Infinite Solutions icon
InfiniteSolutions
Delivery Systems

DORA & Accelerate Principles

Manual chapter for measuring and improving software delivery performance with DORA.

Source: content/manual/01-dora-accelerate/index.md

DORA metrics give leaders a common language for delivery performance. They stay actionable because each metric has a crisp definition, a trusted data pipeline, and a playbook that improves outcomes. Treat this chapter as the north star for deciding what to instrument next and which improvement program deserves investment.

Why leadership should care

  • Customer value delivery: Frequent, low-risk releases keep value flowing and tighten discovery loops.
  • Engineer experience: Shared dashboards end arguments about “velocity” and point directly at system constraints.
  • Executive alignment: DORA metrics make investment cases obvious—slow lead time points at platform gaps, high failure rate highlights quality debt.

Metrics at a glance

Metric Definition Direction of improvement Fastest diagnostic
Deployment frequency How often production receives value Higher Release calendar annotated with change type
Lead time for changes Commit to running in production Lower Value stream map of tooling and approvals
Change failure rate % of releases that cause incidents Lower Incident register tied to deploy IDs
Mean time to restore (MTTR) Duration from incident start to resolution Lower Pager duty timelines, observability alert history

Implementation roadmap

  1. Agree on definitions and scope. Decide what counts as a deploy, which environments matter, and how incidents are classified. Publish the contract in your delivery handbook or internal wiki.
  2. Instrument the value stream. Configure SCM, CI/CD, and incident tooling to emit structured events (webhooks, APIs, data exports). Prefer automation—manual spreadsheets erode trust quickly.
  3. Automate rollups. Use pipelines (GitHub Actions, Dagster, dbt, or your BI stack) to aggregate metrics nightly, tag services and teams, and surface anomalies.
  4. Expose shared dashboards. Grafana, Looker, or even Google Sheets are fine—what matters is a single source of truth with time-series trends and team filters.
  5. Run regular reviews. Incorporate metrics into engineering ops reviews, executive updates, and team retros. Without cadence, metrics become shelfware.

Pair this roadmap with playbooks/measure-dora-metrics/checklist.md for a tactical task list.

Guardrails and anti-patterns

  • Measure in aggregate; do not weaponize individuals. The four metrics describe system health, not personal performance.
  • Avoid vanity targets (“four deploys per day”) unless paired with clear hypotheses and guardrails.
  • Resist redefining metrics when numbers look bad—fix root causes instead.
  • Beware noisy data: reconcile incidents without deploy links, normalize commit emails, and log schema changes that affect calculations.

Choosing the right playbook

Signal Leading diagnosis Recommended playbook
Lead time trending upward Large batch sizes, manual approvals playbooks/shortening-lead-time/index.md
Deployment frequency stalled Bottlenecked pipelines, lack of self-service playbooks/improving-deployment-frequency/index.md
Change failure rate spiking Weak automated testing, brittle rollbacks playbooks/reducing-change-failure-rate/index.md
MTTR above target Slow detection, unclear runbooks playbooks/accelerating-mttr/index.md
No trustworthy metrics Tooling gaps, unclear definitions playbooks/measure-dora-metrics/index.md

Each playbook includes a matching checklist to keep remediation measurable.

Operating cadence checklist

  • Weekly: Review dashboard, capture improvement experiments, and annotate anomalies.
  • Monthly: Pair DORA trends with qualitative DevEx findings (see playbooks/measuring-devex/index.md).
  • Quarterly: Validate definitions, audit data quality, and recalibrate targets based on business outcomes.

Tooling and templates

Reading list

  • “Accelerate” by Forsgren, Humble, and Kim (primary research study).
  • Latest DORA State of DevOps report for industry benchmarks.
  • Charity Majors on MTTR and ownership for cultural framing.

Related assets

Deep dive chapters

Glossary