HCD outcome measurement and learning plan

Purpose

When to use this plan

Use it before committing to a substantial intervention, when defining a service or product outcome, when establishing an HCD capability, or when leaders need evidence to continue, change, scale, or stop an approach.

Begin with the decision the measurement must inform. A dashboard without an owner, threshold, review cadence, or resulting action displays information but does not create a learning system.

Logic

Build an outcome chain before selecting metrics

1. State the current condition and the people affected.
2. Name the intended experience and organizational outcomes.
3. Describe the HCD contribution without claiming sole causation.
4. Connect capability and activity to outputs, behavior, experience, and service effects.
5. Record assumptions and evidence that could disconfirm the chain.
6. Choose the smallest balanced set of measures needed for the decision.

Balanced evidence

Use six dimensions without forcing every metric

Select dimensions based on the decision and risk. Always examine accessibility and unequal effects when averages could hide materially different outcomes.

Operational definition

Make every measure decision-ready

For each measure, record:

• the question it helps answer and its precise definition;
• baseline, target, threshold, or comparison condition;
• source, collection method, population, and segmentation;
• owner, cadence, and the decision date it supports;
• limitations, alternative explanations, and data-quality risk;
• the action triggered when results meet, miss, or complicate expectations.

Reusable artifact

Copy the Markdown plan

Adapt this structure to a governed documentation, analytics, or work-management environment. Do not include personal or sensitive data unless that system and collection method are authorized.

# HCD outcome measurement and learning plan

## Plan control
- Outcome or decision:
- Accountable owner:
- Measurement lead:
- Related effort and decision records:
- Baseline period:
- Review cadence:
- Next decision date:

## Intended change
- People and context:
- Current condition:
- Intended user or stakeholder outcome:
- Intended mission, service, or organizational outcome:
- HCD contribution and other material influences:
- Assumptions:

## Outcome chain
- Capability or input:
- HCD activity:
- Immediate output:
- Expected behavior or experience change:
- Expected service or mission effect:
- Evidence that could disconfirm the chain:

## Balanced measures
For each measure:
- Dimension: Experience | Accessibility and equity | Behavior and adoption | Delivery and quality | Mission, service, and risk | Capability and learning
- Question the measure helps answer:
- Measure and operational definition:
- Baseline:
- Target or decision threshold:
- Source and collection method:
- Population, segment, or disaggregation:
- Owner and cadence:
- Limitations and possible unintended consequences:

## Interpretation and decision
- What changed:
- Confidence and alternative explanations:
- Unequal or unintended effects:
- Evidence gaps:
- Decision or action triggered:
- Owner and due date:

## Learning record
- What should continue:
- What should change:
- What should stop:
- Reusable evidence, pattern, or guidance:
- Next review trigger:

Guardrails

Interpret evidence before claiming impact

• Are activity and output counts clearly separated from outcomes?
• Are displayed values real, current, and labeled when illustrative?
• Could averages conceal accessibility or equity differences?
• Are qualitative findings preserved alongside quantitative measures?
• Could outside changes explain the observed result?
• Are adverse effects, workarounds, and non-adoption visible?
• Does each review produce a decision, action, or documented learning?
• Are reusable findings connected back to methods, patterns, and evidence?