Enterprises using AI-powered observability are seeing tangible gains in engineering productivity, according to new data released by New Relic. The company’s 2026 AI Impact Report draws a clear line between AIOps usage and outcomes such as faster issue resolution, reduced alert fatigue, and significantly higher deployment velocity.
The report is based on aggregated, anonymised data from 6.6 million users of the New Relic Intelligent Observability Platform across 2025. It compares teams using New Relic’s AI capabilities — spanning generative AI and AIOps features — with those relying on traditional, non-AI observability.
Cutting alert noise to reclaim engineering time
One of the most immediate benefits highlighted in the report is noise reduction. Engineers today spend nearly a third of their time firefighting system issues, a burden amplified by the sheer volume of alerts generated in modern environments. In 2025 alone, New Relic customers processed 2.2 billion alert events, close to half of which originated in production systems.
AI-strengthened observability, the report suggests, helps teams cut through this overload. Accounts using New Relic AI achieved twice the signal correlation rate of non-AI users and generated 27% less alert noise on average. By automatically grouping related signals into actionable incidents, AI reduces the manual effort required to triage issues, freeing engineers to focus on higher-value work.
Faster recovery when it matters most
Beyond fewer alerts, AI adoption appears to translate into faster recovery from incidents. Mean Time to Close (MTTC) — a key measure of operational efficiency — was consistently lower for AI-enabled teams. On average, New Relic AI users resolved issues around 25% faster than their peers.
The gap became even more pronounced during high-pressure periods. In May 2025, AI-enabled accounts averaged 26.75 minutes per issue, compared with 50.23 minutes for non-AI users. That 23-minute difference per incident, the report notes, can be critical in preventing investigation delays and preserving engineering momentum during outages.
Deployment velocity sees a step change
Perhaps the most striking finding relates to deployment frequency. By reducing time spent on triage and investigation, AI appears to act as a multiplier for engineering throughput. While non-AI users averaged 87 deployments per day during peak periods, teams using AI-powered observability reached as many as 453 deployments per day — a fivefold increase at the high end.
Overall, the data shows that organizations using New Relic’s AI features shipped code at an average frequency around 80% higher than those without AI. The implication is that operational efficiency gains are being directly converted into faster feature delivery and updates reaching the market sooner.
A new operational baseline
“AI is bringing a level of complexity to modern software operations that is moving beyond human ability to manage. In tandem, AI is also helping solve the problem it created,” said Camden Swita, Head of AI at New Relic. According to Swita, the findings point to a new operational baseline, where teams that effectively apply AI to observability spend less time maintaining systems and more time building new capabilities.
For engineering-led organizations, the report reinforces a growing narrative: AIOps is no longer just about managing complexity, but about unlocking speed. As competitive pressure mounts and release cycles compress, the ability to reduce operational drag may increasingly define how quickly businesses can respond to the market.