Tool sprawl hampers enterprise observability efforts

Enterprises are struggling with fragmented tools, maturity gaps, and cultural hurdles as they try to advance monitoring across their distributed, hybrid environments. Two recent research reports emphasize the need for observability technologies and best practices to evolve from reactive firefighting to proactive intelligence—with AI leading the way.

New Relic’s 2025 Observability Forecast, based on a global survey of 1,700 IT and engineering leaders, highlights the financial stakes of outages and the growing role of AI in observability. BlueCat and Enterprise Management Associates (EMA) provide additional perspective with their report, The Network Observability Maturity Model: How to Plan for NetOps Excellence, which surveyed 250 IT stakeholders and details the steps network operations teams can take to achieve operational excellence. The EMA study found that just 46% of respondents believe they are fully successful with network observability tools.

“Given the challenges that network teams continue to face with their network observability tools, they need a path to follow toward success. A maturity model can help organizations find that path,” the EMA report reads.

When asked by EMA what the biggest complaints they had about their network observability tools, respondents said:

• Limited scope—I can’t monitor everything I need to monitor: 25.1%
• Too expensive: 22.5%
• Lack of customization: 20.8%
• Difficult to implement/maintain: 20.5%
• Insufficient scalability: 18.8%
• Poor data quality: 16.5%
• Too noisy—alert fatigue: 15.7%
• Lack of insights: 14.0%
• Too difficult to use: 12.5%
• Poor customer support: 8.5%
• Other: .6%

According to the New Relic report, the median cost of a high-impact outage is $2 million per hour. For organizations with full-stack observability, that figure drops to $1 million—but doubles for those who have not achieved full-stack observability. Despite the benefits, nearly three-quarters (73%) of organizations surveyed lack full-stack observability. The study also found that 75% of organizations reported positive ROI from observability investments, with nearly one in five (18%) seeing three- to ten-fold returns.

“Even as organizations invest in observability, many remain far from achieving full-stack visibility. Forty-one percent of leaders reported they still learn about service interruptions through inefficient means—customer complaints, incident tickets, or manual checks,” the New Relic report says. 

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.