Phase 2 – Identify the Right Metrics

Purpose

• Define the most appropriate and actionable metrics to answer the performance question

• Align each metric to a specific owner, outcome, and cadence

Objectives

• Identify the right type of metric (SLA, XLA, KPI, OLA, CSF, etc.) for the question

• Validate that the metric is meaningful, measurable, and manageable

• Establish how each metric contributes to decision-making or continuous improvement

Inputs

• Performance question(s) from Phase 1

• Existing metric repositories or dashboards

• Business rules, compliance thresholds, and customer expectations

• Process maps, value streams, or capability models

Outputs

• A curated set of performance metrics linked to the question

• Metric definitions, targets, owners, and collection methods

• Documented rationale for inclusion/exclusion

Tools and Techniques

• Metric Selector Canvas: Match question type to metric form (leading, lagging, SLA, XLA, etc.)

• Measure Validation Grid: Confirm purpose, data availability, and decision-usefulness

• SLA/XLA Mapping Table: Translate service-level commitments into user experience impact

• Measurement Design Workshops: Cross-functional sessions to co-define what good looks like

Step 1: Analyse the Performance Question

• Break the question into its underlying dimensions (time, quality, volume, experience, etc.)

• Determine what kind of insight is required (trend, compliance, impact)

Step 2: Draft Candidate Metrics

• Review existing KPIs, SLAs, XLAs for reuse or revision

• Use the Metric Selector Canvas to explore new metrics where gaps exist

• Consider both quantitative and qualitative indicators

Step 3: Validate Metric Usefulness

• Apply the Measure Validation Grid to test if:

  • The metric answers the question

  • Data is available and reliable

  • The audience understands and acts on it

• Eliminate any measure that adds noise or isn’t owned

Step 4: Define Ownership and Cadence

• Assign a metric owner responsible for accuracy and review

• Define how often the metric is reported and refreshed

• Establish what thresholds or targets apply (or whether trend is more appropriate)

Step 5: Document and Approve

• Record the full metric definition, rationale, owner, and collection method

• Link each metric back to the original performance question

• Finalise for review in governance or performance reporting cycles

• • Don’t confuse measurement with performance: bad metrics can create bad behaviour

  • Focus on metrics that drive questions, not just reporting

  • Limit the number of metrics to those that matter most to the current goal

  • Don’t inherit metrics blindly from other frameworks — make them fit-for-purpose

  • Visualise measure-to-question relationships in workshops to drive shared understanding

  ---

• Efficiency: Time to resolve, cost per transaction, process step counts

• Effectiveness: SLA compliance, defect rates, service availability

• Experience: CSAT, NPS, sentiment analysis, usability feedback

• Flow: Throughput, wait times, handoff count, backlog age

• Value: Outcome attainment, goal alignment, cost vs benefit

• COBIT (MEA01): Provides guidance on performance monitoring practices and key metric types

• ITIL (Service Level Management): Aligns with service measurement and XLAs/SLA design

• Agile/Lean: Encourages flow-based metrics and team-level accountability

• Balanced Scorecard: Supports strategic alignment across multiple perspectives

• • Design & Flow: Helps translate the question into system behaviours and process touchpoints

  • Delivery & Assurance: Ensures selected metrics are consistent, credible, and reportable

  • Governance & Alignment: Provides oversight and validation of metric relevance and risk exposure

  Metrics are not just indicators—they are instruments. When carefully selected, they tune the organisation to respond to the signals that matter most.