Metric Definition Template

Name: [Metric name and abbreviation]

Owner and version: [Name / version]

References: [Discussion, supporting sources and any recorded approval]

#1. Purpose and benefit

• Question: The question the metric answers for a governance decision-maker.
• Expected benefit: The organizational benefit the metric helps assess, and the expected contribution of governance. Examples include growth, responsible autonomy, continuity of legitimate work, accountability, risk reduction or compliance.
• Decision supported: The decision or action the result can inform, and who is responsible for it.

#2. Definition and calculation

• Definition: A short description of what is measured.
• Calculation: The formula, unit and counting unit, such as decisions, invocations, completed operations or spending. For ratios, the numerator and denominator are explicit.
• Time basis: Reporting period, timestamp used and, when reporting change, the comparison period or baseline.
• Counting rules: Treatment of exclusions, duplicates, retries and unknown or missing values. A zero denominator is reported as not calculable, rather than as zero percent.

#3. Evidence

Each input is traceable to a record or another identified source. Producer-record fields, collector-derived values and external inputs or estimates are distinguished, consistent with the GovOps TRACE Profile.

The definition identifies how inputs are linked, what evidence checks are needed and which required inputs are unavailable. Assumptions used for estimates are explicit.

#4. Capability and business context

• Scope: The capabilities, systems, population or business operations covered.
• Capability mapping: Where applicable, how evidence is linked to the ACC, including the mapping version. A broader metric explains its level of analysis.
• Breakdowns: The comparisons needed for interpretation, such as capability, business unit, risk tier or actor type. Unknown or unmapped cases remain visible.

#5. Interpretation and thresholds

• Meaning and action: What different results indicate and what response they may justify. Whether the metric measures the stated benefit directly or is an indicator of it.
• Thresholds: Any target or colour boundary, its applicable context, rationale and who agrees it. Where thresholds are not established or not applicable, that is stated.
• Evidence sufficiency: When a result can be assessed and when it remains unassessed. Missing evidence does not imply a positive result.

An observed change alone does not establish that GovOps caused a business benefit. Any heat-map use identifies the relevant layer; aggregation across metrics follows the proposed heat-map assessment method.

See the draft acceptance process and criteria or return to the draft metrics overview.