Where the time actually goes
A project manager asks for access to a new analytics tool. In ServiceNow that’s one ticket. In practice it’s a license assignment in the SaaS management tool, an account in Okta, a security group in Active Directory, an app pushed from Intune and an update to the CMDB. Five consoles, often three teams. An analyst carries the ticket from one to the next, copying values and chasing whoever owns the next step.
The work in each console takes a few minutes. The request takes days because of the waiting between steps. ITSM tools are built to track that work, and they do it well. They were not built to perform it across other systems.
How a request runs
When the request is submitted, Kinetic reads its type, cost and the requester’s department and picks the approvers and the SLA target. Approvers get the full context and can act from email or the portal.
On approval, the workflow runs the fulfillment steps itself. Steps that don’t depend on each other run at once: the license and the Okta account don’t wait for each other. Steps that do depend on each other wait. The application role isn’t assigned until the Active Directory account exists, and Kinetic checks it exists before moving on.
When a step fails, because a license pool is empty or a system times out, that branch stops and the error goes to the team that owns it with the request attached. Everything else carries on. The ServiceNow ticket is updated at each step, so agents see the same status the requester sees.
A request is marked fulfilled only when every system confirms its step, not when someone remembers to close the ticket.
Missile Defense Agency
At MDA, requests started as emails that waited in inboxes, were re-keyed into a second system and moved through sequential approvals that could take five business days. Kinetic automated the routing and approvals on top of the agency’s existing systems, applying business rules to pick approval paths and escalate exceptions. Across more than 9,000 users, routing cycles dropped from five days to two, exceptions fell 40%, and the agency puts the annual cost impact at $1.1–1.6M. Read the MDA case study.
Building it out
Service request automation usually grows into a full IT service catalog, with each catalog item backed by a workflow like the one above. Access request automation is the most common first category, because access requests are frequent, multi-system and audited. The workflow layer page explains parallel steps, dependencies and error handling in more depth.