Skip to main content
DORA Metrics from Real Deployments

Stop Faking DORA Metrics. Start Measuring Real Deployments.

Most tools estimate deployment frequency from PR merges. That is not a deployment. CodePulse tracks actual deployments via GitHub's Deployments API, release tags, and environment branches, giving you DORA metrics you can actually trust.

GitHub Deployments API
Release Tags
Environment Branches
Read-Only Access

PR Merge ≠ Deployment. Here is Why It Matters.

The Accelerate State of DevOps research defines deployment frequency as how often code reaches production. Most analytics tools count PR merges to main and call that "deployment frequency." But merging code is not the same as releasing it. Conflating the two inflates your numbers and hides real bottlenecks.

What Other Tools Call "DORA"

  • PR merged to main = "deployed"
  • No awareness of staging, QA, or production environments
  • Hotfixes and rollbacks are invisible
  • Lead time stops at merge, not at production

What CodePulse Actually Tracks

  • GitHub Deployments API events per environment
  • Release tags with semantic versioning
  • Environment branch deploys (staging, production, etc.)
  • Full commit-to-production lead time

Four Metrics. One Dashboard. No Guesswork.

Track the four DORA metrics that predict software delivery performance, calculated from your actual deployment events.

DORA Deployment MetricsLast 30 days - Production environmentElite PerformerDeployment Frequency4.2 / day+18% vs last monthLead Time for Changes3.2 hrsImprovingChange Failure Rate2.8%97.2%Success rateMean Time to Recovery47 minIncidentDetectedFix DeployedRecovered

CodePulse deployment analytics dashboard - all four DORA metrics at a glance

Where Does Your Team Stand?

Benchmarks from the Accelerate State of DevOps Report. CodePulse automatically classifies your team against these tiers.

MetricEliteHighMediumLow
Deployment Frequency
How often code reaches production
On-demand
(multiple/day)
Daily to
weekly
Weekly to
monthly
Monthly to
semi-annual
Lead Time for Changes
Commit to production
< 1 hour1 day to
1 week
1 week to
1 month
> 1 month
Change Failure Rate
% of deploys causing incidents
< 5%5 - 15%16 - 30%46 - 60%
Mean Time to Recovery
Incident to service restored
< 1 hour< 1 day< 1 day1 week to
1 month

Source: Google DORA / Accelerate State of DevOps Report. Elite performers achieve 127x faster lead time and 182x more deploys than low performers. Read our complete DORA metrics guide for context on how to use these benchmarks without weaponizing them.

Three Data Sources. Full Deployment Visibility.

CodePulse reads deployment data from every signal GitHub provides, so your metrics are accurate regardless of how you release.

How CodePulse Detects DeploymentsDeployments APIdeployment_statusevents per environmentGitHub Actions, Vercel, etc.Release Tagsv2.4.1, release-2024-03semantic versioninggit tag + GitHub ReleasesEnvironment Branchesmain, staging, productionmerge-to-deploy workflowsHeroku, Railway, etc.CodePulseNormalize + CorrelateDeduplicate across sourcesDeploy Frequency4.2 / dayLead Time3.2 hrsChange Failure Rate2.8%MTTR47 min

CodePulse correlates multiple deployment signals for accurate, deduplicated metrics

What Sets CodePulse Apart

Real Deployment Data

Data comes from GitHub's Deployments API, release tags, and environment branches. No proxying PR merges as deployments. What you see is what actually shipped.

Read-Only. No Agents.

Read-only GitHub API access. No CI plugins to install, no agents to maintain, no webhooks to configure. You are measuring in 5 minutes.

Benchmarks Included

Your team is automatically classified against the Accelerate benchmarks. See whether you are Elite, High, Medium, or Low across every metric, every sprint.

Answers to the Questions You Actually Get Asked

"Are we deploying faster or slower than last quarter?"

Deployment frequency trends with automatic period-over-period comparison.

"How long does a commit actually take to reach production?"

Full commit-to-production lead time, not just merge-to-main.

"Which repos have the highest failure rate?"

Change failure rate broken down by repository and environment.

"How fast do we recover from bad deploys?"

MTTR tracked from deployment failure to successful redeployment.

"Can I present DORA metrics to the board?"

Export-ready dashboards with tier classifications and trend lines.

"Do we have data to justify more engineering headcount?"

Deployment load per engineer and bottleneck identification for staffing conversations.

Get Real DORA Metrics in 5 Minutes

Connect your GitHub organization. CodePulse syncs your deployment history, calculates all four DORA metrics, and classifies your team against the Accelerate benchmarks, automatically.

Free tier includes all 4 DORA metrics. No credit card required.