How Varnel routes every alert to the right engineer

Clear ownership during incidents at Varnel

Alerts without an owner

Varnel had grown into several monitoring tools, each with its own alert channel. When something failed, alerts landed in shared rooms and engineers had to work out who owned the service. On a bad night, that took longer than the fix itself.

Schedules and escalation per team

The site reliability team moved all monitors into Navy and set up an on-call schedule for each of its 40 service teams. Every monitor belongs to a team, and every team has an escalation policy. Critical alerts go to the primary engineer by SMS and PagerDuty, then escalate to the secondary and the team lead if nobody acknowledges.

One timeline for everyone

Navy creates an incident timeline automatically, with response time charts and every alert and acknowledgement in order. After each incident, the owning team writes a postmortem in Navy, and action items are tracked to completion in the weekly review.

Details:

• Industry: Enterprise Cloud Computing
• Location: Santa Clara, California
• Employees: 18,000+
• Identity Provider: Microsoft Entra ID
• Key Workloads: Instance Health Checks, Public APIs, Integrations

With SSO and the audit log on the Business plan, Varnel’s security team signed off on the rollout in a week. Alerts now reach the right engineer the first time, and incidents close faster as a result.

---