Business stakeholders may not explicitly recognize it, but so many of today’s digital initiatives depend upon the reliability of IT operations to consistently keep business chugging along. When IT ops is off its game and when network operations centers (NOCs) are so drowned in alerts that they can’t quickly respond to incidents, the business repercussions reverberate. Productivity takes a hit, internal users become frustrated, and customer experience suffers.
Which is why enterprises are increasingly mashing up the power of artificial intelligence (AI) with IT operations. Experts say that early adopters of AIOps are using these tools to automate mundane tasks and orchestrate complex processes. Doing so is helping IT ops teams streamline operations and address some of the perennial blockers of incident management that threaten digital resilience.
Just as other areas of an organization are leveraging the analytics and machine learning capabilities of AI to make faster business decisions, through AIOps, IT is leveraging those same capabilities to quickly crunch data from IT operations monitoring tools in context of one another for faster response. That contextual analysis—which is usually impossible to achieve using manual methods, helps ops teams to find important anomalies, diagnose problems more quickly, and drive swifter, more automated responses to system problems.
“Most IT practitioners are reporting that AIOps tools are increasing mean-time-to-resolution of incidents by as much as 50%,” says Bhanu Singh, senior vice president at OpsRamp. “That’s clear, tangible value, and frees IT teams to think more strategically.”
Yesterday, OpsRamp released the State of AIOps Report, which examined the IT operational challenges enterprises are trying to address through AIOps and the way that early implementations of AIOps is changing the way ops teams do their work.
Based on a survey of 200 IT professionals, the report showed that the number one challenge that IT teams face in managing incident response has to do with inaccurate and voluminous system data—in other words the dreaded “alert fatigue.” The study showed 68% of “IT teams are struggling to not only suppress alert noise but also extract meaningful insights” that can speed up response times. Right behind that, 65% of organizations say that analyzing incidents to determine the most probable root causes is the second most serious challenge. And coming in third place is the volume of manual tasks it takes to analyze and respond to IT incidents.
According to Gartner, AIOps is still very much an emerging technology. Last year only about 5% of organizations made this kind of tooling available to their ops teams. But that’s changing very quickly. By 2023 Gartner expects close to one in four enterprises to use AIOps.
This rapid adoption is coming from the big upside that the early adopters are reaping from AIOps tools. Yesterday’s study showed that 87% of IT pros using AI ops are getting value out of them. Approximately eight in 10 organizations report that they’re fixing incidents at least 26% to 50% faster due to their use of AIOps. One in 10 organizations report that their mean time to resolution has been sped up by 76% to 100%.
The study showed that the three biggest operational benefits that these teams are taking from these tools are the automation of low-value, repetitive tasks; swifter correlation of alerts to get to the root cause more quickly; and the reduction of incident and ticket volumes.
With those benefits achieved, organizations can then turn IT ops teams toward higher-value work that improves the delivery of digital services.
“Automate those mundane, routine toil activities that you really don’t want to be having your smart, expensive people working on,” suggests Andi Mann, chief technology advocate at Splunk. “Orchestrate complex things as well so your people can focus on what really matters and use their intuition, use their creativity, use their innovation to do the things that only they can do.”
