EMA Research: AI Adoption Cannot Eliminate Enterprise Observability Tool Sprawl
Survey of 356 IT pros finds six to eight tools create staffing and integration burdens that AI alone cannot resolve into a single view.
Key points
- Enterprise IT teams average six to eight observability tools, creating significant integration and staffing challenges.
- Forty-five percent of respondents cite skills and staffing burdens as a primary impact of tool sprawl.
- AI is used for alert correlation and anomaly detection, but vendors do not expect a single-pane solution.
Enterprise organizations typically deploy six to eight distinct observability tools, creating persistent integration and staffing challenges for network operations teams, according to new research from Enterprise Management Associates (EMA). The firm surveyed 356 enterprise IT professionals to assess how tool sprawl impacts operations as teams adopt artificial intelligence to manage complex IT environments.
EMA found that this proliferation of software consumes engineering time, increases staffing demands, and slows incident resolution. Most surveyed organizations are already using AI in specific observability areas, such as alert correlation and anomaly detection. However, the research indicates that AI adoption will not reduce the toolset to a single platform, leaving teams to manage fragmented data sources.
The Limits of AI Unification
Shamus McGillicuddy, vice president of research covering network infrastructure and operations at EMA, stated that no organization achieves a single pane of glass, even with AI assistance. During a web event sharing the findings from the report titled The Reality of Observability Unification in Modern IT Operations, he emphasized that sophisticated environments remain too complex for total consolidation.
Despite these limitations, many enterprises are actively working to unify their observability environments. EMA data shows that 51 percent of respondents have unification efforts under way. Another 32 percent are planning or evaluating such approaches, while only 17 percent have completed unification. Larger organizations reported having more observability tools in use across their environments, exacerbating the complexity.
Operational Impact and Costs
Tool sprawl creates measurable operational friction for IT teams. According to EMA, 45 percent of respondents cited skills and staffing burdens from operating multiple tools as a key issue. Additionally, 44 percent said that integration and API complexity consume significant engineering time, while 41 percent noted that context switching among tools slows investigations and response efforts.
Thirty-four percent of respondents reported that alert noise creates toil and cognitive overload. These issues contribute to higher total cost of ownership, slow incident detection, and inconsistent data quality. Coverage gaps and poor end-to-end visibility across technology domains further complicate troubleshooting for network operations staff.
What should you do?
Network operations teams should audit their current observability stack to identify redundant tools and integration points that consume engineering time. Organizations should evaluate whether AI features in existing tools can improve alert correlation without adding new platforms. IT leaders should prioritize unification efforts that address API complexity and context switching, rather than seeking a hypothetical single-pane solution.
What to do and how to stay safe: Enterprise Management Associates
- Audit current observability tools to identify integration bottlenecks and redundant platforms that increase staffing burdens.
- Leverage existing AI capabilities for alert correlation and anomaly detection to reduce cognitive overload and alert noise.
- Evaluate unification strategies that address API complexity and context switching to improve incident resolution speeds.
- Monitor data quality and coverage gaps across network, cloud, and application domains to ensure end-to-end visibility.
Step-by-step guide: Prompt Injection Warning Signs: How to Spot AI Manipulation
General security guidance from the Threat Times newsroom. It is not confirmed advice from the organisations named in this story.
Frequently asked questions
How many observability tools do enterprises typically use?
Enterprise organizations typically use six to eight observability tools, according to research from Enterprise Management Associates.
Does AI eliminate the need for multiple observability tools?
No, EMA research indicates that AI adoption will not reduce tools to a single platform, and no one gets a single pane of glass.
What percentage of respondents have completed tool unification?
Only 17 percent of surveyed respondents have completed unification of their observability environments.
