GUIDE
Guides to DORA and Flow Metrics
Plain-English explanations of the concepts behind the product — no pitch, just the concept.
No product pitch on these pages — they explain the concepts DORA Metrics and Flow Metrics are built on, useful whether or not you ever use moasy.tech.
Organized in five clusters: one on DORA metrics (deployment frequency, lead time, change failure rate, mean time to restore, and the broader "what does DORA mean" question), one on Flow metrics (cycle time, work in progress, flow distribution, velocity, backlog health, and where the Flow Framework terms come from), one on cost classification (what the CapEx/OpEx split means for engineering work and why it matters beyond accounting), one on team health (toil ratio, bus factor, rework and review health, and cancellation rate — the signals about how sustainable a team's pace actually is, not just how fast it ships), and one on what computing these metrics with AI would actually cost (with an interactive calculator, broken down by integration category). Each cluster starts with an overview guide, then breaks into a dedicated page per metric for anyone searching for one specific term.
What Are DORA Metrics?
DORA metrics (sometimes called DORA4) explained in plain English: deployment frequency, lead time for changes, change failure rate and mean time to restore — what each one measures and why teams track them.
Deployment Frequency
What deployment frequency measures, how to calculate it, and what elite vs. typical performance looks like — one of the 4 DORA metrics.
Lead Time for Changes
What lead time for changes measures, how to calculate it from commit to production, and what elite vs. typical performance looks like — one of the 4 DORA metrics.
Change Failure Rate
What change failure rate measures, how to calculate it, and what elite vs. typical performance looks like — one of the 4 DORA metrics.
Mean Time to Restore
What mean time to restore (MTTR) measures, how to calculate it, and what elite vs. typical performance looks like — one of the 4 DORA metrics.
What Are Flow Metrics?
Flow metrics explained in plain English: cycle time, work in progress (WIP), throughput and flow distribution — what each one measures and how they relate to the Flow Framework.
Cycle Time
What cycle time measures, how to calculate it, and why median matters as much as average — a core flow metric (also called Flow Time in the Flow Framework).
Work in Progress (WIP)
What WIP measures, why it's a leading indicator rather than a lagging one, and how it relates to cycle time — also called Flow Load in the Flow Framework.
Flow Distribution
What flow distribution measures — the mix of feature, defect, debt and risk work — how to calculate it, and why the categories your team uses can differ from another team's.
Velocity
What velocity measures, how it differs from story-point velocity, and why comparing it across teams almost always gives the wrong answer — a core flow metric.
Backlog Health
What backlog health measures — backlog size, aging, stale items and runway — how to calculate each one, and how to read a healthy vs. unhealthy backlog.
Cost Classification
What software cost classification measures, why the CapEx/OpEx split matters beyond accounting, and how cancelled work can quietly distort it.
Team Health
Toil ratio, bus factor, rework and review health, and cancellation rate — the signals about how sustainable a team's pace actually is, not just how fast it ships.
Toil Ratio
What toil ratio measures, why it turns "we're too busy to ship" into evidence a leader can act on, and how to read it against a team's own history.
Bus Factor
What contribution concentration measures, why a low bus factor is a resilience risk rather than a compliment, and how to interpret it without naming names.
Rework & Review Health
What rework rate and PR review health measure, why both catch a quality problem before it becomes an incident, and how to read them together.
Cancellation Rate
What cancellation rate measures, why it's usually a scoping signal rather than an execution one, and why an invisible cancellation rate is worse than a high one.
Cost of AI-Generated Metrics
A worked estimate, with an interactive calculator: what it would actually cost in LLM tokens to compute DORA and Flow metrics with AI instead of deterministic rules.
AI Cost of Ticket Classification
What it would cost in LLM tokens to classify Jira or Linear ticket changes with AI instead of configurable rules — a worked estimate per team, per month.
AI Cost of Analyzing Pull Requests
What it would cost in LLM tokens to analyze GitHub or Azure Repos pull request activity with AI instead of SQL aggregation — a worked estimate per team, per month.
AI Cost of Analyzing Deploys
What it would cost in LLM tokens to analyze deploy events from GitHub Actions, Vercel, ArgoCD or Azure Pipelines with AI — a worked estimate at DORA elite-tier frequency.
AI Cost of Analyzing Incidents
What it would cost in LLM tokens to correlate incidents from Waroom or incident.io with deploys using AI instead of a configurable time window — a worked estimate.