Engineering teams today ship code faster than ever. AI-assisted development has pushed change volume past what teams can manually review before it ships, which makes what happens after deploy the real bottleneck. Automated CI/CD pipelines and AI tools allow teams to deploy changes to production dozens or hundreds of times a day. Modern software delivery requires fast feature releases alongside strong operational reliability. Yet as release speed increases, tracking the exact impact of every code change remains a major challenge.
Without a clear link between feature management and full-stack observability, engineering teams run into constant issues. Features go live without immediate visibility into performance changes, forcing teams to fix incidents after users notice them. When problems happen, finding out whether a performance drop was caused by a feature flag, an infrastructure update, or a downstream dependency requires manual work across separate tools. This lack of clear change tracking slows down detection and recovery, lowering confidence in shipping updates quickly.
LaunchDarkly and New Relic close the loop between feature release and system intelligence.
How the Closed Loop Works
LaunchDarkly manages gradual feature rollouts, targeted access, and quick rollbacks. New Relic provides real-time performance data across the entire system.
Together, LaunchDarkly and New Relic create a continuous feedback loop across every step of software delivery:
- Release: LaunchDarkly controls who sees new features, while New Relic sets the performance baseline to monitor.
- Observe: LaunchDarkly tracks active flag settings across user groups while New Relic monitors latency, error rates, traces, and system performance.
- Evaluate: Detailed data from New Relic gives teams immediate context during feature rollouts, making it easy to check safety rules and identify the cause of changes.
- Act: Teams can pause, reduce, or roll back feature exposure as soon as performance issues appear, limiting user impact in seconds.
Empowering Every Role Across the Software Lifecycle
This partnership delivers practical benefits for every engineering role:
For CTOs and Engineering Leaders: Increase deployment speed and development efficiency without sacrificing system stability, security, or compliance.
For DevOps and SRE Teams: Instantly identify which change caused an issue during releases, shortening detection times and streamlining incident response.
For Platform Engineers: Set up automated release rules and telemetry signals that fit right into existing delivery pipelines.
For Product and Application Teams: Get real-time updates on how each feature release affects app performance and user experience.
Guarded Rollouts Meet Full-Stack Telemetry
This integration enables safer rollouts by embedding New Relic’s telemetry checks directly into LaunchDarkly feature flag evaluations. Features can roll out automatically on set schedules—starting with a small percentage, like 5%, and expanding to 100% over 24 hours. If performance drops during rollout, the system acts like a circuit breaker and automatically triggers an immediate rollback. Throughout the release, error logs and traces are updated with feature flag details to help teams pinpoint root causes right away.
During live release cycles, engineering teams rely on guarded rollouts to strictly limit the blast radius of new code releases. Instead of exposing changes to 100% of the user base at once, rollout schedules begin by exposing the feature flag variation to a small initial segment—such as 5% of users—and progressively scaling to full exposure over a 24-hour period.
As feature flag evaluations execute, New Relic’s telemetry metrics like 500 server errors, caught exceptions, and performance regressions are evaluated directly within the execution context of the LaunchDarkly flag. When a metric threshold is exceeded, the system operates as an automated circuit breaker, immediately triggering a rollback to the flag's prior state to contain impact.
If an incident occurs, teams can precisely identify the exact cohort of exposed users (e.g., 37 affected users) directly from the release audience metrics. This real-time visibility enables rapid incident response, allowing developers to isolate issues and deploy fixes without impacting the broader customer base.
Automated rollback triggered after a payment latency regression.
LaunchDarkly controls what ships. New Relic tells you whether it should keep shipping.
Join the Limited Preview
By combining LaunchDarkly runtime control with New Relic's intelligent observability, teams can automate rollbacks and ship code faster with complete safety, while keeping the blast radius contained. Test out the integration today and submit the interest form to join our limited preview.
The views expressed on this blog are those of the author and do not necessarily reflect the views of New Relic. Any solutions offered by the author are environment-specific and not part of the commercial solutions or support offered by New Relic. Please join us exclusively at the Explorers Hub (support.newrelic.com) for questions and support related to this blog post. This blog may contain links to content on third-party sites. By providing such links, New Relic does not adopt, guarantee, approve or endorse the information, views or products available on such sites.
