Agentic ITSM Tools Are Ready—Enterprises Mostly Aren’t
Vendors have moved fast on agentic ITSM tools. Ivanti has already shipped an autonomous service desk agent that can create incidents, submit requests, and search knowledge bases without analyst intervention. ServiceNow agents are following a similar path, promising end‑to‑end ticket handling and policy‑aware actions. Early adopters see the appeal: Grand Bank’s CIO described how repetitive requests consumed most of the team’s time before Ivanti’s deployment, and expects automation to free staff for higher‑value work. Research from McKinsey reinforces the opportunity, citing a multinational that automated up to 80% of roughly 450,000 annual tickets after redesigning for agent‑led resolution. Yet that redesign is the catch. Most organisations still operate estates optimised for human‑centric, ticket‑based workflows, not autonomous decision‑making. The tools are capable, but the average enterprise IT readiness level—across technology, process, and culture—lags far behind what truly agentic ITSM automation adoption requires.
The ‘Plumbing Problem’: Legacy Estates Block Agentic ITSM
Under the surface, agentic ITSM tools depend on infrastructure that many organisations simply do not have. McKinsey calls it a “plumbing problem”: environments designed around siloed applications and human handoffs struggle to support agents that must cross systems, call APIs, and execute actions under strict policy controls. Their survey suggests 62% of organisations are still only piloting agents, with no more than 10% scaling them in any function. Red Hat’s CEO described a “back‑to‑basics” moment, where even patching discipline must be relearned before layering on autonomy. In unified communications scenarios, the gaps are obvious. A voice‑quality incident can touch network telemetry, device health, carrier status, and ITSM records. If an agent cannot reliably traverse each domain, it has to hand the case back to a human—precisely the inefficiency these deployments are meant to remove. Without modernised plumbing, agentic ITSM tools stall at proof‑of‑concept.
From Automation to Autonomy: New Governance and Observability Demands
Agentic ITSM is not just more automation; it is a different operating model. Traditional ITSM platforms route tickets and trigger scripts, but humans retain the final decision. With ServiceNow agents or Ivanti agentic capabilities, software chooses actions in production environments. That shift raises uncomfortable governance questions: what can agents do, where are the limits, and who is accountable when something goes wrong? Monitoring is another blind spot. Standard tools can confirm a service is running, yet cannot explain what an individual agent decided, which systems it changed, or why. Gartner’s Padraig Byrne highlights this visibility gap as a central risk to scaling, noting that AI decision‑making is often opaque even when the business impact is significant. Gartner expects 40% of organisations deploying AI to adopt dedicated observability tooling by 2028, meaning most IT teams will lack mature oversight for at least the next two years.
Four Technical Prerequisites Most ITSM Teams Still Lack
McKinsey outlines four prerequisites for safe, scalable, agentic ITSM that many enterprises are only beginning to tackle. First is an accurate, trustworthy CMDB so agents can act on current configuration data; stale records lead directly to bad decisions and new incidents. Second, operational actions must be exposed through secure APIs with policy checks built in, giving agents a controlled way to execute changes. Third, organisations need a clear governance model defining which agents can perform which tasks, under what conditions, and with what approvals. Finally, teams must actively monitor inference activity and cost to avoid runaway resource consumption and unexpected side effects. For most ITSM and UC teams, these are not minor tuning tasks but a multi‑year programme of work. Critically, that programme should start before buying agentic ITSM tools, not after pilots fail or generate more tickets than they close.
Assessing Enterprise IT Readiness Across Technology, Process, and Culture
To close the readiness gap, IT leaders need a holistic view that spans technology, process, and culture. Technically, they must evaluate CMDB quality, integration coverage, API maturity, and existing observability capabilities before switching on ServiceNow agents or Ivanti agentic capabilities. On the process side, workflows and customer journeys should be redesigned for agent‑first resolution, not merely bolted onto legacy ticket queues. That includes defining clear fallback paths when agents escalate to humans. Culturally, teams must shift from owning every click to supervising autonomous systems—trusting agents where appropriate, but scrutinising their decisions and outcomes. Governance forums, clear accountability, and change‑management communications all play a role. With AI infrastructure costs expected to rise even as budgets stay flat, the pressure to achieve effective ITSM automation adoption will only grow. Enterprises that invest in readiness now will be positioned to exploit agentic ITSM tools rather than be disrupted by them.