Both reports emphasize that tool sprawl is slowing progress. The New Relic report found that organizations still average 4.4 observability tools, even after a 27% drop in the past two years. More than half (52%) of respondents plan to consolidate onto unified observability platforms. In its report, EMA reported similar findings, with 87% of network operations teams relying on multiple tools, often without meaningful integration. This type of “swivel-chair” troubleshooting—hopping between dashboards to reconstruct incidents—remains widespread. Successful organizations, EMA said, are those investing in integration and automation to streamline workflows.
EMA’s maturity model defines five levels of observability: Ad Hoc/Reactive, Fragmented/Opportunistic, Integrated/Centrally Managed, Intelligent/Automated, and Optimized/AI-Driven. Most organizations today fall into the middle stages, with fewer than half reporting they are fully successful with their observability tools. The leading edge is just beginning to reach the AI-driven observability stage, where end-to-end troubleshooting automation and predictive optimization come into play.
New Relic reports AI monitoring adoption rose from 42% in 2024 to 54% in 2025, marking the first time a majority of organizations are deploying AI for observability. Leaders cited AI-assisted troubleshooting, automated root cause analysis, and predictive analytics as the top use cases. EMA’s maturity model aligns, with advanced organizations using AI for automated remediation, adaptive playbooks, and AI-driven recommendations for proactive capacity management. Those still relying on static thresholds and manual scripts are struggling to keep pace.
EMA found that success correlates with customizable, role-specific dashboards and reporting that spans teams. New Relic found similar results, noting a cultural shift where “reliability becomes everyone’s responsibility.” According to these reports, observability maturity requires more than unified platforms and AI. It also requires alignment across DevOps, NetOps, SecOps, and business stakeholders. Unified, AI-enabled observability can reduce downtime, improve efficiencies, and create resilience.
