The 5 Levels of AIOps Maturity

Understanding where an organization is with AI implementation can help you make smarter decisions about the solutions they need today and how to guide them toward greater value in the future.


While some businesses continue to struggle with challenges from siloed data — and rapidly increasing data volumes — others are turning to artificial intelligence (AI) to provide the visibility into operations and insights they need for data-based decisions. A few years ago, Gartner coined a term, “AIOps” for AI working in concert with IT operations to give organizations these advantages.

Sean McDermott, CEO and founder of Windward Consulting, explains, “I see IT operations as both an organization and a function. Often, you have a ‘director’ or ‘manager’ of an operations team. However, operations are also a function, such as network or security operations. AIOps is a long-term strategy to bring AI and machine learning to the operations team and functions. So, really AIOps is a subset of both, driven by technology.”

“When we look at AIOps as a strategy, it is just not root cause analysis; it’s bringing AI and machine learning to the prescriptive practice and the automation stages,”  he explains. “In other words, most AIOps platforms are focused on event handling and processing, but we view AIOps as a broadening strategy, implementing AI/ML in all stages of the detection, isolation, prescription, and remediation stages.”

Steps in the AIOps Evolution  

In general, businesses and organizations evolve through five stages on their way to AIOps maturity:

  1. Reactive: At this stage, organizations have siloed operations and collect data on events so they can react to them. In general, there is no dialog between systems and the business. This puts the company in a constant “fire-fighting mode.”
  2. Integrated: As organizations mature in their AIOps implementation, they integrate data sources into unified architecture and improve IT service management (ITSM). At this stage, they also break down silos and create a dialog within the business.
  3. Analytical: Stage 3 involves implementing a unified analytics strategy with data transparency for all stakeholders. Organizations also optimize ITSM processes and define measurement points and baseline metrics.
  4. Prescriptive: Organizations that have matured to this point increase automation, often introducing machine learning. ITSM also leverages automation with human interaction. Businesses at this stage also use comparative analytics to measure improvements and business value.
  5. Automated: At this final stage of maturity, businesses have the capability for data sharing among all stakeholders, full automation with no human interaction, machine learning based on predictive models, full transparency of analytics, and proactive decision making based on business value.

McDermott comments, “Most businesses are in stages 2 and 3, integrated and analytical. However, we do see companies that have capabilities in all three stages, but they’re not fully deployed.”

He says that although AIOps isn’t widely adopted now, “We believe it is definitely heading there.

AIOps Exchange, a private forum for the exchange of ideas related to AIOps, reports 84 percent of organizations were planning and budgeting for an AIOps project in 2019, and going into 2020, 50 percent had already leveraged AIOps to improve customer experiences.  “Most companies implementing AIOps are focused on specific areas and will ultimately convince other groups to come along,” says McDermott, “but there are business processes changes that are necessary, and some organizations are resistant to change.”

Advice for Software Developers Involved in the AIOps Evolution

McDermott’s advice to software developers is to approach AIOps one step at a time. “As teams maneuver through each level of AIOps maturity, it’s essential to keep the long-term AIOps strategy and goals at the forefront to fully unlock the true potential of AIOps.

“As they evolve, they must develop a foundation to identify how they will track this success and deploy machine learning and automation to drive business value and transparency across the entire organization,” he concludes.

Jay McCall

Jay McCall is an editor and journalist with 20 years of writing experience for B2B IT solution providers. Jay is co-founder of XaaS Journal and DevPro Journal.

Jay McCall

Jay McCall is an editor and journalist with 20 years of writing experience for B2B IT solution providers. Jay is co-founder of XaaS Journal and DevPro Journal.