New Relic Smart Alerts Uses Historical Telemetry to Recommend Alert Coverage

From a blank page to informed recommendations

In a blog post by Douglas Braun, a Product Marketing Manager at New Relic, the company argues that autonomous operations starts with knowing what deserves attention. Modern observability environments generate large amounts of telemetry across applications, services, Kubernetes clusters, databases and cloud infrastructure. That visibility, the post says, does not by itself establish which conditions matter enough to investigate.

Historically, according to the post, deciding what to monitor has been manual work. An engineer decides what should be monitored, creates alert conditions and repeats the process as new services and infrastructure appear. That becomes hard to sustain when organizations manage thousands or tens of thousands of entities that are constantly changing.

Smart Alerts is meant to change this. Instead of asking engineers to begin every monitoring decision from a blank page, New Relic can use operational data already in the platform to recommend alert conditions, coverage and monitoring best practices.

What Smart Alerts does and does not do

New Relic is explicit about the limits of the feature. Smart Alerts does not remove engineers from the process. It also does not continuously change alert thresholds as application behavior changes. Threshold-based alerting, the post says, remains an important part of observability.

What changes is the manual effort needed to establish appropriate alert coverage in the first place. The stated goal is not to create more alerts, but to help organizations set up meaningful coverage more quickly and consistently, while engineering teams keep control over the monitoring strategy that suits their environments.

The company says the value shows most at scale. In cloud-native environments, teams often solve essentially the same monitoring problem again and again across applications, services, clusters, infrastructure and databases. Historical telemetry gives engineers a more informed starting point and reduces repetitive configuration work.

Smart Alerts also does not perform incident investigation or decide the appropriate response. According to New Relic, it contributes a stronger detection layer that engineering teams and other New Relic capabilities can use as a source of better operational signals.

Why detection matters for AI

The post links the feature to the growing role of AI in software operations. AI can help engineers investigate incidents, interpret complex telemetry, surface relevant context and assist with determining what may have happened. But the usefulness of that investigation depends in part on the quality and relevance of the signals that start it.

If monitoring coverage is incomplete, important conditions may never get the attention they deserve. If alerting produces too many low-value notifications, responders must separate meaningful signals from noise before they can address the underlying problem. In both cases, the post says, everything that follows detection can suffer.

New Relic describes autonomous operations not as a single feature or an AI agent that handles every stage of an incident. It calls it an operating model in which telemetry, detection, trusted operational context, AI-assisted investigation, specialized capabilities, workflows, governance and human judgment work together.

What it means for teams

The company frames a progression across the operational lifecycle: better telemetry supports better detection, better detection provides more useful signals for investigation, and rich operational context helps AI and engineers interpret those signals. Teams then decide on a response using the controls, workflows and human oversight their organizations require.

For New Relic customers, the practical claim is narrower than the broader vision. Smart Alerts turns historical telemetry into recommendations that make meaningful alert coverage easier to establish and scale. The post offers no pricing, availability dates or performance figures, so the benefit rests on New Relic's description of the approach rather than on measured results.