a substitute of construct is used, it is flagged per cell and carried at a lower confidence tier rather than silently absorbed; and where the underlying data cannot support a composite at all, none is published.