I85PREP
⌘ K
CASM study materials
CASM study resources

CASM: Metrics, Measurement, and Reporting for Agile Service Management

CASM · Work through your exam, one topic at a time.

All guide sections

Metrics, Measurement, and Reporting for Agile Service Management

Metrics, Measurement, and Reporting for Agile Service Management

This domain tests metric selection, calculation and interpretation, including the discipline to reject averages that mislead.

Flow and delivery metrics

  • Lead time: request (or commitment point) to delivery. Cycle time: work start to done. Always state both endpoints.
  • Throughput: items completed per period. WIP: items started and not finished.
  • Flow efficiency: active time divided by elapsed time.
  • DORA four keys: deployment frequency, lead time for changes, change failure rate, time to restore. They are outcome indicators, never targets for individuals.
  • SLI (measured indicator), SLO (internal target), SLA (contractual commitment). Error budget = 1 − SLO, expressed in time or events.

Interpretation rules

Use distributions, not averages: report the 85th or 95th percentile, because the average hides the tail customers feel. Prefer leading indicators (WIP age, review queue length, error budget burn rate) for steering, and lagging ones for confirming. Beware Goodhart's Law: when a measure becomes a target, it stops being a good measure. Report at team or value stream level and never compare velocity across teams.

Worked example

An SLO of 99.9% availability over a 30-day month allows 43.2 minutes of downtime (30 × 24 × 60 × 0.001). Incidents have consumed 35 minutes, leaving about 8 minutes. A proposed release carries an assessed 20-minute exposure, which would exhaust the budget. Reasoning: the budget is a decision tool, so the release should be split into canary steps with bounded exposure, or the SLO renegotiated deliberately, before proceeding. Reporting that the service was 'mostly up' would conceal the arithmetic that should drive the decision.

Traps

Averaging lead time; using utilisation as a proxy for productivity; treating an SLO as an SLA; setting SLOs with no agreed consequence when the budget is exhausted; vanity metrics such as story points delivered or tickets closed that reward volume over outcomes; comparing teams' velocities.