For a concise summary of What’s New in the Vancouver release for ITOM see Jason Smith’s excellent blog. The purpose of this blog is to go a little deeper into how the release fits into our overall AI strategy. This blog is part of a two-part series. The second will focus on Generative AI Alert enhancement.
AIOps User Experience
Business leaders feel the urgency of the upcoming AI revolution. They want to adopt AI-powered innovations fast enough to stay ahead of competitors, but judiciously enough to limit immense risks to security and privacy.
ServiceNow has provided trusted, serious, and secure AI for many years. Our AI is already transforming our customer’s operations. It helps them automatically map their application services, automatically group incidents and alerts, automatically reduce noise, and automatically address issues with tailored workflow before they become outages.
Our customers asked us for two additional things: (1) they wanted to be able to clearly demonstrate the value of these innovations with clear metrics and meaningful insights, and (2) they wanted a compelling user experience that was equal to the power of their ITOM AI capabilities.
ServiceNow’s original interfaces were built on a classic business web design, solid but time-bound. The new AIOps Experience makes full use of the available screen, takes full advantage of AI-powered data processing, and puts the full array of the tool’s features directly under the user’s fingertips.
With these new releases, companies who are anxious for the AI revolution now have compelling interfaces to match the industry-leading application of AI to their critical operations.
At ServiceNow, we believe that AI is most effective when used to empower humans, not replace them. The new AI Experience will enable your organization’s technical heroes to:
- Identify and address issues from a single screen: The alert list has been reconceptualized and then carefully redesigned as the Express List. Operators will move more rapidly and efficiently through clustered alert groups to the underlying issue. Without leaving a single screen, located just off the Service Operations Workspace, they will be able to take in the full blast radius of a problem and then take action to quickly remediate the root issue, even if that involves updating many records at once.
- Extract the value of other tools: One of the enduring strengths of the ServiceNow platform has been its interoperability with other third-party systems. The Integration Launchpad is a visually engaging but straightforward new interface in the AIOps experience that encourages users to find and quickly configure integrations for alerts, metrics, and health logs.
These may be one of numerous out-of-box integrations, or, when not available, the Integrations launchpad provides a series of screens that walk administrators through the steps necessary to build and then manage a custom integration (or webhook). At the conclusion of the process, the launchpad can supply an automatically generated URL to install the connector on a third-party system. After adoption or creation, the integrations launchpad then serves as a central point to view and manage existing integrations.
- Clarify outcomes—clarify value: A robust implementation of AI Operations has many more dimensions than simply bringing in alerts or metrics and processing them. New AI Operations dashboards provide administrators with comprehensive, real-time windows into these key performance dimensions and clearly identify the underlying value to the enterprise.
Every dashboard is informed by the principle that it should clarify necessary outcomes and clearly measure the value of those outcomes over time. This helps administrators to make better decisions and realize demonstrable ROI. It also helps simplify what can be otherwise complex technological dependencies.
Separately designed AIOps dashboards display meaningful measures of operational data enabling administrators to have full awareness of current operations and meaningful insights to set future direction and strategy.
These dashboards include a trends dashboard that displays current operational metrics compared to trends: current noise reduction against the trend, current alert grouping percentage against the trend, and the current incident compression against the trend.
Another operational dashboard focuses on agent health showing the percentages of agents up and the percentage of agents down, as well as breakdowns of agents by version, MID Server, and by the OS on which they are deployed.
A third operational dashboard gives important metrics on HLA processing over time, showing the number of logs processed in the last 24 hours against the trend, the active alerts created by those logs, and the level of those alerts. Log trends help notify administrators of issues when anomalous readings may indicate a deviation from normal so the root of the issue can be squashed before becoming a problem.
By contrast, a series of technology dashboards illustrate the composition of the technology being managed. Administrators can peruse separate technology dashboards breaking down metrics and alerts for VMWare, Windows Servers, or Azure VM’s among other technology platforms. Each technology type is designed with tailored and appropriate visualizations and filters.
Finally, an overall business value realization dashboard provides bottom-line metrics to quantify the impact of AI Operations. These include:
- noise reduction percentage (events to alerts compression)
- average MTTR Incidents created by Event Management, and
- most critical services impacted (in hours)
Also dropping into the Innovation Lab in the Vancouver release are alerts enriched by generative AI. We’ll be covering this innovation in Part Two of this blog. Stay tuned!
