Product Guides / AlertsstableUpdated 2026-07-06

Routing & Escalation

Configure who gets notified and when escalation happens.

Configure who gets notified and when escalation happens.

## What this page is for

Use this page when you need to understand what the Routing & Escalation surface does in Kadryn, how it supports the operating workflow, and how it connects to the rest of the product.

## Who should use it

- FinOps users who need operational visibility. - Engineering and platform teams that own AI workloads. - Admins who configure workspace rules, access and governance. - Executives or finance stakeholders when the page supports review or reporting.

## What you can see

- Active and resolved alert states

  • Severity, status and evidence
  • Routing and escalation context
  • Links to costs, guardrails and runbooks

## What you can do

- Triage cases

  • Resolve or reopen where permitted
  • Mute noise when appropriate
  • Follow runbooks and escalate incidents

## Permissions and plan access

Visibility and editability can differ. Kadryn can show a read-only or locked view when the workspace plan, billing state or role does not allow mutation. Treat UI locks as guidance and backend guards as the source of truth.

## How to use it

  1. 1. Open the matching Kadryn menu.
  2. 2. Confirm the workspace, time range, environment and filters.
  3. 3. Review the summary cards before acting on individual rows.
  4. 4. Use filters to isolate the relevant project, provider, team, owner or status.
  5. 5. Open details before making a governance, billing or operational decision.
  6. 6. Verify the result through the linked Costs, Alerts, Guardrails, Developers or Audit surface.

## How Kadryn uses this data

- Uses cost signals, usage events and governance events

  • Connects alerts to owners and routes
  • Feeds operational review and incident learning

## Common mistakes

- Treating a summary card as the source of truth without opening the detail view. - Confusing product entitlements with provider billing. - Editing a setting without checking the affected scope. - Comparing environments without confirming metadata quality. - Ignoring stale, missing or synthetic data indicators.

## Troubleshooting

- If data is missing, check Developers → Logs & Traces and Costs → Data Health. - If an action is disabled, check your role, plan and billing state. - If a policy or cap blocks traffic unexpectedly, inspect Guardrails → Policies and Developers → Logs & Traces. - If counts look wrong, verify filters, archived/deleted toggles and the selected time range.

## Related pages

- /docs/concepts/alerts-vs-guardrails-vs-optimizations