Enterprise AI Measurement Guide

CI/CD

Deploy MTTR

How does Deploy MTTR measure recovery time from deployment failures in the CI/CD tab, and how does it relate to Change Failure Rate?

What it shows

Deploy MTTR (Mean Time to Recovery) measures how long it takes the engineering team to restore production service after a deployment failure, the fourth DORA metric. A shorter MTTR means the team can detect, diagnose, and resolve production issues quickly, limiting the blast radius of any individual failure. In AI-assisted engineering environments, MTTR often improves even as Change Failure Rate rises, because the same AI tools that generate more failures also accelerate the debugging and hotfix cycle.

Why it matters

MTTR improvement is sometimes the one bright spot in a CI/CD picture where other metrics are moving the wrong direction after AI agent adoption. Larridin customer data shows Deploy MTTR improving (down 121%) in the same window where Change Failure Rate rose 83%, faster recovery even as failures increased. That pattern is valuable context: it distinguishes an engineering org that breaks production more often but recovers quickly from one where failures are both more frequent and more damaging. The distinction matters for incident risk assessment, SLA management, and board reporting.

The Larridin angle

Larridin surfaces MTTR alongside Change Failure Rate in the same tab, making the recover-faster-even-as-we-break-more pattern visible as a combined signal. That combination is the one CTOs need to present to boards and CISOs when explaining that rising failure rates are accompanied by shrinking recovery windows, which changes the risk profile materially.

Related CI/CD Metrics

• Deployment Frequency
• Change Failure Rate
• Change Lead Time
• CI Success Rate / Cancellation Rate / Median CI Duration
• CI Run Outcomes Chart
• Deploy Outcomes Chart
• Deployment Log