Part 6 · 1 chapters · ~8 min

Accelerate

The research behind the DORA metrics: four measures of delivery performance, the finding that speed and stability go together, the link to organisational outcomes, and the technical, cultural and lean capabilities that drive it, with how to compute each metric and the Goodhart trap.

Accelerate

Accelerate is the research behind the DORA metrics. It matters because it replaced opinion with survey data: you do not have to trade speed for stability.

metric	measures	how to compute it from your systems
deployment frequency	throughput	production deploys per day, from the CD system
lead time for changes	throughput	merged commit timestamp → deploy timestamp containing it
change failure rate	stability	deploys followed by an incident, rollback or hotfix ÷ all deploys
time to restore service	stability	incident start → resolved, from the incident tool

code

-- lead time for changes from the deploy log (one row per commit per deploy)
SELECT percentile_cont(0.5) WITHIN GROUP (ORDER BY deployed_at - merged_at) AS median_lead_time
FROM deploy_commits
WHERE environment = 'production' AND deployed_at > now() - interval '30 days';

where it has aged, and the trap

The research is survey-based, and later DORA reports added reliability as a fifth metric and refined the clusters. The trap is Goodhart's law: once the metrics become targets for individuals or teams, deploys get split into meaningless pieces and incidents get reclassified. Use the metrics to improve the system, as the book says.

ACCELERATE

Nicole Forsgren, Jez Humble and Gene Kim, 2018: the four key metrics and what drives them

swipe the figure sideways, or tap expand for full screen

1/6

Delivery performance is measurable

The four key metrics: deployment frequency and lead time for changes (from commit to running in production) measure throughput; change failure rate and time to restore service measure stability. Together they describe delivery performance without counting lines or story points.