We published our Now Tech: Artificial Intelligence For IT Operations, Q2 2022 at the beginning of June, but we collected a lot more data in that effort that needs to be accounted for in future AIOps decisions. One — and probably the biggest topic that surfaced — is that the influx of new vendors into the AIOps space is wreaking havoc with the traditional perspective for AIOps implementations. I will have more details on that in an upcoming blog and report. For now, however, I want to focus on the wealth of data collected in the latest AIOps Now Tech and three items from that report in particular. The AIOps Data Overview, Q2 2022 report gets into more detail on several other topics beyond these three, so be sure to check it out.

Digital Experience Is Still Lagging

More and more people are talking about business outcomes, but that can’t be fully realized until we have more pervasive digital experience capabilities. The vendor support for collecting and generating this type of sensory data has not yet arrived in AIOps tools, according to the data collected. We looked at 15 different capabilities in the digital experience category. Only approximately 40% percent of the vendors supported half of those capabilities. This needs to become a more widespread capability to give organizations better insights into their user behaviors and drive positive business outcomes.

Interoperability Via OpenTelemetry Is Growing

There is tremendous pressure on vendors to be everything to everyone, and interoperability is no exception. They need to provide native sensory and telemetry capabilities that offer their clients opportunities to reduce technical debt. They must also provide a wide breadth of third-party integration capabilities, because organizations have a plethora of already installed software and hardware that isn’t going away anytime soon. Don’t underestimate the daunting task they have in this arena, but that doesn’t mean that they shouldn’t be pushed to do more. It is a challenge they must overcome if they want to rise or stay at the top of the AIOps vendor charts. Fortunately, more than half of the Now Tech vendors surveyed already offer some level of OpenTelemetry support. An additional 25% stated that they have plans in their roadmap for full support of OpenTelemetry.

Audit Trails Aren’t Reliable If They Can Be Manipulated

As we continue our move into automated and AI/machine-learning (ML)-driven settings, audit trails, and auditable data, are going to become even more vital to trust and transparency. When asked to describe their audit trail design, the vast majority of vendors stated that they did in fact have formal audit trails for some, if not all, of the data in their AIOps offering. It varied between all the data points being auditable to a subset, configurable by the client able to be audited. What was surprising was that only roughly half of those same vendors said that their audit trail was fully locked. So I ask you, how reliable is an audit trail that can be manipulated? Is an unlocked audit trail truly an audit trail if its accuracy can’t be ensured? We’ll delve deeper into that in the future when we ask, “How will your AI/ML engine handle being deposed in a court of law?” For now, be sure to scrutinize how your technology offering handles audit trails.

Snippets Are Good, But The Full Collection Is Better

Above, I called out three specific items out of the seven referenced in the data overview report. Our Now Tech research went even further in depth on countless other topics (for example, which of the 24 different sensory and telemetry data types that vendors were able to collect both natively and via third parties). If you’re an organization heading toward an AIOps implementation, this report can provide insight into what you need to ask. If you’re a vendor in, or trying to get into, this market, this report gives you some good insight into how you can assess your solution in the field against your peers. In both cases, we all become better informed about what’s available today.

 

Join The Conversation

I invite you to reach out to me through social media if you want to provide general feedback. If you prefer more formal or private discussions, email inquiry@forrester.com to set up a meeting! Click Carlos at Forrester.com to follow my research and continue the discussion.